lcov-1.14/0000755000000000000000000000000013436014710011077 5ustar rootrootlcov-1.14/CHANGES0000644000000000000000000037133213436014710012103 0ustar rootrootcommit 6c7ad581ab9ed66f80050970c0d559c6684613b7 (HEAD, tag: v1.14, origin/master, origin/HEAD, master) Author: Peter Oberparleiter Date: Thu Feb 28 18:01:39 2019 +0100 lcov: Finalize release 1.14 Signed-off-by: Peter Oberparleiter commit 29814f18ec207ebaefa7b41f6e5acc4eca6d7a7a Author: Peter Oberparleiter Date: Thu Feb 28 17:31:17 2019 +0100 geninfo: Fix missing FN: entries in result files geninfo sometimes fails to correctly collect function starting lines for some source files, resulting in output files with missing FN: lines. Also such functions are missing from the function list in HTML output. The problem occurs when a) multiple source files contribute to a function implementation (e.g. via including code), and b) the source file that contains the initial function definition is not the source file that contains the most function definitions The problem occurs due to a heuristic in function graph_find_base() that incorrectly determines the source file for a function in this situation. Fix this by using the first file that contributes to a function as the base source file for that function. Only apply this change to data collected using GCC versions 4 and above since earlier versions did not produce stable file orders in graph files. Signed-off-by: Peter Oberparleiter Reported-by: Joshua Cranmer commit 74bae96e8ef724eb9dbdf126adad17505375e149 Author: Peter Oberparleiter Date: Thu Feb 28 16:15:22 2019 +0100 Makefile: Make Perl path install-time configurable Add support for specifying the Perl interpreter path used in installed Perl scripts. If no path is specified, the default '/usr/bin/perl' is used. Set variable LCOV_PERL_PATH to specify a different path, for example: make install LCOV_PERL_PATH=/usr/local/bin/perl Unset this variable to keep the current path: make install LCOV_PERL_PATH= Signed-off-by: Peter Oberparleiter commit 0b378cba2c0f93d728627aa8750849d3c33de0e1 Author: Peter Oberparleiter Date: Thu Feb 28 14:21:18 2019 +0100 bin,test: Use /usr/bin/env to locate script interpreters Make use of the /usr/bin/env tool to locate script interpreters. This is needed to locate the correct interpreter in non-standard environments. Signed-off-by: Peter Oberparleiter Suggested-by: Bjørn Forsman Suggested-by: Mario Costa commit 2ff99aefbd0c80fe0cfddf1e09a596d7344533e1 Author: Peter Oberparleiter Date: Thu Feb 28 14:09:42 2019 +0100 bin/*: Remove '-w' from interpreter specifications Replace '-w' flag from Perl interpreter specifications with 'use strict' directive. This is done in preparation of using a more flexible interpreter specification. Signed-off-by: Peter Oberparleiter commit 3b378b0e76be95971680056d864d0e13f4a08557 Author: Peter Oberparleiter Date: Wed Feb 27 16:33:42 2019 +0100 geninfo: Fix errors while resolving /././ path components Trying to collect code coverage data for source code that contains repeat ./ references in a path components fails with the following error message: geninfo: WARNING: cannot find an entry for .gcov in .gcno file, skipping file! This is caused by a bug in path normalization function solve_relative_path() that does not correctly process adjacent ./ references. Fix this by repeating the resolution of ./ references in path components. Signed-off-by: Peter Oberparleiter Reported-by: Joshua Cranmer commit 42b55f5a497d2286566d0dd411e3e52fd4d50469 Author: iignatyev Date: Wed Feb 6 11:35:02 2019 -0800 geninfo: preserve-paths makes gcov to fail for long pathnames geninfo uses '--preserve-paths' gcov option whenever gcov supports it, this forces gcov to use a whole pathname as a filename for .gcov files. So in cases of quite large pathnames, gcov isn't able to create .gcov files and hence geninfo can't get any data. The fix replaces usage '--preserve-paths' with '--hash-filenames' when it is available. Signed-off-by: Igor Ignatev commit 04335632c371b5066e722298c9f8c6f11b210201 Author: Peter Oberparleiter Date: Fri Jan 11 13:53:33 2019 +0100 geninfo: Fix "Can't use an undefined value" error When run on data for source code that causes gcc 8 to generate artificial functions, geninfo emits warnings and eventually aborts processing: geninfo: Use of uninitialized value in hash element at /usr/local/bin/geninfo line 3001. geninfo: Can't use an undefined value as an ARRAY reference at /usr/local/bin/geninfo line 2889. This problem was introduced by commit 9aa0d14a ("geninfo: Ignore artificial functions during --initial"). It is the result of an incomplete removal of artificial functions from internal data. Fix this by explicitcly removing known artificial functions after parsing of graph files completes. Signed-off-by: Peter Oberparleiter Reported-by: Steven Peters commit 9aa0d14af4446ef46d80356849a97bc961a91f97 Author: Peter Oberparleiter Date: Thu Jan 10 13:20:15 2019 +0100 geninfo: Ignore artificial functions during --initial Graph files generated by gcc 8 may contain "artifical" functions that do not exist in a source file. geninfo incorrectly generates coverage data for these functions when run with option --initial. Fix this by filtering out artifical functions when generating initial coverage data. Signed-off-by: Peter Oberparleiter Reported-by: Marcin Konarski commit 1e0df571198229b4701100ce5f596cf1658ede4b Author: Peter Oberparleiter Date: Thu Jan 10 11:39:07 2019 +0100 geninfo: Fix data collection for files containing templates When using gcc 8, lcov/geninfo produces corrupt coverage output for source code that contains templates or other constructs that cause gcov to produce multiple versions of output for some lines and branches. This is caused by an incorrect check for duplicate output in function read_gcov_file() that is triggered when a template consists of multiple lines, or contains branches. Fix this by ensuring that duplicate lines in per-instance gcov output are correctly ignored. Only the initial occurrence of each line containing the combined coverage of all instances will be processed by geninfo. Note that for branch coverage, gcov doesn't provide a combined view and geninfo processes all branches provided. This should not be a problem though as genhtml will combine the branch data when generating HTML output. Signed-off-by: Peter Oberparleiter commit abd8bed2b013334d4ef978abadbfff6cc6f3d55d Author: MarcoFalke Date: Tue Jan 8 12:49:00 2019 +0100 genhtml: Unconditionally include anchor for each named line This helps with referencing the line in the html when sharing links. Signed-off-by: MarcoFalke commit 28675dc7564aaa1ad231a7ac23106512a3956d68 Author: Peter Oberparleiter Date: Tue Dec 18 13:07:58 2018 +0100 genhtml: Use gmtime for SOURCE_DATE_EPOCH conversion By changing that localtime to gmtime the "Known bug" section of the commit message can be removed. Signed-off-by: Peter Oberparleiter Suggested-by: Bjørn Forsman commit 180286bec651928c41de4d6ce3a8760678b38f60 Author: Bjørn Forsman Date: Tue Dec 4 14:30:28 2018 +0100 genhtml: honor the SOURCE_DATE_EPOCH variable Implement the SOURCE_DATE_EPOCH specification[1] for reproducible builds. If SOURCE_DATE_EPOCH is set, use it as timestamp instead of the current time. In this context, reproducible builds means reproducible HTML coverage reports. Known bug: the specification[1] says to defer converting the timestamp to local timezone at presentation time. This is currently not happening; it's converted at build time. [1] https://reproducible-builds.org/specs/source-date-epoch/ Signed-off-by: Bjørn Forsman commit 41e07cadeeae3054ac22202d5b0b0f0ef6e26467 Author: Bjørn Forsman Date: Tue Dec 4 14:30:27 2018 +0100 Tolerate CDPATH being set If CDPATH is set, cd will print the path it enters, resulting in TOOLDIR containing the path twice, separated by a newline. Signed-off-by: Bjørn Forsman commit a3bbe8f0398a3c36b4228cc173e4739d27a863e1 Author: Peter Oberparleiter Date: Mon Dec 10 13:58:10 2018 +0100 CONTRIBUTING: Clarify patch format requirements Signed-off-by: Peter Oberparleiter commit e6750800fe4cb89eda1ff80b7a5fe70fe87ede36 Author: Peter Oberparleiter Date: Tue Nov 13 17:28:17 2018 +0100 geninfo: Fix accounting of basic blocks in exceptional paths Basic blocks that are not executed and are only reachable via exceptional paths are marked with a '%%%%%' marker in the GCOV output of current GCC versions. Fix geninfo to also recognize this marker. Signed-off-by: Peter Oberparleiter Reported-by: trotux (github user) commit 94eac0ee870e58630d8052dca1181b0cf802525f Author: Peter Oberparleiter Date: Mon Jul 16 13:24:58 2018 +0200 lcov: Fix branch coverage summary When combining two data files A (without branch coverage data) and B (with branch coverage data), lcov will incorrectly report no branch coverage data for the resulting file in program output, even though the resulting file contains branch coverage data. This only happens when A is specified first during the add operation. This is due to a bug in lcov that loses the correctly combined branch coverage data internally in function brcount_db_combine() when its first parameter is undefined. Fix this by ensuring that the first parameter is an empty hash reference instead. Signed-off-by: Peter Oberparleiter commit a5dd9529f9232b8d901a4d6eb9ae54cae179e5b3 Author: Peter Oberparleiter Date: Wed Mar 7 14:18:55 2018 +0100 geninfo: Add gcc 8 support Fix errors and incorrect data when trying to collect coverage data for programs compiled with gcc 8. Covers the following gcov-related changes in gcc: .gcov-file format: - Line coverage data can appear multiple times for the same line - Line coverage count can be suffixed by '*' to indicated unexecuted basic blocks in that line .gcno-file format: - new header field 'support unexecuted blocks flag' - new function record fields 'column number', 'ending line number', and 'compiler-generated entity flag' Signed-off-by: Peter Oberparleiter commit c30d88a3a8096dbb3f968de999480c3dc2dedb5f Author: Peter Oberparleiter Date: Tue Jan 30 15:12:09 2018 +0100 genhtml: Implement option to show miss counts Add new command line option --missed that can be used to show the number of missed lines, functions, or branches. Signed-off-by: Peter Oberparleiter commit 999abf2447b4df373b135dc3f8ee317350bd95f8 Author: Benoit Belley Date: Fri Oct 6 10:01:28 2017 -0400 Adding the --include and --exclude options to lcov and geninfo * The --include command-line option allows the user to specify a regular expression for the source files to be included. The command-line option can be repeated to specify multiple patterns. The coverage information is only included for the source files matching at least one of the patterns. The "lcov --capture --include" (or "geninfo --include") option is similar in functionality to the "lcov --extract" command-line option. But, by directly using applying the pattern while capturing coverage data one can often avoid having to run "lcov --extract" as a second pass. * The --exclude command-line option allows the user to specify a regular expression for the source files to be excluded. The command-line option can be repeated to specify multiple patterns. The coverage information is excluded for source files matching at least one of the patterns. The "lcov --capture --exclude" (or "geninfo --exclude") option is similar in functionality to the "lcov --extract" command-line option. But, by directly using applying the pattern while capturing coverage data one can often avoid having to run "lcov --remove" as a second pass. * On one of our code base at Autodesk, this speeds-up the generation of HTML code coverage reports by a factor of 3X. Signed-off-by: Benoit Belley commit b6a11368c3cdc86c4e147ccd8e539918dfe37900 Author: Ziqian SUN (Zamir) Date: Wed Jul 19 10:58:24 2017 +0800 Resolve some rpmlint issue in SPEC. Following messages reported by rpmlint on RHEL is fixed by this patch: lcov.src: W: invalid-license GPL lcov.src:9: W: hardcoded-path-in-buildroot-tag /var/tmp/%{name}-%{version}-root lcov.src: E: specfile-error warning: bogus date in %changelog: Fri Oct 8 2002 Peter Oberparleiter (Peter.Oberparleiter@de.ibm.com) lcov.noarch: W: non-conffile-in-etc /etc/lcovrc Signed-off-by: Ziqian SUN (Zamir) [oberpar@linux.vnet.ibm.com: Corrected license to GPLv2+] commit a77a7628ef5377c525a0d4904cc0b73eeede4d7c Author: Peter Oberparleiter Date: Fri Apr 7 15:43:28 2017 +0200 genhtml: Reduce path resolution overhead Reduce overhead when reading coverage data files by consolidating calls to Cwd:cwd(). Signed-off-by: Peter Oberparleiter commit 526c5148ac0add40ef1224d2cdabdec73ce3f899 Author: Peter Oberparleiter Date: Fri Apr 7 15:37:52 2017 +0200 genhtml: Reduce load times for complex coverage data files genhtml uses a significant amount of time loading coverage data files containing complex branch coverage data (i.e. data with a large number of branches per basic block). Most of this time is spent storing branch coverage data in a vector-based data representation, with an unnecessary amount of cross-checking being done for existing branch data. Fix this by replacing the vector based data representation by two separate representations, scalar for storage and hash for processing, and by moving cross-checking out of the hot path. This results in a significant speedup at the cost of a minor increase in memory usage. Test results for "make -C genhtml_output/ SIZE=large": Original: 6 tests executed, 6 passed, 0 failed, 0 skipped (time 768.4s, mem 893.8MB) Patched: 6 tests executed, 6 passed, 0 failed, 0 skipped (time 202.3s, mem 908.10MB) Signed-off-by: Peter Oberparleiter commit 0f07133f184af6670bdf1edf39fca9d2e90e9ad2 Author: Peter Oberparleiter Date: Fri Apr 7 14:38:22 2017 +0200 test: Add self-tests for genhtml Add some tests for checking basic functionality of genhtml. Signed-off-by: Peter Oberparleiter commit 544a6951db25679792bb0648006a897ea564d883 Author: Peter Oberparleiter Date: Fri Apr 7 14:32:47 2017 +0200 genhtml: Ensure stable block order in branch output Sort order of basic blocks in output of branch coverage data. This allows for a better comparison of output between test cases. Signed-off-by: Peter Oberparleiter commit 477957fa4c6c104d5842911682ec17d6ad2d2980 Author: Peter Oberparleiter Date: Thu Apr 6 12:28:11 2017 +0200 lcov: Reduce load times for complex coverage data files lcov uses a significant amount of time loading coverage data files containing complex branch coverage data (i.e. data with a large number of branches per basic block). Most of this time is spent storing branch coverage data in a vector-based data representation, with an unnecessary amount of cross-checking being done for existing branch data. Fix this by replacing the vector based data representation by two separate representations, scalar for storage and hash for processing, and by moving cross-checking out of the hot path. This results in a significant speedup at the cost of a minor increase in memory usage. Test results for "make test SIZE=large": Original: 17 tests executed, 17 passed, 0 failed, 0 skipped (time 1883.9s, mem 2459.0MB) Patched: 17 tests executed, 17 passed, 0 failed, 0 skipped (time 283.6s, mem 2544.2MB) Note that this fix only applies to the lcov tool. The same work is necessary for genhtml. This approach was inspired by a patch by creich.3141592@gmail.com. Signed-off-by: Peter Oberparleiter commit 3b397a3f3acdb62080e8366130758cb34703cfbf Author: Peter Oberparleiter Date: Thu Apr 6 09:01:36 2017 +0200 test: Improve test framework Various improvements to lcov's self-test framework: - Add test case for lcov --diff - Add new verbosity level - Enable normalization of coverage data files from stdin - Fix lcov_add_concatenated4 test name Signed-off-by: Peter Oberparleiter commit 53a6ce8ef604173b6de874a534a30121392d7cd0 Author: Peter Oberparleiter Date: Thu Mar 30 15:42:56 2017 +0200 lcov: Add self-tests Add some tests for checking basic functionality of lcov. To run these tests, type: make test in either the top-level directory, or the test/ sub-directory. Signed-off-by: Peter Oberparleiter commit 9753d5c0da107919537e91e504551e4ab3bccc2f Author: Peter Oberparleiter Date: Thu Mar 30 15:31:34 2017 +0200 lcov: Fix output on stderr for --summary Some functions of lcov erroneously print informational output to stderr instead of stdout as expected. Fix this by inverting the "to_file" logic in lcov to a "data_stdout" logic. Affected functions are --summary, --reset and --list. Signed-off-by: Peter Oberparleiter commit 25f5d38abad20eeaa407f62f53c3c00dfbbd0bf3 Author: Peter Oberparleiter Date: Mon Mar 6 09:51:00 2017 +0100 lcovrc.5: Add genhtml_demangle_cpp default and CLI reference Signed-off-by: Peter Oberparleiter commit 66db744a1d63c5d3b1dee2d8a2ce76e6e06c7255 Author: Katsuhiko Nishimra Date: Fri Mar 3 17:47:48 2017 +0900 Support passing demangle-cpp option via lcovrc This patch allows users to passing the demangle-cpp option to genhtml via lcovrc, alongside with CUI. Signed-off-by: Katsuhiko Nishimra commit b6fb452addaa6a33dcb37c101879b8b5e1e0c34c (tag: v1.13) Author: Peter Oberparleiter Date: Mon Dec 19 15:20:40 2016 +0100 lcov: Finalize release 1.13 Signed-off-by: Peter Oberparleiter commit daca8d9febe52ccf1976240a3b48ffc350dec902 Author: Peter Oberparleiter Date: Mon Dec 19 14:36:00 2016 +0100 geninfo: Fix 'unexpected end of file' error Use the compiler version as stored in the .gcno file to determine if the file contains function records with split checksums. This fixes the following problem that can occur when lcov is run using a gcov tool of GCC version 4.7 and above on .gcno files compiled with a version below 4.7: # lcov -c -d . -o test.info --initial [...] geninfo: ERROR: test.gcno: reached unexpected end of file Also add missing lcov version to --debug output. Signed-off-by: Peter Oberparleiter commit a90d50e97cb49ea712c94d91cdef1cc21a3c7986 Author: Peter Oberparleiter Date: Wed Dec 14 11:00:08 2016 +0100 lcov: Remove use of install -D option Some versions of the install tool don't support the -D option, causing a 'make install' call to fail. Fix this by replacing the -D option with two calls to install, first to create all target directory components, then to install the actual files. Signed-off-by: Peter Oberparleiter Reported-by: commit 6ec3f2398d22e605c1a8019541fb32d26d18044b Author: Peter Oberparleiter Date: Fri Oct 7 09:47:35 2016 +0200 genhtml: Fix warning with small genhtml_line_field_width On systems with Perl versions 5.21 and above, genhtml prints a warning similar to the following during processing: genhtml: Negative repeat count does nothing at bin/genhtml line 3854, line 4. This is due to size calculations resulting in a negative number of padding characters when genhtml_line_field_width is lower than the size of the strings to pad (9). Fix this by disabling padding in these cases. Reported-by: xaizek@openmailbox.org Signed-off-by: Peter Oberparleiter commit d7cc7591b3a7cc1ec95371d04e4fc46f10b3fd54 Author: Peter Oberparleiter Date: Tue Oct 4 09:50:52 2016 +0200 geninfo: Fix gcov version detection for XCode 8.0 The LLVM gcov version included in XCode 8.0 reports its version in a format that is not understood by geninfo, resulting in the wrong format of coverage data files being expected. Fix this by reworking gcov version detection in geninfo to be more robust. Signed-off-by: Peter Oberparleiter commit 68320d932c5ee5537ae1c287fe52603ae2fecf8c Author: Peter Oberparleiter Date: Mon Aug 22 15:54:56 2016 +0200 lcov: Update installation mechanism Change default installation location to /usr/local to prevent conflicts with files installed by package managers (reported by Gregory Fong). To achieve this, rename PREFIX to DESTDIR and introduce actual PREFIX Makefile variable and update spec file to install packaged files to previous locations. Also fix spec file to not announce ownership of system directories (reported by and based on patch by Jiri Kastner ). Signed-off-by: Peter Oberparleiter commit 04a3c0ed1b4b9750b2ac5060aac0e6d5a3b9da7f Author: Benoit Belley Date: Mon Apr 4 18:16:54 2016 -0400 Pass --no-strip-underscore to c++filt on OS X * The --no-strip-underscope flag is necessary on OS X so that symbols listed by gcov get demangled properly. From the c++filt man page: "On some systems, both the C and C++ compilers put an underscore in front of every name. For example, the C name "foo" gets the low-level name "_foo". This option tells c++filt not to remove the initial underscore. Whether c++filt removes the underscore by default is target dependent." Signed-off-by: Benoit Belley commit 632c25a0d1f5e4d2f4fd5b28ce7c8b86d388c91f Author: Peter Oberparleiter Date: Tue Mar 8 10:51:51 2016 +0100 lcov: Fix output files being created in / directory When a warning is emitted by lcov before creating an output file, e.g. when a negative coverage count was found while combining tracefiles, lcov tries to create the output file in the root directory (/) instead of the current working directory. This is a result of lcov's warn handler calling a temp file cleanup routine that changes directories to / before trying to remove its temporary directory. Fix this by removing the temp cleanup call from the warn handler. Signed-off-by: Peter Oberparleiter commit e32aab1b4c85503a6592a91326c4b362613e1d66 Author: Gabriel Laskar Date: Wed Feb 10 09:56:18 2016 +0100 lcov: Fix --remove pattern matching The --remove option of lcov now consider the pattern passed as parameter as a full path, and not only a part of the filename. This behavior was discovered by using AX_CODE_COVERAGE[1] m4 macro from a directory in $HOME/tmp. The macro itself calls lcov with `--remove "/tmp/*"`. [1]: https://www.gnu.org/software/autoconf-archive/ax_code_coverage.html Signed-off-by: Gabriel Laskar commit 79e9f281ea893b2f6498b4bad79173b1414aa055 Author: Reiner Herrmann Date: Fri Oct 30 20:26:59 2015 +0100 lcov: use UTC to get timezone-independent date The date is used for updating the time inside manpages. If localtime is used, the date could vary depending on the user's timezone. To enable reproducible builds, UTC is used instead. Signed-off-by: Reiner Herrmann commit de33f51b49dc6d01a285aa73990f03e7d982beb2 (tag: v1.12) Author: Peter Oberparleiter Date: Mon Oct 5 17:37:40 2015 +0200 lcov: Finalize release 1.12 - Use full git describe output as tool version - Update version numbers and last-changed-dates in man pages, spec and README file - Replace static CHANGES file with git log - Switch Makefile logic to use mktemp for generating a temporary directory Signed-off-by: Peter Oberparleiter commit 1ad4f7779b7721e311e552209e110e08bbf18fa1 Author: Denis Abramov Date: Mon Sep 21 09:29:20 2015 +0200 geninfo: Added support for Xcode 7.0 gcov version handling With Xcode 7.0 LLVM gcov keeps version information on the first line. E.g. gcov --version yields: Apple LLVM 7.0.0 (clang-700.0.65) Signed-off-by: Denis Abramov commit c3602ea8e598deda4afff603bb123caa98eef159 Author: Peter Oberparleiter Date: Mon Aug 3 11:05:51 2015 +0200 genhtml: Allow prefix paths with spaces Signed-off-by: Peter Oberparleiter commit a3572971367198ef0febe476052640bd09bec931 Author: Gilles Gouaillardet Date: Thu Jul 30 14:11:57 2015 +0900 genhtml: support a comma separated list of prefixes the --prefix option of genhtml now takes a comma separated list of prefixes instead of a single prefix. this can be required when running lcov vs projects configure'd with VPATH and in which source files are both in the source and build directories. Signed-off-by: Gilles Gouaillardet commit 997f32ae85717cd47d2305d7cd7ccce3ffa1abe6 Author: Gilles Gouaillardet Date: Tue Jun 23 14:28:22 2015 +0900 Fix find command line find xxx -name \*.gcda -type f -o type l does return : - all files with the .gcda suffix - all symbolic links the updated command line now returns - all files with the .gcda suffix - all symbolic links with the .gcda suffix Signed-off-by: Gilles Gouaillardet commit 533db4e78b54ae01e023d00c1fec5dddaaaf37e6 Author: Peter Oberparleiter Date: Wed Jun 17 17:54:20 2015 +0200 lcov: Fix capture for package files containing graph files Depending on whether package files contain graph files, data should be collected from the unpacked package file directly, or from the build directory after linking data files. This approach fixes problems when capturing coverage data via a package from a directory containing graph files. Signed-off-by: Peter Oberparleiter commit a2a8b376ec5e9e5082a0cbb935137d6a8f526870 Author: Peter Oberparleiter Date: Wed Jun 17 17:34:33 2015 +0200 lcov: Fix .build_directory file not being deleted Using option --to-package while capturing coverage data creates a temporary file named ".build_directory". Currently this file is not properly removed at the end of processing due to a changed CWD. This patch fixes this problem by reverting to the original CWD before trying to remove the temporary file. Signed-off-by: Peter Oberparleiter commit b9de825f1fe018f381c8859ee0f3f4af15122c7a Author: Peter Oberparleiter Date: Tue Jun 16 13:53:00 2015 +0200 lcov: Enable userspace package capture with only data files Previously lcov's --from-package capture mechanism required that .gcno files and source were present on the test machine. This patch modifies --from-package capturing to work when only .gcda files are present in the package captured on the test machine. It works by linking the .gcda files collected on the test machine into their natural location on the build machine. This requires existing .gcda files to be removed. Signed-off-by: Peter Oberparleiter commit 0e4f0908aed3e1a071d5435c36c18cd493f0c309 Author: Peter Oberparleiter Date: Tue Jun 16 13:33:54 2015 +0200 lcov: Make package handling more robust Apply some changes to --from-package and --to-package handling to better handle failures: - Abort if tar tool is not available - Abort if no data file is found in package file - Ensure that temporary directories can be deleted Signed-off-by: Peter Oberparleiter commit f87d980929a5a06d49d0a6856f6c3314418c27ef Author: Peter Oberparleiter Date: Tue May 12 17:28:44 2015 +0200 genhtml: Rework c++filt name demangling When running genhtml with command line option --demangle-cpp, do not merge function call data based on demangled function names. Instead mark duplicate function entries in the function view with a version suffix (.). This resolves problems with entries for functions that demangle to the same name, but begin on different lines according to GCC (e.g. virtual destructors). Reported-by: Lukasz Czajczyk Signed-off-by: Peter Oberparleiter commit 2e872175cbba2c09c9025da2660edf0b4abb55cb Author: Daniel Fahlgren Date: Wed Apr 22 15:17:10 2015 +0200 geninfo: make line exclusion markers configurable This patch exposes the variable $excl_line and $excl_br_line so they can be set in the configuration file. It is not always possible to add the exclusion markers to the code with reasons like third party code, company policy, legacy code, no commit access etc. One obvious use case is to exclude assert() from the branch coverage and abort() from line coverage. They are never meant to be triggered unless something is wrong. Other use cases can be custom error handling macros or macros that depend on endianness, like htons(). Signed-off-by: Daniel Fahlgren commit 10b11eaa178976d1433007adb2188d05b8605be6 Author: Peter Oberparleiter Date: Mon Nov 10 17:17:23 2014 +0100 geninfo: Ignore empty .gcno files with --initial Some versions of GCC create empty .gcno files which cause geninfo to abort processing with an error message: geninfo: ERROR: dummy.gcno: reached unexpected end of file Fix this problem by skipping empty .gcno files. Reported-by: Maarten Hoes Signed-off-by: Peter Oberparleiter commit f9d8079646aa906518c4ab7d326504e6837532a7 Author: Peter Oberparleiter Date: Mon Nov 10 16:54:08 2014 +0100 lcov: Fix warning when specifying --rc Current Perl versions report the following warning when using the --rc option of lcov: lcov: Use of each() on hash after insertion without resetting hash iterator results in undefined behavior Fix this warning by not modifying the hash variable that is being iterated on. Also add the missing whitespace fix-up of --rc parameters to genhtml. Reported-by: Maarten Hoes Signed-off-by: Peter Oberparleiter commit 2a634f6caa98f979606189ec3ee98f4cac270b97 Author: Philip Withnall Date: Mon Nov 10 14:58:34 2014 +0000 genhtml: Support relative source filenames in SF keys Some tools which generate .info files generate relative filenames for the ‘SF’ keys. For example, nodeunit’s lcov output does. When genhtml is run with --output-directory, it calls chdir() which breaks relative lookup of the source files. Fix that by resolving all source filenames to absolute paths when loading an info file, resolving any relative ones using the info file’s path as a base. Signed-off-by: Philip Withnall commit b4344c6a5d3c434ca0d801c197a09cfdeecb3f32 Author: Peter Oberparleiter Date: Fri Sep 26 13:11:18 2014 +0200 man: Add description for --precision and genhtml_precision Add man page sections for genhtml's command-line option --precision and lcovrc configuration setting genhtml_precision. Also add an example configuration setting in lcovrc. Signed-off-by: Peter Oberparleiter commit aa1217412f1e8b540010fea5ca9844b9e4699e54 Author: Euccas Chen Date: Fri Sep 26 12:53:29 2014 +0200 genhtml: Implement option to specify coverage rate precision Add command line support and config file support for specifying the coverage rate precision, valid precision range: [1,4]. Signed-off-by: Euccas Chen commit 4d4eba1a8b5e7d2a6c5e93c0a50264da1a5c5540 Author: Peter Oberparleiter Date: Wed Jun 25 09:41:59 2014 +0200 get_version.sh: Remove - characters from release string Replace - with . in release strings to fix the following build error in the dist Makefile target: error: line 4: Illegal char '-' in: Release: 4-g1d44b2a make: *** [rpms] Error 1 Signed-off-by: Peter Oberparleiter commit ffbd3e08cc0871842b2205b0b73c2ae8f3ad02e8 Author: Peter Oberparleiter Date: Wed Jun 25 09:25:50 2014 +0200 genhtml: Improve demangle error message Improve error message that is shown when there are mangled function name entries on different lines that demangle to the same clear text function name. Signed-off-by: Peter Oberparleiter commit 1d44b2a090aa933b15e4cafc1a440ccb390df92e Author: Peter Oberparleiter Date: Tue Jun 24 17:45:34 2014 +0200 geninfo: Fix error when using --demangle-cpp Using genhtml's --demangle-cpp option on data produced with recent GCC versions (at least 4.8 and 4.9) can result in an error message similar to the following: genhtml: ERROR: Demangled function name _ZN3subD2Ev maps to different lines (5 vs 4) The reason for this error is an unexpected sequence of lines records in a .gcno file. These records mention line numbers as belonging to a function which occur before the initial line number of that function as reported by the corresponding function record. Fix this problem by retaining the order of lines belonging to a function as found in the .gcno file. This way geninfo will consistently use the initial line number as reported by the function record when merging function data during the demangling process. Reported-by: Alexandre Duret-Lutz Signed-off-by: Peter Oberparleiter commit 566e5ec7e69a03612e1ed4961779d939af180d66 Author: Peter Oberparleiter Date: Wed Jun 18 16:05:29 2014 +0200 lcov: Remove unused files Signed-off-by: Peter Oberparleiter commit c76172bfe630520e217ecc0bca8f18481c4c33b0 Author: Peter Oberparleiter Date: Wed Jun 18 16:01:05 2014 +0200 README: Fix typo Signed-off-by: Peter Oberparleiter commit a6b10a41056cd10c7b735e259fee81f1865c2109 Author: Peter Oberparleiter Date: Wed Jun 18 15:50:04 2014 +0200 lcov: Remove CVS artifacts Replace CVS specifics in the build environment and tools source with Git mechanisms: * CONTRIBUTING and README file now refer to github for the primary source location * When run from a Git repository, the tools dynamically determine the Git version using 'git describe' * When installed into the file system, the version information is fixed with the current Git version * When preparing distribution files, the version at the time of preparing the files is written to file ".version" Also add a .gitignore file to filter out the most frequent temporary file types. Signed-off-by: Peter Oberparleiter commit fa2a991cf6fad37fec7650b95be705df143e058a (tag: v1.11) Author: Peter Oberparleiter Date: Fri May 23 08:56:17 2014 +0000 lcov: finalizing release 1.11 commit e2729beea0d7769ef0e992c27a294b0742a6ac77 Author: Peter Oberparleiter Date: Fri May 23 08:47:10 2014 +0000 CHANGES: update commit 866d187602bfc2e3a8199f4e9e9430ef38f106a8 Author: Jeffrey Hutzelman Date: Tue May 20 14:12:55 2014 +0000 lcov: Sort branches in unnamed blocks first When processing branch coverage data, consider branches in "unnamed" blocks to come before other blocks on the same line, so that they appear in the correct order in HTML output. This is accomplished by using block number -1 for unnamed blocks, instead of 9999 as was previously done. In branch data vectors, this is reprsented by the value $BR_VEC_MAX, which is defined to be the largest value representable in the field width used. This same value is also used in .info files, for backward-compatibility with regular expressions used to parse these files. As a result, .info files generated by versions of lcov with this change can be read by older versions, though branch results will still appear out of order. Signed-off-by: Jeffrey Hutzelman commit 17c0edec32193b9e8058908447d3eb403d76c8de Author: Peter Oberparleiter Date: Thu May 15 10:23:45 2014 +0000 lcov: Update man page Add missing description for command line parameter value. Reported-by: sylvestre@mozilla.com commit c0958139e015805cce15b60b740c735690ad4002 Author: Peter Oberparleiter Date: Mon Apr 14 12:14:55 2014 +0000 genhtml: Implement option to allow HTML in test description Add lcovrc directive genhtml_desc_html to allow using HTML markup in test case description text. Signed-off-by: Peter Oberparleiter commit 4f2c3aefcfcf816806da83a8609bd743eb227d37 Author: Peter Oberparleiter Date: Mon Apr 14 11:24:05 2014 +0000 genhtml: Check for proper description file format Ensure that description files contain test name lines before test description lines. This fixes a "use of uninitialized value" warning in genhtml. Signed-off-by: Peter Oberparleiter commit 3a68239905c28a7c3bfac52172a254872d6a7aa7 Author: Jonah Petri Date: Mon Apr 14 11:06:21 2014 +0000 lcov: make geninfo compatible with LLVM's gcov These changes are needed to make geninfo compatible with LLVM's gcov: * Use --version rather than -v to probe version info * Convert LLVM gcov version numbers to the GCC gcov version they emulate * Translate short options into their equivalent long option capabilities Signed-off-by: Jonah Petri commit a74bdeeae0383b197b1dafa44d01a54129fb3d7c Author: Peter Oberparleiter Date: Wed Jan 8 13:26:04 2014 +0000 genhtml: Reduce hash copying while adding up files Reduce copying effort and memory usage. Based on similar patch for lcov by olly@survex.com. Signed-off-by: Peter Oberparleiter commit c6b4d91fdf667cfca17213742e2e04f6281ebed4 Author: Olly Betts Date: Wed Jan 8 13:14:05 2014 +0000 lcov: Avoiding copying hashes passed to add_counts function This patch reduces memory usage - without it lcov was failing every time for me with out of memory errors in a VM with 1GB of RAM and 1GB of swap, but with it lcov completes every time. It's presumably also faster to avoid these copies. Signed-off-by: Olly Betts commit cf6f2e685510da62bd2eb1f386f71d57c41f4594 Author: Peter Oberparleiter Date: Fri Dec 13 16:09:05 2013 +0000 geninfo: Tolerate garbage at end of gcno file Some versions of gcc produce garbage at the end of a gcno file when recompiling a source code file after removing some lines. This patch makes geninfo's gcno file parser more robust by assuming end-of-file when it finds a record that extends beyond the end-of-file. Signed-off-by: Peter Oberparleiter commit 14286b29d076208452da6021c792ebf43552ac2c Author: Peter Oberparleiter Date: Fri Dec 13 15:23:27 2013 +0000 geninfo: make gcov tool version detection more robust Don't consider gcov tool version information in parenthesis when determining the gcov tool version. This fixes problems where the version string contains a different version number in parenthesis before the actual gcov version. Signed-off-by: Peter Oberparleiter commit 0bde87338cd155af46804d77701c93ef263c3d53 Author: Sebastian Stigler Date: Fri Dec 13 15:09:58 2013 +0000 geninfo: add exclude marker for branch coverage Sometimes it can be helpful to generally use branch coverage but to disable it for some lines of code without excluding the line or function coverage too. For example if you make heavily use of assertions in your code (which is generally a good idea) you will see that for each 'assert(...)' exist one branch which is taken and one that is not. Similarly you can see the same phenomenon for 'delete' in C++ code. If you use the 'LCOV_EXCL_LINE' marker in such a situation both of these branches will be omitted from the output. But in doing so, you loose the ability to determine if this peace of code is genuine 'dead code' or not because the line coverage is omitted too. The newly introduces 'LCOV_EXCL_BR_LINE', 'LCOV_EXCL_BR_START' and 'LCOV_EXCL_BR_STOP' marker address this problem. The usage is similar to the 'LCOV_EXCL_LINE' etc. markers. Signed-off-by: Sebastian Stigler commit 119be727596f567e83b03de384b4150b926911a3 Author: Peter Oberparleiter Date: Thu Dec 12 14:58:44 2013 +0000 geninfo: Fix handling of non-english locales geninfo expects gcov output in the default C locale. This isn't always given, for example when running in an environment where variable LANGUAGE is set to a non-english locale. In such cases gcov output cannot be correctly parsed, resulting for example in the absence of branch coverage data. gcov uses gettext() for writing internationalized messages. The info page for gettext mentions the order in which locale-defining environment variables are evaluated: LANGUAGE LC_ALL LC_MESSAGES LANG In addition, gettext implements special handling where LC_ALL=C takes precedence over LANGUAGE. geninfo currently only specifies LANG=C. Fix the issue by specifying LC_ALL=C instead. Based on fix suggestion by Sebastian Stigler. Reported-by: Sebastian Stigler commit 0f7bb3ebc8487b83ce9b7047c81a3655135876ea Author: Peter Oberparleiter Date: Mon Dec 9 15:49:35 2013 +0000 lcov: Added contribution guidelines commit f83688fe27f133ef02e9ab47a435d6a5d2074932 Author: Peter Oberparleiter Date: Fri Aug 2 07:29:20 2013 +0000 geninfo: fix --no-external not working with --initial When running lcov --capture --initial together with --no-external. the --no-external has no effect. Fix this by applying the external filtering also for graph files. Reported-by: malcolm.parsons@gmail.com commit 6a8a678046bd75aa81d30484b1817425022d71e5 Author: Peter Oberparleiter Date: Mon Jul 1 11:49:46 2013 +0000 lcov: fix --config-file not being passed to geninfo Calling lcov to capture coverage data while specifying --config-file will result in the configuration directives of that file not being used during data collection. Fix this by ensuring that --config-file is passed on to geninfo. Reported-by: liuyhlinux@gmail.com commit c3be5b6859ef280b469b6b75cf4709fc35f91ced Author: Peter Oberparleiter Date: Thu May 2 11:02:24 2013 +0000 lcov: fix whitespace handling in --rc command line option Specifying blanks around --rc options results in the options not being correctly recognized, for example: This doesn't work: geninfo . -o - --rc="geninfo_adjust_src_path = /tmp => /usr" This works: geninfo . -o - --rc="geninfo_adjust_src_path=/tmp => /usr" Fix this by automatically removing whitespaces at the start and end of --rc options and values. commit 4699f8d391325335777ed234e388be2e2f87478c Author: Peter Oberparleiter Date: Fri Apr 12 07:51:34 2013 +0000 README: improve usage hint commit 36e0539737198ad1bee51103f47842f13c575239 Author: Peter Oberparleiter Date: Wed Mar 13 10:28:07 2013 +0000 genhtml: add time to date string Add the current time to the date information in the HTML output generated by genhtml. This way users can differentiate results when creating HTML output multiple times a day. Based on patch by sylvestre@debian.org. commit 38fbe93c8cd8402be8e4821825fdeeaa23e8367c Author: Peter Oberparleiter Date: Fri Feb 22 14:09:08 2013 +0000 geninfo: don't warn about missing .gcov files Newer versions of gcc remove .gcov files for source files that do not contribute instrumented lines. Remove the WARNING: no data found for file.c warning that geninfo issues in this case. commit 29346542c30af221a2ffdfe097fbd858044b712a Author: Peter Oberparleiter Date: Fri Feb 1 11:44:03 2013 +0000 genhtml: fix handling of user-specified prefixes with trailing / A trailing / in a user-specified prefix is not correctly recognized. Fix this by removing any number of trailing / in a user-specified prefix. Reported by ahmed_osman@mentor.com. commit 5241e2afadca5f172bd0b8cafe61e20d2153f0bf Author: Peter Oberparleiter Date: Wed Jan 30 11:46:42 2013 +0000 lcov: fix bug when converting function data in --diff operation When a patch is applied to a tracefile using command line option --diff and the patch changes the list of functions, the operation aborts with the following error: lcov: Use of freed value in iteration at lcov line 3718. Fix by applying missing calls to keys() when iterating function data hashes. Reported by Nasir.Amanullah@us.fujitsu.com. commit 9ce8d8cb4f978eb80fb88ecafd52e869fab75d8f Author: Peter Oberparleiter Date: Fri Jan 25 16:30:25 2013 +0000 lcov/genhtml: fix outdated comment regarding data structure commit c85e73a36e3f8c4e7fab888ac1536bee94a6fe56 Author: Peter Oberparleiter Date: Fri Jan 25 16:29:30 2013 +0000 genhtml: merge function data during demangling Merge function execution counts when multiple function names demangle to the same name. commit 2dfafc99c1eccbb81066436845e06a868eb3c434 Author: Peter Oberparleiter Date: Fri Jan 25 11:46:28 2013 +0000 genhtml: improve function table sorting In the function table view, the initial view should show the functions sorted by execution count because - unlike with file names - the function name is not a natural order for functions (the line number would be, but that is not available). Also sort functions with the same execution count alphabetically for a stable order. Base on a suggestion by paul.bignier@hotmail.fr. commit 331a29011709a27d2ec11c6cbd6ac51dfdaf70c6 Author: Peter Oberparleiter Date: Wed Jan 23 16:52:06 2013 +0000 genhtml: consolidate calls to c++filt When using --demanglecpp, call c++filt only once instead of per function. This approach can reduce the run-time for source files with a lot of overloaded functions significantly. Based on idea by olly@survex.com. commit 49b877160b1d28cd6c3d8332d5d47c9c74420070 Author: Peter Oberparleiter Date: Thu Jan 10 09:02:32 2013 +0000 geninfo: make empty data directories non-fatal Emit a warning when no data file is found in a data directory to allow processing of additional directories. Based on suggestion by rich_drake@yahoo.com. commit 3836c162c2864ed180df7d80fa03c70d17102edc Author: Peter Oberparleiter Date: Tue Nov 13 09:58:53 2012 +0000 geninfo: fix parsing of gcc 4.7 gcov format GCC 4.7 changes the gcov format for lines which can only be reached by exceptions to "=====" instead of "#####". This results in the following warning: geninfo: Argument "=====" isn't numeric in numeric gt (>) at geninfo line 1281. Fix this by handling "=====" correctly. commit b932f94cc83c3df169c76689533336bba4de4dba Author: Peter Oberparleiter Date: Wed Oct 10 09:14:17 2012 +0000 lcov.spec: back to CVS version commit 6af00fa26e1a91a39c873ff9fa6df7fb8830ec42 Author: Peter Oberparleiter Date: Wed Oct 10 09:12:42 2012 +0000 lcov.spec: fix Perl dependency commit 4eac16e93db328e86e44da40e3d5e96a0301d361 Author: Peter Oberparleiter Date: Wed Oct 10 08:36:16 2012 +0000 lcov: update CVS version to 1.11-pre1 commit b5c1bdddd1380be3ad12952ed2747df3744e227e Author: Peter Oberparleiter Date: Wed Oct 10 08:20:21 2012 +0000 lcov: finalizing release 1.10 commit 089861768a94d0f6e827539c828f19141092f529 Author: Peter Oberparleiter Date: Wed Oct 10 08:07:54 2012 +0000 CHANGES: update commit 9037de17458c5d9767d201bd0599d40347a9bc41 Author: Peter Oberparleiter Date: Wed Oct 10 08:07:01 2012 +0000 genhtml: handle source files in root directory gracefully commit 68dd0f19da0d8d6e82375e09b97f7ffc22847db4 Author: Peter Oberparleiter Date: Tue Oct 9 13:58:22 2012 +0000 geninfo: add automatic detection of base directory Add a heuristic to automatically determine the base directory when collecting coverage data. This heuristic should cover many, if not most cases of build systems moving files around during compilation (e.g. libtool, automake, etc.). The heuristic can be enabled or disabled using the configuration file directory 'geninfo_auto_base'. commit 631d2b11bfde56ffca4568382abf5d90653c4141 Author: Peter Oberparleiter Date: Mon Oct 8 15:03:23 2012 +0000 geninfo: fix missing line data after last commit commit b1e14c4a1a0f3ccaad0c665f439624cf4588a68d Author: Peter Oberparleiter Date: Mon Oct 8 13:02:45 2012 +0000 lcov: add missing help text for option --rc commit a432efff6ee8485ec0724aca4eae79a4c390a328 Author: Peter Oberparleiter Date: Fri Oct 5 15:53:09 2012 +0000 lcov: updated CHANGES file and copyright years commit 897322ecdb858f18e4a12f4716bbb08c067b6c9c Author: Peter Oberparleiter Date: Fri Oct 5 15:20:41 2012 +0000 geninfo: fix warning about unhandled .gcov files gcov will sometimes create .gcov files that contain no instrumented line. When geninfo reads .gcno files it filters out such files, resulting in the following warning: geninfo: WARNING: cannot find an entry for #path#to#file.gcov in .gcno file, skipping file! Avoid this warning by not filtering out non-instrumented lines. commit 37d381ae99a66f59ea55d966f1da13a726d2efe8 Author: Peter Oberparleiter Date: Fri Oct 5 15:09:58 2012 +0000 genhtml: fix source path prefix calculation Fix the following problems of the algorithm used to identify an optimal source path prefix: - the last two path components (filename and first parent directory) are ignored when trying to identify the optimal prefix - if a path prefix matches a longer path prefix, the weight of the filenames associated with the latter is incorrectly attributed to the former commit 263de2b40e21193ef8d11e899eb55aa52b17225d Author: Peter Oberparleiter Date: Fri Oct 5 12:00:57 2012 +0000 lcov: set default for branch coverage data to disabled Collecting branch coverage data can significantly slow down coverage data collection and processing of data files. Assuming that most users are more interested in line/function coverage, change defaults to not collect/process branch coverage data. Users can still override this default using lcov_branch_coverage=1 in the lcovrc file or command line option --rc lcov_branch_coverage=1 commit 7e04a152683ff66e24b87f2125474c6765d4524b Author: Peter Oberparleiter Date: Fri Oct 5 11:49:30 2012 +0000 geninfo: fix problems with adjust_src_path option Fix the following problems with adjust_src_path: * specifying --compat libtool=on and geninfo_adjust_src_path unexpectedly sets --compat libtool=off * path components that are assembled from sub-directory names are not correctly adjusted commit 74e4296b6e2a0b0f164c6828c28cc82449344f08 Author: Peter Oberparleiter Date: Fri Oct 5 08:23:06 2012 +0000 lcov: add setting to disable function and branch coverage Add two new configuration file settings: * lcov_function_coverage and * lcov_branch_coverage When set to zero, lcov will skip the corresponding coverage data type from being collected or processed, resulting in reduced memory and CPU time consumption and smaller data files. commit 37bc1a1a5f721c6b88fff4c63121c1cbb794c14f Author: Peter Oberparleiter Date: Tue Oct 2 14:29:57 2012 +0000 lcovrc: clarify meaning of geninfo_external in man page commit fc4b9e21efe8f3409d9b0b90cfe7a3e8bc59a74c Author: Peter Oberparleiter Date: Tue Oct 2 09:12:38 2012 +0000 geninfo: fix processing of pre-3.3 gcov files When trying to collect coverage data for programs compiled with GCC versions prior to 3.3, geninfo skips each data file with the following warning: geninfo: WARNING: cannot find an entry for test.c.gcov in .bb file, skipping file! Fix this by deriving the source code filename from the gcov filename in case the gcov files do not follow the GCC 3.3 format. Reported-by: georgysebastian@gmail.com commit d1014dfcabfee2f305278a14ec8e5343e3889139 Author: Peter Oberparleiter Date: Fri Aug 24 11:43:18 2012 +0000 lcov: fix problems with --rc option Fix error "Invalid option linkage for \"rc=s%\"" when running lcov with an older version of the Getopt::Long module. Also pass --rc options through lcov to geninfo. commit a9f08b79e2e7ec2b4a5c9ad27a077df8dfb46890 Author: Peter Oberparleiter Date: Tue Jul 24 15:41:38 2012 +0000 geninfo: implement rc option geninfo_adjust_src_path Provide a new lcovrc file option geninfo_adjust_src_path that users can use to change incorrect source paths. Inspired by patch by ammon.riley@gmail.com. commit 108f805788590defda99fdf252bfb71cb749f31e Author: Peter Oberparleiter Date: Thu Jul 19 13:12:35 2012 +0000 lcov: implement command line option --rc Users can now use command line option --rc to override configuration file directives. commit eeeeeca74706e88a9b8ecfef2bb3451957512e20 Author: Peter Oberparleiter Date: Wed Jul 18 12:56:21 2012 +0000 lcovrc: add description for geninfo_compat setting commit f842e46149b48ff316e80f68f630bf94085e4d19 Author: Peter Oberparleiter Date: Wed Jul 18 12:40:56 2012 +0000 lcov: improve --compat description commit 392a690ba31092857f7d21d0008783d87954ebce Author: Peter Oberparleiter Date: Wed Jul 18 12:13:00 2012 +0000 lcov: add missing documentation for --compat option Add missing sections in the geninfo and lcov man-pages for the newly introduced command line option --compat. Also set the default value for the hammer compatibility mode to 'auto' to keep the behavior of previous releases. commit 691cab3e3aaebc295c2cfe91c43c6a7c48f1ec2b Author: Peter Oberparleiter Date: Wed Jul 18 10:40:12 2012 +0000 lcov: fix extra closing parenthesis in comment commit cef6f0ff8baa9b2b3dfb437463e7a88d3380b555 Author: Peter Oberparleiter Date: Tue Jul 17 11:37:13 2012 +0000 lcov: make 0%/100% exclusive to no/full coverage rate Ensure that coverage rates 0% and 100% are only used when no or all lines/functions/branches are hit respectively. This approach is implemented to allow better identification of boundary cases, and to be in accordance with the behavior of the gcov tool. Based on suggestions by: Paul.Zimmermann@loria.fr and vincent@vinc17.net commit 9cec8f7e332258c9128f1c53d61acb9f0bc17085 Author: Peter Oberparleiter Date: Wed Jul 11 14:09:27 2012 +0000 geninfo: more improvements to the .gcno format auto-detection Suggestions by garnold@google.com: - rename command line setting - simplify logic commit 0bbca3bd0c1ad3e3d3fd0b6eebfc3afbbc212a85 Author: Peter Oberparleiter Date: Fri Jul 6 14:29:27 2012 +0000 geninfo: rename compatibility setting to compatibility mode commit f30fb978662996e29517c733218292a91f5fd12b Author: Peter Oberparleiter Date: Fri Jul 6 09:03:27 2012 +0000 geninfo: improve detection of gcc 4.7 function records Suggestions by garnold@google.com: - perform detection only once - add warning in case detection is off but overlong strings are found Misc: - add help text for --compat - isolate detection heuristic into separate function - rename corresponding compatibility setting to "gcc_4_7" - allow "android_4_4_0" as alias for "gcc_4_7" commit 01321c3f170e5d24ffb3bb998441c99f5b775836 Author: Peter Oberparleiter Date: Wed Jul 4 16:06:10 2012 +0000 geninfo: enable auto-detection of gcc-4.7 function record format gcc-4.7 introduced a modified function record format. This format is in use by android toolchains and has also been ported to some pre-4.7 versions of gcc. Introduce a heuristic-based auto-detection to correctly handle .gcno files in these cases. commit d929600a0e2133168085e5ddea7ee832afd902b7 Author: Martin Hopfeld Date: Fri Jun 8 14:19:49 2012 +0000 geninfo: Make geninfo work more reliably on MSYS Using the lcov tools on Win7 with MSYS and MinGW 4.5.1/4.5.2 raised some issues for us: geninfo created in the for one source file in the 'SF:' line paths starting with a lowercase drive letter and sometimes starting with uppercase drive letters. This lead to inaccurate coverage results on the MSYS platform. This patch fixes this issue. commit 5b2751854aa19e6443fdc5fecc139595988d1e99 Author: Peter Oberparleiter Date: Mon May 7 16:04:49 2012 +0000 lcov: add perl version dependency to RPM spec file lcov CVS (1.10 pre) seems to be broken on MSYS with perl 5.6.1. The issue is the following: genhtml: Unknown open() mode '>>&' at /usr/bin/genhtml line 5512. $> perl genhtml --version genhtml: LCOV version 1.10 pre (CVS 1.58) $> perl --version This is perl, v5.6.1 built for msys Fortunately perl v5.8.8 is available for MSYS and genhtml works like a charm with that 'new' version. Reported-by: Martin Hopfeld commit 83957a145d243cad0f8060e4a9ccc6cb8ed8fc09 Author: Peter Oberparleiter Date: Tue Apr 10 11:48:52 2012 +0000 geninfo: add support for gcc 4.7 .gcno file format Based on patch by berrange@redhat.com. commit 91c91dbc63d1e880d106919300c2fb37737697b0 Author: Peter Oberparleiter Date: Fri Jan 20 11:53:57 2012 +0000 lcov: add new command line option --compat Add new option to lcov and geninfo to specify compatibility settings. Supported settings: libtool: same as --compat-libtool hammer: gcc3.3 hammer patch compatibility android_4_4_0: android toolchain 4_4_0 compatibility commit 9588355790a302da680eff2f664058f78439a03e Author: Peter Oberparleiter Date: Thu Aug 11 08:29:21 2011 +0000 lcov: fix problem with Objective-C functions Fix geninfo not recognizing function entries for Objective-C functions. Based on patch by abrahamh@web.de: current version of lcov unfortunately not support Objective-C files. In details the count of tested function is zero always and the annotated lines have an offset by one if the Objective-C method have one ore more arguments. commit e1acd78d1e88fe51aad96badf32555c470ee029b Author: Martin Hopfeld Date: Mon May 23 08:03:13 2011 +0000 geninfo: Make geninfo handle MinGW output on MSYS. This patch converts path mixtures from MinGW when running on MSYS to correct MSYS paths. In solve_relative_path() an additional conversion step will be inserted when running on MSYS. This will extract the drive letter and convert the remaining path from Windows pathnames to Unix Paths, which are used by MSYS. Additionally, if no drive letter is found, the (relative) path is converted to Unix style. There may be the case where Windows and Unix path separators are intermixed within one path string. commit ed161e3db5cd5a7c6c8b2113930c729f001cdd4e Author: Peter Oberparleiter Date: Thu Dec 16 08:11:22 2010 +0000 genpng: handle empty source files Generating an overview PNG image for an empty source code file fails. Handle this case by assuming a single empty line when run for an empty source code file. Reported by: sylvestre@debian.org commit 95e2c5c337d281b4e88144d95d29bbec183c8728 Author: Peter Oberparleiter Date: Tue Dec 7 08:40:09 2010 +0000 genhtml: add note to further explain branch coverage output commit b1c66916151dd4b20998c79f81edf174659ebb14 Author: Peter Oberparleiter Date: Tue Dec 7 08:29:45 2010 +0000 genhtml: fixed incorrect description of default coverage rates commit 1994be7d8ed472772b884063af74235f2f25ab39 Author: Peter Oberparleiter Date: Fri Nov 19 16:33:25 2010 +0000 geninfo: add missing man page sections Add sections describing options --derive-func-data and --no-markers to the geninfo man page. commit 01a393ef76092e43ebd2d8bf7892ebf375481a84 Author: Peter Oberparleiter Date: Fri Nov 19 16:15:27 2010 +0000 geninfo: remove help text for unimplemented parameter Parameter --function-coverage was removed but the help text still mentions it. Fix this by removing the option from the help text as well. commit b92f99d9db0af131080c462300dc9baf292a8ff6 Author: Peter Oberparleiter Date: Fri Nov 19 16:00:22 2010 +0000 genhtml: handle special characters in file and directory names HTML special characters (e.g. '<') found in file or directory names are not correctly shown in HTML output. Fix this by correctly escaping such characters. commit 17e158d4569d25218e79901e2d8cd03bfc7752fc Author: Peter Oberparleiter Date: Fri Nov 19 15:45:01 2010 +0000 gendesc/genhtml/geninfo/genpng/lcov: handle '<' in filenames Use 3-arg open mode to prevent that a special character (e.g. '<') found in a user-specified filename interfers with the required open mode for that file. commit b87e40e475c560bdc88206df4de6dc8cf094d91f Author: Peter Oberparleiter Date: Fri Nov 19 15:11:53 2010 +0000 geninfo: ignore .gcov files The gcov tool will sometimes create a file .gcov for code which was added by gcc itself during compilation. Since there isn't any source available for such code, geninfo will fail. Fix this by skipping these files while capturing code coverage data. commit 398d8f385423927b5675c1429f58c67b6a89a1a8 Author: Peter Oberparleiter Date: Thu Oct 28 14:17:57 2010 +0000 geninfo: add function comment Add comment explaining data structures used by function derive_data. commit f5c2072e0e7195d35455db50705884e7f6c5fbe5 Author: Peter Oberparleiter Date: Thu Oct 28 14:16:34 2010 +0000 geninfo: apply exclusion marker to derived function data When option --derive-func-data is used together with exclusion markers, function data for excluded lines is still included. Fix this by only deriving function data for lines which are instrumented and not excluded. Reported by: bettse@gmail.com commit 82280b8a5a78e8a147c333c8850a556729d9d96d Author: Peter Oberparleiter Date: Tue Aug 31 08:19:03 2010 +0000 geninfo: improve --debug output commit 6375a03010cb1bb22490b9d19a176188940e2f8b Author: Peter Oberparleiter Date: Tue Aug 31 08:17:23 2010 +0000 gcov: add configuration file option to not use gcov's -a option lcov calls gcov while specifying its --all-blocks option to get more detailed branch coverage data per line. It turns out that this option is broken on many versions of gcov, resulting in an endless loop while processing some gcov data files. There's also a slight performance penalty when specifying -a. lcov users can opt to not use the -a option by setting configuration option geninfo_gcov_all_blocks to 0 in the lcovrc file. commit 7706fb73ebef8060fbbd92c0e08b5d68a2cd284e Author: Peter Oberparleiter Date: Tue Aug 24 16:15:53 2010 +0000 lcov: add option to specify a configuration file Provide an option for users to specify a configuration file to lcov. This option may be useful when there is a need to run several instances of a tool with different configuration file options in parallel. commit a404dafc2da12608a936afeb095d68410fa49b0a Author: Peter Oberparleiter Date: Mon Aug 23 16:14:37 2010 +0000 lcov: add option to display summary coverage information Provide an option for users to determine the summary coverage information of one or more tracefiles. Example output: Summary coverage rate: lines......: 26.0% (78132 of 300355 lines) functions..: 34.9% (8413 of 24081 functions) branches...: 16.9% (32610 of 193495 branches) commit 526b5b6a43f2b29f11eb02c1dd8f645293d8c295 Author: Peter Oberparleiter Date: Mon Aug 23 14:47:43 2010 +0000 lcov: add option to exclude external files Implement an option for users to specify that external source files should be excluded when capturing coverage data. External source files are files which are not located in the directories specified by the --directory and --base-directory options of lcov/geninfo. commit c2255a0344648dc6eaef0189c53f345fdc70ed4e Author: Peter Oberparleiter Date: Fri Aug 20 14:58:48 2010 +0000 lcov: pass --no-recursion to geninfo When specifying --no-recursion, make sure that the option is also passed to the helper tool geninfo. commit 83543f3d21b5a5496b57c8d73e8e9c1819f82f34 Author: Peter Oberparleiter Date: Fri Aug 20 14:31:59 2010 +0000 genhtml: fix HTML page title for directory pages commit b77df8ef1a69de3809e0b0bfa5cbbe5a84f313ae Author: Peter Oberparleiter Date: Fri Aug 20 14:27:19 2010 +0000 genhtml: make HTML charset specification customizable Provide a configuration file setting to adjust the charset specification used by all generated HTML pages. Also change the default charset to UTF-8. commit 1ff260462a67c440dc709d34c1fadf7d64760120 Author: Peter Oberparleiter Date: Fri Aug 20 13:14:50 2010 +0000 lcov: follow Perl naming guidelines commit f637eb8c6ecb793b64eeb6bea57c6be8501d1484 Author: Peter Oberparleiter Date: Fri Aug 20 13:08:25 2010 +0000 genhtml: add --ignore-errors option Provide a means for users to specify that genhtml should not abort if it cannot read a source code file. Also make handling of --ignore-errors parameter consistent accross lcov, geninfo and genhtml. commit 617bced393d5bb97e3409ec140768d9c8a2e2bfb Author: Peter Oberparleiter Date: Fri Aug 6 11:25:12 2010 +0000 lcov: update CVS version to 1.10 commit 4dcb4f0ed014ca0f49859ef84fc9ced650f6deb8 Author: Peter Oberparleiter Date: Fri Aug 6 11:14:38 2010 +0000 lcov: finalizing release 1.9 commit 594779e047eed2f534905ac40912969955d3797f Author: Peter Oberparleiter Date: Thu Aug 5 16:17:44 2010 +0000 lcov: update CHANGES file in preparation of new release commit fbbd9034e7a4ea4bc59342b22bfbe9612dd4bdb8 Author: Peter Oberparleiter Date: Thu Aug 5 15:11:56 2010 +0000 lcov: introduce configuration file parameters for list output Make some aspects of list output customizable via configuration file parameters. Also introduce special handling, if the root directory is chosen as prefix. commit c6e783c1a1d3fb6db7419af95f9e2dcb89836fe9 Author: Peter Oberparleiter Date: Thu Aug 5 14:07:35 2010 +0000 lcov: switch coverage rate and number columns in list view To be more consistent with the order of output in the "Overall coverage rate" case, rates are now shown first in the list output. commit 3c87b66c68c2e06811c9be479c6813cb409e5461 Author: Peter Oberparleiter Date: Thu Aug 5 11:22:12 2010 +0000 lcov: fix display of total line coverage rate in list view commit 3cb6bc4ae0ef34aa63931d63f659f1ef43804c77 Author: Peter Oberparleiter Date: Wed Aug 4 16:15:19 2010 +0000 lcov: more lcov --list improvement Further improve list output to increase readability. commit dd98ff68ad143b985a728fc585c86d69e6027bd8 Author: Peter Oberparleiter Date: Wed Jul 28 14:49:47 2010 +0000 lcov: minor list improvement commit d4778c75ce8cf3c9d44607b6fd0e385db71126dd Author: Peter Oberparleiter Date: Wed Jul 28 14:48:25 2010 +0000 geninfo: remove unneeded functions commit 65a15afef3430c49c9c7c0d151cc2afec5fc83cc Author: Peter Oberparleiter Date: Wed Jul 28 14:19:09 2010 +0000 geninfo: print note on branch coverage data only once commit bd8ab633298ec27acf5f7db4b2cc4766baf1f153 Author: Peter Oberparleiter Date: Wed Jul 28 14:17:59 2010 +0000 geninfo: remove incorrect overall coverage rate calculation geninfo output showing the overall coverage rate of its current operation is incorrect since it may count lines, functions and branches for included files multiple times. Remove the output and associated code until a fixed version is available. commit 8c54de96a1326b7ee0632773816c52eda43393e8 Author: Peter Oberparleiter Date: Wed Jul 28 13:56:26 2010 +0000 lcov: more list output fixes commit 7e5fa9900d991320677c381db747c764495b2cc2 Author: Peter Oberparleiter Date: Wed Jul 28 13:52:01 2010 +0000 lcov: fix list output Fix list output for directories with short filenames. commit badd4790c70bd8ef8b991a9d56d0e062b28006a8 Author: Peter Oberparleiter Date: Mon Jul 26 13:33:18 2010 +0000 lcov: fix problem when using --initial and --kernel-directory Fix a problem in lcov that resulted in --kernel-directory options being ignored when specifying --initial at the same time. Reported by hjia@redhat.com. commit a06c2038babb2f6d3e0a634cd298b0434041f834 Author: Peter Oberparleiter Date: Mon Jul 19 16:06:15 2010 +0000 genhtml: change wording for branches which were not executed Since gcov sometimes reports both branches which were never executed as well as branches which were executed in a single block, the wording of the HTML alt text needs to be adjusted accordingly. commit e6b2491823ffd84c85406145031646af675170ee Author: Peter Oberparleiter Date: Mon Jul 19 15:50:02 2010 +0000 geninfo: handle branches in unnamed blocks gcov will sometimes report branches outside of a block. In that case, account these branches to a special block so that they are not accidentally merged with subsequently reported blocks. commit d6c82edf2117ce8b6232c998baf06c7a87269081 Author: Peter Oberparleiter Date: Mon Jul 19 15:23:10 2010 +0000 genhtml: fix branch formatting code Fix the vertical alignment of the HTML representation of branches in the source code view. commit 44ac74a47e25064ad1b421f65a28d057fdb9925d Author: Peter Oberparleiter Date: Mon Jul 19 14:27:08 2010 +0000 lcov: improve list output Improve list output by separating directory and file names. Also provide an option to show full path names. commit 0ab6f7507f3c4f074bec41e571ff1afbeb943185 Author: Peter Oberparleiter Date: Mon Jul 19 12:12:43 2010 +0000 genhtml: fix large numbers being shown as negative in html output genhtml uses a "%d" format string for printing execution counts. For counts exceeding integer range, the output becomes negative. Fix this by using the "%.0f" format string instead. Reported by kkyriako@yahoo.com. commit bbf0ef40a51dd716c544f91576cffde7986bb6ec Author: Peter Oberparleiter Date: Mon Jun 7 12:22:18 2010 +0000 geninfo: ensure that exclusion markers apply to --initial Fix a problem where exclusion markers are ignored when gathering initial coverage data. Problem was reported by ahmed_osman@mentor.com. commit b371fc59fa52f7176f62f382457fba498f39f4b2 Author: Peter Oberparleiter Date: Tue Jun 1 13:48:29 2010 +0000 lcov: fix problem with relative path names Fix a problem where coverage data is missing because gcov produces output files starting with a dot. Problem reported by weston_schmidt@open-roadster.com. commit 93c70ddd0edbc2b0addf9d135dfd76871cc7a160 Author: Peter Oberparleiter Date: Sun Feb 28 20:57:37 2010 +0000 lcov: fix problem with relative paths in build paths When binaries are built using relative paths, lcov cannot find any coverage data. Instead, warnings similar to the following are printed: geninfo: WARNING: cannot find an entry for ^#src#test.c.gcov in .gcno file, skipping file! The reason for this is that File::Spec::rel2abs does not remove ../ from paths which results in lcov not being able to match the relative and absolute versions of the corresponding filenames. Fix this by using the internal function solve_relative_path instead. commit fad24a75cc69364d002d40e4fb75736b0efbdb37 Author: Peter Oberparleiter Date: Sun Feb 21 14:57:52 2010 +0000 geninfo: write all debugging output to STDERR commit c0943385fa0acb927f63f9f78c9aeaebe3a8ece1 Author: Peter Oberparleiter Date: Sun Feb 21 14:56:46 2010 +0000 geninfo: fix problem with some .gcno files Some .gcno files contain more data in a line record than expected. Skip unhandled bytes of a .gcno file record. This prevents the following unexpected error message: geninfo: ERROR: file.gcno: reached unexpected end of file commit 4b9ee7598e991b503425148eb43a35de2702aded Author: Peter Oberparleiter Date: Sun Feb 7 13:07:09 2010 +0000 lcov: add COPYING file commit de0e43a098ade45d6624ea43a53e6fad9a176469 Author: Peter Oberparleiter Date: Fri Jan 29 11:07:25 2010 +0000 lcov: update CVS version to 1.9 commit 4a33269fa3a73ea2577f7616d90bd3f1d14ae460 Author: Peter Oberparleiter Date: Fri Jan 29 10:09:53 2010 +0000 lcov: finalizing release 1.8 commit 310ffb28d8847f96e02b5a5db3d16bdcb406a876 Author: Peter Oberparleiter Date: Fri Jan 29 10:01:35 2010 +0000 lcov: updated CHANGES file commit 9e12808e6108e05dca42b5e682bd8be121f3608d Author: Peter Oberparleiter Date: Fri Jan 29 09:21:22 2010 +0000 genhtml: use sans-serif font for function table commit 71baabb6a1c15228213f8b25359346ee202300ce Author: Peter Oberparleiter Date: Fri Jan 29 09:12:55 2010 +0000 lcov: improve list output commit cc61a28dbc3c46ac84340141fafbfa559e1bf318 Author: Peter Oberparleiter Date: Fri Jan 29 08:56:19 2010 +0000 lcov: fix overall rate display for tracefiles with more than one testcase commit b89028529db5110b3b76d117df788768a593d7dd Author: Peter Oberparleiter Date: Fri Jan 29 08:44:47 2010 +0000 lcov/genhtml: fix warning while merging branch data commit b7c69f31d9b1bfbd4bfc0fcb880cb8e514bcdb3f Author: Peter Oberparleiter Date: Thu Jan 28 15:59:23 2010 +0000 lcov: fix branch coverage related issues - warnings when combining certain combinations of branch data - branches are not merged correctly when multiple input files are specified to genhtml or when lcov -a is used commit 817875459df122fa3536a5e57c05ddfae19a089e Author: Peter Oberparleiter Date: Wed Jan 27 16:37:50 2010 +0000 gendesc: fix problem with single word descriptions commit 33f60f48747b5ba12a6fdfb505bb662c922496bd Author: Peter Oberparleiter Date: Wed Jan 27 12:10:04 2010 +0000 lcov: remove temporary files when creating a package commit 6775457cbd3fa86acba4655d77b4ba2054b13253 Author: Peter Oberparleiter Date: Wed Jan 27 12:00:05 2010 +0000 lcov: correctly retain information about converted test data commit f4d13eccc54f31a53ad109c3c4b86e4b52d6dfcb Author: Peter Oberparleiter Date: Wed Jan 27 10:17:43 2010 +0000 lcov. fixed overview output for function data commit aa00c65b7514c93320c1c787b848c8277593dcb0 Author: Peter Oberparleiter Date: Tue Jan 26 09:36:19 2010 +0000 genhtml: don't use too much gcc-specific terms (basic block -> block) commit 3562f60b9500d8ad167c4629e9d95485308aa665 Author: Peter Oberparleiter Date: Fri Jan 22 16:17:37 2010 +0000 lcov: consolidate coverage rate classification limits Classifying coverage rates per coverage type (line, function or branch coverage) is not useful in most cases. Also the respective color legend takes up too much space in the HTML output. Remove function and branch coverage rates from the documentation and from the color legend. Instead the original limits will be applied to those coverage types as well. The per type rates can still be used if required but it is recommended to only use one rate set. commit d77dc6a0adf259e322ac9f35c93241d446269a5b Author: Peter Oberparleiter Date: Fri Jan 22 16:16:47 2010 +0000 lcov: minor code cleanup - remove unused function definitions and declarations - remove unused CSS declarations - add missing function declarations - fix function prototypes commit b3243d1fdc17571ca9b1ed6a1ea975a9b3f1b86b Author: Peter Oberparleiter Date: Fri Jan 22 16:16:15 2010 +0000 geninfo: consolidate similar functions commit 739e2bca054c69975594c2570049e8aa9ae1b5ce Author: Peter Oberparleiter Date: Fri Jan 22 16:15:35 2010 +0000 lcov: add coverage result output to more operations commit 0a31d3c0696015c5e4878e821529eba45451c3dd Author: Peter Oberparleiter Date: Fri Jan 22 16:14:57 2010 +0000 lcov: minor cosmetic HTML changes - top level view is now named "top-level" - use sans-serif font for coverage values in file list - use smaller font for show/hide details link - use smaller font for function/source view link - use smaller font for show descriptions link commit b631fa0cb9aabdf18f9365423f0b0bf85d6b8e16 Author: Peter Oberparleiter Date: Fri Jan 22 16:14:27 2010 +0000 lcov: improve color legend Move color legend closer to the table containing coverage rates. commit 2aeeeafb31c36ccd1a51051f040e29a9fcf59df2 Author: Peter Oberparleiter Date: Fri Jan 22 16:13:58 2010 +0000 lcov: implement branch coverage commit 49dfe22f41b6c3edcb774dfb89b1a807ce7aee6c Author: Peter Oberparleiter Date: Fri Jan 22 16:13:34 2010 +0000 genhtml: implement branch coverage commit 6aa2422401bb854c9710f5ed2936f06e487848c5 Author: Peter Oberparleiter Date: Fri Jan 22 16:13:07 2010 +0000 geninfo: implement branch coverage commit ca2c9781b0a512bd6789eac2b6840405e2d87330 Author: Peter Oberparleiter Date: Fri Jan 22 16:12:27 2010 +0000 geninfo: consolidate handling of extra gcov parameters commit 9d9c964eb6ece00b15ef068f176c68cb0eedfda0 Author: Peter Oberparleiter Date: Thu Jan 21 11:26:34 2010 +0000 lcov: minor fix for lcov --diff commit 4306f81d1e8446a89fe83d20cd71abe075a3cd61 Author: Peter Oberparleiter Date: Thu Jan 21 10:23:35 2010 +0000 lcov: improve lcov --list output commit 3242ce1bae94cfd859c3bc964fab11f85bd7d1ed Author: Peter Oberparleiter Date: Wed Jan 20 17:13:28 2010 +0000 lcov: unify data order in tracefiles commit 8f53b2e8dbbe5580050fbe0c604bd9a9322735a7 Author: Peter Oberparleiter Date: Wed Jan 20 16:05:56 2010 +0000 lcov: fix bug when applying baseline files without function coverage data Fix the following error that occurs when genthml's --baseline-file option is used on files which do not contain any function data: genhtml: Can't use an undefined value as a HASH reference at ./lcov/bin/genhtml line 4441. commit 96fcd676d5ac9c1eb9f83f3dc4c3089ba478baad Author: Peter Oberparleiter Date: Wed Jan 20 15:28:21 2010 +0000 lcov: resolve short-name option ambiguities commit f1d34d49b394a13c33c7a5b51f04e5dfbded5d26 Author: Peter Oberparleiter Date: Wed Jan 20 14:47:50 2010 +0000 lcov: fix error messages commit 89ff61aa7cd2ca23b8cacd649288ecf7f67746de Author: Peter Oberparleiter Date: Wed Jan 20 08:35:20 2010 +0000 lcov: fix bug when using genhtml's --baseline-file option Fix the following error message when trying to use genhtml's --baseline-file option: genhtml: Undefined subroutine &main::add_fnccounts called at /home/oberpar/bin/genhtml line 4560. Reported by Brian DeGeeter commit c3df3a8504b06ca32b9863fdb2abb8cf0ce62251 Author: Peter Oberparleiter Date: Mon Jan 18 09:17:17 2010 +0000 lcov: ensure LANG=C before calling gcov Fix problem calling lcov when LANG is not set to an english locale. Reported by benoit_belbezet@yahoo.fr. commit d945f23345e02ca535d740782e7ae10cb3396b8c Author: Peter Oberparleiter Date: Wed Nov 18 09:39:21 2009 +0000 lcov: more version fixup commit 413249e6336cff432083954e6ed47236dd35f647 Author: Peter Oberparleiter Date: Wed Nov 18 09:38:03 2009 +0000 lcov: fix version fixup commit d0b7148e2d76164e5ea091fe56035c24f7dce22a Author: Peter Oberparleiter Date: Wed Nov 18 09:34:45 2009 +0000 lcov: add more CVS versioning commit 4e0219f918a15cbc9ff40d0e0e4dab91ac073f72 Author: Peter Oberparleiter Date: Wed Nov 18 09:14:56 2009 +0000 lcov: add CVS revision number to version output commit 34154c2d48497d9aad41ec1452ba94dd4cbce881 Author: Peter Oberparleiter Date: Fri Oct 30 14:18:45 2009 +0000 lcov: further clarification in the README commit 7a4ab1340dd7f88ba0fb56a7b0eb368bf2d0112e Author: Peter Oberparleiter Date: Fri Oct 30 13:58:56 2009 +0000 lcov: update README to mention required -lgcov switch during linking commit 3fa5b311b123af84debbd774baa4a1cd30e7085b Author: Peter Oberparleiter Date: Tue Oct 27 16:54:41 2009 +0000 lcov: remove further unneeded warning ... + use correct source for list of filenames commit cd4051719e72129f4abf1ad177269bf14031f83a Author: Peter Oberparleiter Date: Tue Oct 27 16:19:05 2009 +0000 lcov: fix problem with matching filename - used correct source for filenames - converted match_filenames to portable version commit 0d0ff8a9945260eebed6d316aa08c0021faf3549 Author: Peter Oberparleiter Date: Tue Oct 27 15:29:41 2009 +0000 lcov: remove unnecessary warning commit 6c711d664c38d18f788ee8a5239586cd4a5b77d9 Author: Peter Oberparleiter Date: Mon Oct 26 14:21:40 2009 +0000 lcov: improve derive-func-data option - rewrite graph file handling - make derive data look at all lines belonging to a function to find out whether it has been hit or not - introduce --debug option to better debug problems with graph files commit 214cda20c4b591a823045f35b73f2a16221c9aa1 Author: Peter Oberparleiter Date: Thu Oct 1 15:26:58 2009 +0000 lcov: introduce new options --derive-func-data When using a gcov version that does not provide function data, this option will attempt to guess the function coverage data for a function by looking at the number of times that the first line of that function was called. commit 9a75125895fd07a775a2a25f2cbe66b9fbf332d6 Author: Peter Oberparleiter Date: Thu Oct 1 11:49:53 2009 +0000 lcov: ignore incomplete function names in .bb files - don't abort processing when an incomplete function name is encountered in a .bb file (gcc 2.95.3 adds those) - fix filename prefix detection commit d5ab6076a0bfc5ad80652ba592583f7fc7946dc6 Author: Peter Oberparleiter Date: Mon Sep 28 12:27:09 2009 +0000 lcov: improve detection of gcov-kernel support commit 3cca782fcac9c4ea54adcebe75e1f047a8dca636 Author: Peter Oberparleiter Date: Tue Sep 22 13:44:04 2009 +0000 lcov: fix problem with CONFIG_MODVERSIONS Make geninfo work with Linux 2.6.31 and CONFIG_MODVERSIONS. commit 8af873f44c104cd214b796e13b916718fc8f6f99 Author: Peter Oberparleiter Date: Wed Sep 16 15:24:51 2009 +0000 lcov: remove default for gcov_dir so that auto-sensing works Fix problem with lcov not finding kernel coverage data at /sys/kernel/debug/gcov because the default system-wide lcovrc file contained a specification for the gcov directory which prevented auto-detection from working. commit 50f90681af4d105a52b5b0dbf4f0bfd04369ffd2 Author: Peter Oberparleiter Date: Thu Aug 27 10:14:23 2009 +0000 lcov: apply excluded lines also to function coverage data commit 4aeb840d25c85a419171970e1a445aeb81079e53 Author: Peter Oberparleiter Date: Thu Aug 27 09:23:13 2009 +0000 lcov: fix help text typo commit c17a783f87aa8e42949131d2fbc1c540bb3751a3 Author: Peter Oberparleiter Date: Thu Aug 27 09:22:43 2009 +0000 lcov: add exclusion markers Users can exclude lines of code from coverage reports by adding keywords to the source code. commit 445715c88337c13ce496bd05423ee5e58d84705c Author: Peter Oberparleiter Date: Fri Aug 14 08:19:26 2009 +0000 lcov: ignore gcov errors for unnamed source files When specifying "--ignore-errors gcov", lcov/geninfo should not abort when they cannot read a .gcov file. Fix this by introducing warnings in the respective places. commit 0e23f03a9ce130e8ebec679fb5a9a6f854efbee5 Author: Peter Oberparleiter Date: Thu Aug 6 12:34:04 2009 +0000 lcov: improvements - added --from-package and --to-package options - improved gcov-kernel handling commit 17a05bdf646870cd61794274c7165211c93c82f9 Author: Peter Oberparleiter Date: Thu Jul 23 12:45:15 2009 +0000 lcov: fix kernel capture for new gcov-kernel version - fix problems when compiling without O= commit 64e302b9134b6098852cad2e6180e0722f2dea41 Author: Peter Oberparleiter Date: Tue Jul 21 15:42:44 2009 +0000 lcov: improve lcov -l output commit cea6941ef36d0860330b6e94f8c6096dca78ca58 Author: Peter Oberparleiter Date: Tue Jul 21 09:10:49 2009 +0000 lcov: add support for the linux-2.6.31 upstream gcov kernel support commit 04470d2b25808f195d338112155b9f7db405d902 Author: Peter Oberparleiter Date: Wed Apr 22 09:13:12 2009 +0000 genhtml: fix warning about undefined value used nikita@zhuk.fi: genhtml.patch checks that $funcdata->{$func} is defined before using it - I got few "undefined value used" warnings without this check. commit a12d4f9a5d36232b928be12b7cbfaa9a00b3a923 Author: Peter Oberparleiter Date: Wed Apr 22 09:08:19 2009 +0000 genpng: fix runtime-warning - when called from within genhtml, genpng would warn about warn_handler being redefined commit d0b5641c62bbdac89757b9ff185a7aa3f38fc0bb Author: Peter Oberparleiter Date: Fri Mar 13 09:58:00 2009 +0000 lcov: improve function name filtering Only remove those characters from function names which would conflict with internal delimiters. commit fbafa4a5628a639544e83f88083082c685677c36 Author: Peter Oberparleiter Date: Fri Feb 13 15:04:40 2009 +0000 genhtml: minor man page update commit 085a2150e38a3c1bdadb5af23c0a8a8a79dc4b0d Author: Peter Oberparleiter Date: Fri Feb 13 14:56:45 2009 +0000 genhtml: added --demangle-cpp option - used to convert C++ internal function names to human readable format - based on a patch by slava.semushin@gmail.com commit 53f3ed4afb45a2a4248314b677d36377598cc73c Author: Peter Oberparleiter Date: Fri Feb 13 14:07:46 2009 +0000 genhtml: update comment commit 3c2b2e8541387506fd514d183f9a4a63c07c0aa4 Author: Peter Oberparleiter Date: Thu Feb 12 17:01:19 2009 +0000 genhtml: fix error when combining tracefiles without function data - genhtml: Can't use an undefined value as a HASH reference at genhtml line 1506. - bug reported by richard.corden@gmail.com commit 22397370ada6893b6e9a1c3f6ad0aba7f4864f81 Author: Peter Oberparleiter Date: Wed Feb 11 09:31:24 2009 +0000 lcov: fix error when combining tracefiles without function data - lcov: Can't use an undefined value as a HASH reference at lcov line 1341. - bug reported by richard.corden@gmail.com commit 24ec53ae83acdd35682ba757adae23750bd4c623 Author: Peter Oberparleiter Date: Mon Feb 9 16:15:49 2009 +0000 lcov: fix warning when $HOME is not set - based on patch by acalando@free.fr commit 5da3521d5a438db0a21e93b0d14ea5a3cdab14d9 Author: Peter Oberparleiter Date: Mon Feb 9 12:41:44 2009 +0000 lcov: use install -pD -m for file installation commit bdce1bda2ac1a86aa6dfefae8e18353ba57afe4b Author: Peter Oberparleiter Date: Mon Feb 9 09:46:00 2009 +0000 lcov: fix double-counting of function data commit ea62c4e701abb05dd560ef22b52a4a72c17660e8 Author: Peter Oberparleiter Date: Wed Jan 21 16:33:29 2009 +0000 geninfo: need to add CR removal to geninfo as well ... or checksumming will fail commit 70be5df7d58a393e27cee178df669c12ec670c5a Author: Peter Oberparleiter Date: Wed Jan 21 16:24:01 2009 +0000 lcov: modify end-of-line CR removal - s///g is 10% slower than s/// - \r may be 0x10 or 0x13 (see man perlport) commit d8df4b0f83ff175f1a06afb693903ee1a93ec377 Author: Michael Knigge Date: Tue Jan 20 11:41:39 2009 +0000 lcov: remove CRLF line breaks in source code when generating html output - added patch by michael.knigge@set-software.de commit 442cca7e69356e7f8ba03bd95f7813576bd197cc Author: Peter Oberparleiter Date: Mon Nov 17 14:11:20 2008 +0000 lcov: updated CVS version to 1.8 commit 5c5c85a1c090360facd50cb089b8af98f0b37c47 Author: Peter Oberparleiter Date: Mon Nov 17 13:55:52 2008 +0000 lcov: version + date updates commit 9f6a735809c23559b861e97a20af55a66b6b96bb Author: Peter Oberparleiter Date: Mon Nov 17 13:49:43 2008 +0000 lcov: fix spec file bug commit 11483dc0b56d326718edcd31d06458143add858f Author: Peter Oberparleiter Date: Mon Nov 17 13:44:38 2008 +0000 lcov: update error and warning messages commit 4dd11b80d14e34fee2e75b3fe8c7aa163f61ad1d Author: Peter Oberparleiter Date: Mon Nov 17 12:48:03 2008 +0000 lcov: preparations for release 1.7 commit b847ed6f3103a4c9f0a48417b9c3f160b9e00557 Author: Jeff Connelly Date: Fri Oct 10 07:54:47 2008 +0000 lcov: geninfo chokes on spaces in the directory name In lcov 1.6, geninfo fails to find gcno/gcda files if the source directory has spaces in the name, because it uses backticks to shell out to "find", passing $directory on the command-line. Attached is a patch that double-quotes the variable, allowing geninfo to operate on directories with spaces in their name. The fix isn't perfect; it won't work on directories with a " character, but it works fine for my purposes (I don't have any directories with quotes). A better fix would be to use IPC::System::Simple's capturex from http://search.cpan.org/~pjf/IPC-System-Simple-0.15/lib/IPC/System/Simple.pm #runx(),_systemx()_and_capturex(). capturex() is a multiple-argument form of the backticks, so it avoids any interpolation errors. commit ee3cdd554ee4e6d3ef5bdc9c5dcfee50de6375a7 Author: Peter Oberparleiter Date: Mon Aug 18 07:12:33 2008 +0000 lcov: change sorting order to low-to-high coverage commit fe665ca5ccf9d73d9ebdae17de88e181c1b9b0eb Author: Peter Oberparleiter Date: Fri Aug 15 08:38:21 2008 +0000 lcov: several changes - update download link - unify webpage links - provide --sort and --function-coverage switch + documentation commit 14137c5456f307982fed418e1e8fac65d7f086c3 Author: Peter Oberparleiter Date: Wed Aug 13 15:57:23 2008 +0000 lcov: fix function view page creation when --no-func is specified commit e59f7d15ffc7f1b3794a4212c53d0fb97ac7fb2a Author: Peter Oberparleiter Date: Wed Aug 13 15:35:48 2008 +0000 lcov: updated versioning mechanism ... + fixed some man page bugs commit e933698b31bc2fb4a750d89a5755bb8155313da2 Author: Peter Oberparleiter Date: Wed Aug 13 14:08:23 2008 +0000 lcov: updated rpm description ... + summary and version strings commit 5a9660585ce39a77fa38607d0c2d2440955e7242 Author: Peter Oberparleiter Date: Wed Aug 13 13:53:50 2008 +0000 lcov: integrated function coverage patch ... by Tom Zoernen + sorting function commit d10ede8179747cfd675a3989578350c710e9bdd5 Author: Peter Oberparleiter Date: Wed May 7 15:08:12 2008 +0000 lcov: --norecursion becomes --no-recursion + added docs commit 4096130608b9faf74c5b5feac554a10b5d9f83ce Author: Peter Oberparleiter Date: Thu Feb 21 10:20:33 2008 +0000 lcov: fix error when trying to use genhtml -b genhtml fails when the data file contains an entry which is not found in the base file. commit 9578099e13388344a6179c7cce54bfa094fd9b08 Author: Peter Oberparleiter Date: Wed Feb 20 17:21:51 2008 +0000 lcov: fixed problem with pre gcc-3.3 versions read_gcov_headers does not return valid results for pre gcc-3.3 versions. Due to an unnecessary check, parsing of gcov files was aborted. Fix by removing check. commit 16ec76b48fbc50c32890919e5bd0c30653719af9 Author: Peter Oberparleiter Date: Tue Feb 5 09:18:50 2008 +0000 lcov: adding support for gzipped html ... based on patch by dnozay@vmware.com dnozay@vmware.com: genhtml is a great tool to generate html, but the more files, the more space it takes (here I have over 113MB of html generated), add to that I need to have different sets, and space usage increases dramatically (2.7GB). we are using browsers with htmlz support, so it would be nice to have support for that in genhtml, relying on 'gzip -S z' to do the job. commit f2c98a8c8581180533508eb4af41720d8566049e Author: Peter Oberparleiter Date: Mon Jan 7 16:33:57 2008 +0000 Filter non-word characters in function name ... as they would break our file format which uses comma and '=' as field separator. commit 37725fc78fcacaf06e6240971edc3bdd7fe3d142 Author: Peter Oberparleiter Date: Thu Nov 1 16:29:39 2007 +0000 lcov: fix for problem resulting in lcov aborting with "ERROR: reading string" commit 48f13fcec1b521d2daba6202ccd7ec0ec8c5ece9 Author: Peter Oberparleiter Date: Thu Oct 4 08:18:07 2007 +0000 lcov: workaround for gcc 4.1.0 .gcno file oddness scott.heavner@philips.com: I'm trying to use lcov 1.6 with gcov/gcc 4.1.0. The geninfo parser was aborting on a small number of .gcno files. I've patched my local copy so that geninfo prints out the offset of the error and skips the remainder of the problem file commit 1a805ea068db29b63a83c801f3bb1840fda8dd35 Author: Peter Oberparleiter Date: Fri Aug 24 08:50:26 2007 +0000 lcov: add experimental option "--norecursion" commit 194de5071db1d9903d22164432448b73c1ec6cd0 Author: Peter Oberparleiter Date: Thu Aug 23 11:08:39 2007 +0000 lcov: Makefile for post-release commit 0750f8a3e5235833711d616a3763c04103cf55a5 Author: Peter Oberparleiter Date: Thu Aug 23 11:04:30 2007 +0000 lcov: Makefile for release 1.6 commit cb911f7a79593c89a730dc93fa54179fbf1df363 Author: Peter Oberparleiter Date: Mon Aug 20 10:29:35 2007 +0000 lcov: fixed spec file commit 62cefebdda87784140eb5f997ae4e575d2338298 Author: Peter Oberparleiter Date: Fri Jul 6 07:38:47 2007 +0000 lcov: add new option --initial to get zero coverage data from graph files commit f0b6927f1ab1052b00081c662ced614a6e5f9ed7 Author: Peter Oberparleiter Date: Wed Jul 4 14:38:59 2007 +0000 lcov: fixed bug that would not delete .gcda files when using -z commit 13941c3a159caf7dc6ba18a5b13e43c20fc18f2b Author: Peter Oberparleiter Date: Wed Jul 4 14:18:26 2007 +0000 lcov: another update in preparation for a new release commit d25e630a77ef2d0f69139058322269387866e414 Author: Peter Oberparleiter Date: Wed Jul 4 13:13:22 2007 +0000 lcov: man page update commit 7844b915af5402441df9ab0423e4c20ef9a2632f Author: Peter Oberparleiter Date: Tue Jul 3 16:43:05 2007 +0000 lcov: update manpage commit 5adaa72bfb32737d18c328492777c1c6116d4a9e Author: Peter Oberparleiter Date: Mon Jul 2 15:29:02 2007 +0000 lcov: preparations for new release - updated CHANGES file - added compat-libtool + no-compat-libtool option - changed libtool default to on (due to popular request) - added checksum option - changed checksum default to off (to reduce cpu time + file size) - added geninfo_checksum option to lcovrc, deprecated geninfo_no_checksum - added geninfo_compat_libtool option to lcovrc - minor update of README file commit 6cbfd5022703a6198e1a1e2a2ddddcc0b90f5334 Author: Peter Oberparleiter Date: Tue May 22 08:11:44 2007 +0000 lcov: minor help text update commit 2416ed02ba299c4d0bceb1e47c214b7dec066d7a Author: Peter Oberparleiter Date: Wed Mar 7 14:59:25 2007 +0000 lcov - add --ignore-errors option to lcov/geninfo - add --gcov-tool option to lcov/geninfo - remove s390 test case modification in geninfo - restructured help text for lcov/geninfo commit a13375811717d3ada718e6f52364e4344a7e3187 Author: Peter Oberparleiter Date: Mon Jan 8 17:07:21 2007 +0000 lcov - re-added libtool compatibility workaround patch by thomas@apestaart.org - added new lcov/geninfo-option --compat_libtool to activate libtool compatibility patch commit 14871d7b097282819db60266d8b8a38506d7b14a Author: Peter Oberparleiter Date: Tue Nov 14 11:45:17 2006 +0000 lcov Fix for problem found by Joerg Hohwieler: lcov -k doesn't work if -k is specified more than once. commit 43b52b37006822c0fca12548bc72fecc957342ca Author: Peter Oberparleiter Date: Mon Jun 26 15:48:52 2006 +0000 lcov: new version for prerelease rpms commit 89e9d59709c9d9d8722170c86251090adc3b96c9 Author: Peter Oberparleiter Date: Wed Jun 7 09:31:57 2006 +0000 lcov: removed autoupdate of copyright date (second thoughts) commit bb0cf1c9d0ed58b37c1551fea765fb1622bcacde Author: Peter Oberparleiter Date: Wed Jun 7 09:20:37 2006 +0000 lcov: minor cleanup (release preparations) commit 527693d753d11ac2b59fe26b923662c99e6e3715 Author: Peter Oberparleiter Date: Wed Apr 5 10:10:05 2006 +0000 lcov - added base-directory documentation - updated CHANGES file commit 11ef9338cc4124801c8b61e3edd51a02e50b4c68 Author: Peter Oberparleiter Date: Mon Mar 20 17:09:50 2006 +0000 genhtml: added html-extension option commit 93d22308ffb410327248059b7dcdb592f85e249e Author: Peter Oberparleiter Date: Mon Mar 20 16:39:25 2006 +0000 genhtml - adding html-prolog and html-epilog options (based on patch by Marcus Boerger) - specified behavior when both --no-prefix and --prefix options where provided - small whitespace diff commit dcac095cdc00cc65930285bb6fc01d0f257ee4ed Author: Peter Oberparleiter Date: Wed Feb 15 16:02:07 2006 +0000 lcov: added check for invalid characters in test names commit d89e561dfd9c5fde43350af1b145b1892d0710d0 Author: Peter Oberparleiter Date: Fri Dec 2 06:38:16 2005 +0000 lcov - updated Makefile so that people building RPMs from the CVS version get a correct build version. Note: this needs to be adjusted after each release! commit 1960123050f9098690768d10cd2490dd49b995f7 Author: Peter Oberparleiter Date: Thu Nov 10 13:10:09 2005 +0000 lcov - fixed bug: .info file generation with new gcc 4.x compilers may fail for programming languages that allow ':' in function names (c++, objective c) - removed special handling for libtool .libs files - libtool should work with currently undocumented option --base-directory commit 479d446d3bf20a84c2933100ead279c79eeaf5c4 Author: Peter Oberparleiter Date: Wed Sep 7 16:24:39 2005 +0000 lcov - implementation of new option --base-directory (untested, undocumented) - minor fix for link-traversal when looking for object directory TODO: document option (man page, online help), add to config file, check whether libtool fix still works commit 770b94a3172f206de7f194c7497ebae14348b521 Author: Robert Williamson Date: Mon Jul 11 17:54:25 2005 +0000 Applied patch from Stefan Kost when running lcov over an uninstalled user-space apps tests, it finds the .da file in the .libs directories, but does not look for the sources one hierarchy up. Libtool places the object in the .libs dirs. when running gcov manually one can specify -o.libs/ to produce a source.c.gov file. I now have attached a patch that fixes the problem for me. please do not just ignore this report. the lcov tool is so nice and it would be a shame if it can not be used for normal apps. commit 79f2ff2c168150e7532046c2cdbc1e42c8b4708f Author: Peter Oberparleiter Date: Tue Jun 14 11:34:59 2005 +0000 lcov - renamed support for modified compilers (gcc 3.3 hammer patch) - fixed bugs in the support for modified compilers commit fb7dab3494fdd8b093e6a84f088f6ea07fcefe6e Author: Peter Oberparleiter Date: Tue Mar 15 18:02:54 2005 +0000 lcov Emil.Jansson@oss.teleca.se: lcov 1.4 does not work with the gcc version in Mandrake Linux 10.0 >> gcc --version gcc (GCC) 3.3.2 (Mandrake Linux 10.0 3.3.2-6mdk) This patch for geninfo fixes the problem: commit ae3fe899d824e8af8a16736a0c8104c903565a56 Author: Peter Oberparleiter Date: Tue Mar 8 14:23:06 2005 +0000 lcov - added optional legend to HTML output - changed background color for "good coverage" entries to green for consistency reasons commit 18b73d39fd9d6bc8829395baa612a6ed98b89efe Author: Peter Oberparleiter Date: Wed Mar 2 14:49:47 2005 +0000 lcov - fixed rpm build process to exclude unnecessary directories in RPM commit ef6ee74df5bf1d1d104322f8fff36b5c6fda34b4 Author: Peter Oberparleiter Date: Wed Mar 2 12:48:29 2005 +0000 lcov - added man page for configuration file lcovrc commit be3afe2626d6bc72256e1873d409c737ac4391c9 Author: Peter Oberparleiter Date: Mon Feb 28 16:31:51 2005 +0000 lcov - Updated CHANGES file in preparation for a new release commit dc68ce9c804ef21bc8e149d9b468e18c1619bb54 Author: Peter Oberparleiter Date: Tue Nov 2 15:48:45 2004 +0000 lcov - temporary fix for a problem which occurs when trying to parse C++ coverage data generated with vanilla gcc 3.3.3 commit efedc5b930ab6743ea9f47ce4ea4a1a75bd739ff Author: Peter Oberparleiter Date: Mon Sep 27 13:13:51 2004 +0000 lcov - fix for minor bug in geninfo (access to uninitialized variable) related to SLES9 compatibility test and test for existing source code files commit 47943eedfbec7a12c52e7a8ccbcfaf8d0706f142 Author: Peter Oberparleiter Date: Mon Sep 20 14:11:16 2004 +0000 lcov - minor fix for regular expression used to parse .gcov files - caused problems when parsing branch coverage data and when using custom gcov versions commit ce6335ebd92ce017b75ee3e194e9e3ca7bc7e1f3 Author: Peter Oberparleiter Date: Tue Sep 14 15:52:38 2004 +0000 lcov - fixed bug in geninfo which would not report any FN: data for data generated with gcc versions 3.4.0 and above commit 58df8af3a62fa4e60569ef300e0ddd0073bf109e Author: Peter Oberparleiter Date: Tue Aug 31 15:57:41 2004 +0000 lcov - added support for modified GCC version provided by SUSE SLES9 commit 69f3bc3a0c59b35eb6882205286a68b04a8a8d22 Author: Peter Oberparleiter Date: Tue Aug 31 15:48:32 2004 +0000 lcov - fixed bug in lcov RPM spec file which would not include the global config file in the package list commit 5d10ca22144ad2be885405c3683b20c0976f7562 Author: Peter Oberparleiter Date: Mon Aug 9 14:32:23 2004 +0000 lcov - fixed a bug which would cause generation of incorrect line checksums when source code is not available while capturing coverage data - changed default directory for temporary files from . to /tmp commit 8ee3061f23f17a5074deda0777c66c3e82b5d852 Author: Peter Oberparleiter Date: Mon Aug 9 11:15:02 2004 +0000 lcov - added configuration file support - fixed Makefile error for target "uninstall" commit 58af07f0b0ca1af8c9f2b90ad1683447bb560165 Author: Peter Oberparleiter Date: Fri Aug 6 11:36:33 2004 +0000 lcov - fixed bug which would cause an error when lcov was used on a source directory which contained perl regular expression special characters - simplified regular expression character escaping - removed unnecessary function escape_shell from lcov commit 69a6918d4cd386aff2fbff093a6e0b5ddcc46602 Author: Peter Oberparleiter Date: Tue Mar 30 13:27:55 2004 +0000 lcov: - added --path option to fix --diff functionality commit cbc6cb11b532e525ae8b0c0742a4fd41189ca7c2 Author: Peter Oberparleiter Date: Mon Mar 29 12:56:08 2004 +0000 lcov - Added compatibility for gcc-3.4 - Modified --diff function to better cope with ambiguous entries in patch files - Modified --capture option to use modprobe before insmod (needed for 2.6) commit 1cf9a02c3ea0e98cc1d8b626eaa0a2a1cbd96cf1 Author: Peter Oberparleiter Date: Fri Jan 30 09:42:13 2004 +0000 lcov - updated CHANGES file - changed Makefile install path (/usr/local/bin -> /usr/bin) commit c60f0668059032cf4dc5f6c556fd6117925f535f Author: Peter Oberparleiter Date: Wed Jan 14 10:14:10 2004 +0000 lcov-patch by Laurent Deniel avoids aborting the geninfo processing when an empty .bb file is encountered (e.g. source code with no profiled statement) commit 7f2966f8f874a6c905b4d31e5aaf0f4654929044 Author: Peter Oberparleiter Date: Fri Dec 19 16:22:52 2003 +0000 lcov: updated references to lcov webpage to reflect recent site changes commit a3893f4eb2b4fadc4d7350324d74fa453a5ba0f3 Author: Peter Oberparleiter Date: Fri Dec 19 12:50:28 2003 +0000 Added changes by Laurent Deniel - a small patch to lcov 1.1 that introduces the --follow option (in lcov & geninfo) to control whether or not links should be followed while searching for .da files. - a workaround for a gcov (3.2) bug which aborts with empty .da files (gcov 3.3 is fixed but many distributions include gcc 3.2) commit d44f2f8e8672e31cc104c0598b0556a5949dc067 Author: Paul Larson Date: Fri Nov 21 19:34:59 2003 +0000 Fixed two buglets that caused geninfo to break with some versions of gcov. 1. Return value for gcov --help might not be 0, expect -1 when it doesn't exist 2. use -b instead of expanded (--branch-coverage or whatever it was) commit 5a1a33a840a665c77409f799be91cc2dce5cd3b2 Author: Peter Oberparleiter Date: Tue Nov 18 14:06:47 2003 +0000 lcov - fixed function which interprets branch possibility data in geninfo (branch x taken = y% would not be interpreted correctly) - deactivated function which would add 'uname -a' output to testname in geninfo (output in genhtml/showdetails looked unreadable, there needs to be some better solution) commit e0ea03fedf43a3232c35708f882d7058998b2b3d Author: Peter Oberparleiter Date: Fri Oct 10 14:18:32 2003 +0000 New function and bug fix update. Makefile: - Added rule to build source rpm lcov.spec: - Modified to support building source rpms genhtml: - Fixed bug which would not correctly associate data sets with an empty test name (only necessary when using --show-details in genhtml) - Added checksumming mechanism: each tracefile now contains a checksum for each instrumented line to detect incompatible data - Implemented new command line option '--nochecksum' to suppress generation of checksums - Implemented new command line option '--highlight' which highlights lines of code which were only covered in converted tracefiles (see '--diff' option of lcov) geninfo: - Added checksumming mechanism: each tracefile now contains a checksum for each instrumented line to detect incompatible data - Implemented new command line option '--nochecksum' to suppress generation of checksums - Added function to collect branch coverage data lcov: - Fixed bug which would not correctly associate data sets with an empty test name (only necessary when using --show-details in genhtml) - Cleaned up internal command line option check - Added info() output when reading tracefiles - Added checksumming mechanism: each tracefile now contains a checksum for each instrumented line to detect incompatible data - Implemented new command line option '--nochecksum' to suppress generation of checksums - Implemented new command line option '--diff' which allows converting coverage data from an older source code version by using a diff file to map line numbers genpng: - Added support for the highlighting option of genhtml - Corrected tab to spaces conversion commit c17af02b4a856d8733a763e6c0685c31f3c7fb74 Author: Nigel Hinds Date: Fri Sep 19 21:51:06 2003 +0000 capture branch coverage data from GCOV. commit e2fc88f85254017bcf1fb04a3c935395a9b7a4a1 Author: James M Kenefick Jr Date: Thu Sep 4 16:56:10 2003 +0000 Initial checking of the galaxy map commit dfec606f3b30e1ac0f4114cfb98b29f91e9edb21 Author: Peter Oberparleiter Date: Sat Jul 5 13:48:45 2003 +0000 LCOV: Fixed negative count handling - Negative counts are treated as zero - Warning is issued when encountering negative counts commit a2ee105a07b19c52efe7a3e6e5b11a27b4b60ef8 Author: Paul Larson Date: Wed Jul 2 19:37:50 2003 +0000 Small fixes before the release commit 72860625dd904f84909253b20a7fc024b4e3377e Author: Peter Oberparleiter Date: Mon May 5 08:32:04 2003 +0000 Adjusted example program and README file ... to reflect renaming of lcov option '--reset' to '--zerocounters'. commit cbd9e315832960604d2949439326b30f4061e512 Author: Peter Oberparleiter Date: Wed Apr 30 15:47:51 2003 +0000 Renamed lcov option '--reset' to '--zerocounters' - Included '--remove' in help text of lcov - Adjusted man pages to include option changes - Extended info() change to geninfo and genhtml (infos are now printed to STDERR) commit 8155960cb5db0359470d2a5f652bdc744e9ecfcd Author: Peter Oberparleiter Date: Wed Apr 16 15:43:31 2003 +0000 Modified read_gcov so that it can also parse the new gcov format which is to be introduced in gcc 3.3. commit 382440f781b12ade8f1f7962a0eb1cfc0525f2a5 Author: Paul Larson Date: Tue Apr 15 16:06:59 2003 +0000 Added --remove option info() now prints to stderr commit 62760fa1840326e849c7e58892ce671f510bb0af Author: Peter Oberparleiter Date: Mon Apr 14 09:31:51 2003 +0000 Check-in of updated LCOV version (to be released as 1.1). Includes fixes and modifications by Mike Kobler, Paul Larson and myself. A quote from the CHANGS file: - Added CHANGES file - Added Makefile implementing the following targets: * install : install LCOV scripts and man pages * uninstall : revert previous installation * dist : create lcov.tar.gz file and lcov.rpm file * clean : clean up example directory, remove .tar and .rpm files - Added man pages for all scripts - Added example program to demonstrate the use of LCOV with a userspace application - Implemented RPM build process - New directory structure: * bin : contains all executables * example : contains a userspace example for LCOV * man : contains man pages * rpm : contains files required for the RPM build process - LCOV-scripts are now in bin/ - Removed .pl-extension from LCOV-script files - Renamed readme.txt to README README: - Adjusted mailing list address to ltp-coverage@lists.sourceforge.net - Fixed incorrect parameter '--output-filename' in example LCOV call - Removed tool descriptions and turned them into man pages - Installation instructions now refer to RPM and tarball descriptions.tests: - Fixed some spelling errors genhtml: - Fixed bug which resulted in an error when trying to combine .info files containing data without a test name - Fixed bug which would not correctly handle data files in directories with names containing some special characters ('+', etc.) - Added check for empty tracefiles to prevent division-by-zeros - Implemented new command line option --num-spaces / the number of spaces which replace a tab in source code view is now user defined - Fixed tab expansion so that in source code view, a tab doesn't produce a fixed number of spaces, but as many spaces as are needed to advance to the next tab position - Output directory is now created if it doesn't exist - Renamed "overview page" to "directory view page" - HTML output pages are now titled "LCOV" instead of "GCOV" geninfo: - Fixed bug which would not allow .info files to be generated in directories with names containing some special characters lcov: - Fixed bug which would cause lcov to fail when the tool is installed in a path with a name containing some special characters - Implemented new command line option '--add-tracefile' which allows the combination of data from several tracefiles - Implemented new command line option '--list' which lists the contents of a tracefile - Implemented new command line option '--extract' which allows extracting data for a particular set of files from a tracefile - Fixed name of gcov kernel module (new package contains gcov-prof.c) - Changed name of gcov kernel directory from /proc/gcov to a global constant so that it may be changed easily when required in future versions commit ec94ed71838a9780e82ea8bd67742bde2f4eeb47 Author: Paul Larson Date: Fri Mar 7 20:28:15 2003 +0000 Fix lcov.pl to work with the new gcov-kernel module ... ,documentation fixes in readme.txt commit e70d9abdb60b83de7174815371259c63fa75bf76 Author: Robert Williamson Date: Tue Feb 18 20:05:09 2003 +0000 Applied patch from Mike Kobler: One of my source file paths includes a "+" in the directory name. I found that genhtml.pl died when it encountered it. I was able to fix the problem by modifying the string with the escape character before parsing it. commit 69ef6f1b607670589aae1ae1e6c78ef1b5d204e3 Author: Peter Oberparleiter Date: Fri Sep 6 09:04:34 2002 +0000 Replaced reference to "cat" cvs directory ... and to .zip package. commit c641f6e694e2bebf9ef0a507091460026463d169 Author: Manoj Iyer Date: Thu Sep 5 19:14:51 2002 +0000 Coverage analysis files. Peter worked on this version. lcov-1.14/.version0000644000000000000000000000004113436014710012560 0ustar rootrootVERSION=1.14 RELEASE=1 FULL=1.14 lcov-1.14/README0000644000000000000000000001073713436002743011772 0ustar rootroot------------------------------------------------- - README file for the LTP GCOV extension (LCOV) - - Last changes: 2019-02-28 - ------------------------------------------------- Description ----------- LCOV is an extension of GCOV, a GNU tool which provides information about what parts of a program are actually executed (i.e. "covered") while running a particular test case. The extension consists of a set of Perl scripts which build on the textual GCOV output to implement the following enhanced functionality: * HTML based output: coverage rates are additionally indicated using bar graphs and specific colors. * Support for large projects: overview pages allow quick browsing of coverage data by providing three levels of detail: directory view, file view and source code view. LCOV was initially designed to support Linux kernel coverage measurements, but works as well for coverage measurements on standard user space applications. Further README contents ----------------------- 1. Included files 2. Installing LCOV 3. An example of how to access kernel coverage data 4. An example of how to access coverage data for a user space program 5. Questions and Comments 1. Important files ------------------ README - This README file CHANGES - List of changes between releases bin/lcov - Tool for capturing LCOV coverage data bin/genhtml - Tool for creating HTML output from LCOV data bin/gendesc - Tool for creating description files as used by genhtml bin/geninfo - Internal tool (creates LCOV data files) bin/genpng - Internal tool (creates png overviews of source files) bin/install.sh - Internal tool (takes care of un-/installing) man - Directory containing man pages for included tools example - Directory containing an example to demonstrate LCOV lcovrc - LCOV configuration file Makefile - Makefile providing 'install' and 'uninstall' targets 2. Installing LCOV ------------------ The LCOV package is available as either RPM or tarball from: http://ltp.sourceforge.net/coverage/lcov.php To install the tarball, unpack it to a directory and run: make install Use Git for the most recent (but possibly unstable) version: git clone https://github.com/linux-test-project/lcov.git Change to the resulting lcov directory and type: make install 3. An example of how to access kernel coverage data --------------------------------------------------- Requirements: get and install the gcov-kernel package from http://sourceforge.net/projects/ltp Copy the resulting gcov kernel module file to either the system wide modules directory or the same directory as the Perl scripts. As root, do the following: a) Resetting counters lcov --zerocounters b) Capturing the current coverage state to a file lcov --capture --output-file kernel.info c) Getting HTML output genhtml kernel.info Point the web browser of your choice to the resulting index.html file. 4. An example of how to access coverage data for a user space program --------------------------------------------------------------------- Requirements: compile the program in question using GCC with the options -fprofile-arcs and -ftest-coverage. During linking, make sure to specify -lgcov or -coverage. Assuming the compile directory is called "appdir", do the following: a) Resetting counters lcov --directory appdir --zerocounters b) Capturing the current coverage state to a file lcov --directory appdir --capture --output-file app.info Note that this step only works after the application has been started and stopped at least once. Otherwise lcov will abort with an error mentioning that there are no data/.gcda files. c) Getting HTML output genhtml app.info Point the web browser of your choice to the resulting index.html file. Please note that independently of where the application is installed or from which directory it is run, the --directory statement needs to point to the directory in which the application was compiled. For further information on the gcc profiling mechanism, please also consult the gcov man page. 5. Questions and comments ------------------------- See the included man pages for more information on how to use the LCOV tools. Please email further questions or comments regarding this tool to the LTP Mailing list at ltp-coverage@lists.sourceforge.net lcov-1.14/test/0000755000000000000000000000000013436014706012063 5ustar rootrootlcov-1.14/test/lcov_misc/0000755000000000000000000000000013436014706014041 5ustar rootrootlcov-1.14/test/lcov_misc/Makefile0000644000000000000000000000014513067205540015477 0ustar rootrootinclude ../common.mak test: @test_run lcov_version lcov --version @test_run lcov_help lcov --help lcov-1.14/test/profiles/0000755000000000000000000000000013436014706013706 5ustar rootrootlcov-1.14/test/profiles/medium0000644000000000000000000000254013067205540015110 0ustar rootroot# Profile of a medium-sized source code project. Use with mkinfo to generate # fake test source code and coverage data. [tests] # List of test names names = test1 test2 test3 [files] # Create this many files numfiles = 50 # Generate paths from these components (top/sub/subsub/prefix_suffix.ext) top = lib tools test bin img scripts sub = build debug release include target sys config subsub = work www utils gui info log basic prefix = main misc report tune mem list suffix = a b c top work proto final fast ext = .c .h [lines] # Generate line coverage data enabled = 1 # Line coverage rate covered = 80 # Percentage of lines covered instrumented = 50 # Maximum number of lines per file maxlines = 1000 [functions] # Generate function coverage data enabled = 1 # Function coverage rate covered = 60 # Percent of instrumented lines containing function definitions perinstrumented = 5 # Generate function names from these components (verb_adj_noun) verb = get set find read write stat add sub combine adj = first last best min max avg noun = bit byte file str num obj data [branches] # Generate branch coverage data enabled = 1 # Branch coverage rate covered = 20 # Percent of instrumented lines containing branches perinstrumented = 50 # List of blocks to use blocks = 0 4294967295 # Distribution of number of branches per block (num:probability) branchdist = 2:50 3:50 lcov-1.14/test/profiles/small0000644000000000000000000000252613067205540014744 0ustar rootroot# Profile of a small source code project. Use with mkinfo to generate fake test # source code and coverage data. [tests] # List of test names names = test1 test2 [files] # Create this many files numfiles = 5 # Generate paths from these components (top/sub/subsub/prefix_suffix.ext) top = lib tools test bin img scripts sub = build debug release include target sys config subsub = work www utils gui info log basic prefix = main misc report tune mem list suffix = a b c top work proto final fast ext = .c .h [lines] # Generate line coverage data enabled = 1 # Line coverage rate covered = 80 # Percentage of lines covered instrumented = 50 # Maximum number of lines per file maxlines = 500 [functions] # Generate function coverage data enabled = 1 # Function coverage rate covered = 60 # Percent of instrumented lines containing function definitions perinstrumented = 5 # Generate function names from these components (verb_adj_noun) verb = get set find read write stat add sub combine adj = first last best min max avg noun = bit byte file str num obj data [branches] # Generate branch coverage data enabled = 1 # Branch coverage rate covered = 20 # Percent of instrumented lines containing branches perinstrumented = 50 # List of blocks to use blocks = 0 4294967295 # Distribution of number of branches per block (num:probability) branchdist = 2:50 3:45 50:5 lcov-1.14/test/profiles/large0000644000000000000000000000253713067205540014730 0ustar rootroot# Profile of a large source code project. Use with mkinfo to generate fake test # source code and coverage data. [tests] # List of test names names = test1 test2 [files] # Create this many files numfiles = 500 # Generate paths from these components (top/sub/subsub/prefix_suffix.ext) top = lib tools test bin img scripts sub = build debug release include target sys config subsub = work www utils gui info log basic prefix = main misc report tune mem list suffix = a b c top work proto final fast ext = .c .h [lines] # Generate line coverage data enabled = 1 # Line coverage rate covered = 80 # Percentage of lines covered instrumented = 80 # Maximum number of lines per file maxlines = 2000 [functions] # Generate function coverage data enabled = 1 # Function coverage rate covered = 60 # Percent of instrumented lines containing function definitions perinstrumented = 10 # Generate function names from these components (verb_adj_noun) verb = get set find read write stat add sub combine adj = first last best min max avg noun = bit byte file str num obj data [branches] # Generate branch coverage data enabled = 1 # Branch coverage rate covered = 20 # Percent of instrumented lines containing branches perinstrumented = 5 # List of blocks to use blocks = 0 4294967295 # Distribution of number of branches per block (num:probability) branchdist = 2:50 3:25 5:20 100:5 lcov-1.14/test/bin/0000755000000000000000000000000013436014706012633 5ustar rootrootlcov-1.14/test/bin/test_run0000755000000000000000000000432513435760116014432 0ustar rootroot#!/usr/bin/env bash # # Copyright IBM Corp. 2017 # # Usage: test_run # # Announce a test case, run it, and record the resulting output in the # test log file. Must be run after testsuite_init. # TOPDIR=$(realpath $(dirname $0)/..) && source "$TOPDIR/bin/common" EXCERPTLEN=10 TESTNAME="$1" shift TIME=$(which time 2>/dev/null) if [ ! -z "$TIME" ] ; then TIME="$TIME -v -o $TIMEFILE" if ! $TIME true 2>/dev/null ; then TIME="" fi fi t_announce "$TESTNAME" let POS=$(stat -c %s "$LOGFILE")+1 t_detail "COMMAND" "\"$*\"" >>"$LOGFILE" t_detail "OUTPUT" "" >>"$LOGFILE" # Run command $TIME bash -c "$*" 2>&1 | t_indent >>"$LOGFILE" RC=$? # Evaluate output of time command ELAPSED= RESIDENT= SIGNAL= if [ ! -z "$TIME" ] ; then while read LINE ; do case "$LINE" in "Command terminated by signal"*) SIGNAL=${LINE##* } ;; "Elapsed"*) ELAPSED=$(elapsed_to_ms ${LINE##* }) ;; "Maximum resident"*) RESIDENT=${LINE##* } ;; "Exit status"*) RC=${LINE##* } ;; esac done < "$TIMEFILE" rm -f "$TIMEFILE" fi t_detail "EXITCODE" "$RC" >>"$LOGFILE" # Show result if [ $RC -eq 0 -a -z "$SIGNAL" ] ; then RESULT="pass" t_pass "$TESTNAME" else if [ -z "$SIGNAL" ] ; then RESULT="fail" t_fail "$TESTNAME" else RESULT="kill" t_kill "$TESTNAME" fi fi if [ ! -z "$SIGNAL" ] ; then t_detail "SIGNAL" "$SIGNAL" >>"$LOGFILE" fi if [ ! -z "$ELAPSED" ] ; then echo -n " (time $(($ELAPSED/1000)).$(($ELAPSED%1000/100))s, " echo "elapsed $TESTNAME $ELAPSED" >> "$COUNTFILE" fi if [ ! -z "$RESIDENT" ] ; then echo -n "mem $(($RESIDENT/1024)).$((($RESIDENT%1024)/100))MB)" echo "resident $TESTNAME $RESIDENT" >> "$COUNTFILE" fi echo # Show log excerpt on failure or if requested if [ $RC -ne 0 -o "$V" == "1" ] ; then LEN=$(tail -c "+$POS" "$LOGFILE" | wc -l) if [ "$LEN" -gt "$EXCERPTLEN" -a "$V" != "1" ] ; then tail -c "+$POS" "$LOGFILE" | head -n $EXCERPTLEN | t_indent let LEN=$LEN-$EXCERPTLEN echo " ..." echo " Skipping $LEN more lines (see $LOGFILE)" else tail -c "+$POS" "$LOGFILE" | t_indent fi fi # Log more details [ ! -z "$ELAPSED" ] && t_detail "TIME" "${ELAPSED}ms" >>"$LOGFILE" [ ! -z "$RESIDENT" ] && t_detail "MEM" "${RESIDENT}kB" >>"$LOGFILE" t_detail "RESULT" "$RESULT" >> "$LOGFILE" lcov-1.14/test/bin/norminfo0000755000000000000000000001271613435760116014421 0ustar rootroot#!/usr/bin/env perl # # Copyright IBM Corp. 2017 # # Usage: norminfo [] # # Normalize coverage data file (ensure stable order), perform some sanity # checks, and apply optional multiplier to execution counts. # use strict; use warnings; sub ferr($$$) { my ($pos, $filename, $msg) = @_; if (defined($pos)) { $pos .= ":"; } else { $pos = ""; } die("$0:$filename:$pos $msg"); } sub print_sorted($$$) { my ($fd, $info, $multi) = @_; my (%fn, %fns, %fnda, %brda, %da); my ($fnf, $fnh, $brf, $brh, $lf, $lh); while (my $line = <$fd>) { $line =~ s/(^\s*|\s*$)//g; if ($line =~ /^end_of_record$/) { last; } elsif ($line =~ /^FN:(\d+),(.*)$/) { my ($lineno, $fnname) = ($1, $2); if (exists($fn{$lineno})) { ferr($., $info, "Duplicate FN: entry\n"); } $fn{$lineno} = $fnname; if (exists($fns{$fnname})) { ferr($., $info, "Duplicate function name\n"); } $fns{$fnname} = $lineno; } elsif ($line =~ /^FNDA:(\d+),(.*)$/) { my ($count, $fnname) = ($1, $2); if (exists($fnda{$fnname})) { ferr($., $info, "Duplicate FNDA: entry\n"); } $fnda{$fnname} = int($count * $multi); } elsif ($line =~ /^FNF:(\d+)$/) { if (defined($fnf)) { ferr($., $info, "Duplicate FNF: entry\n"); } $fnf = $1; } elsif ($line =~ /^FNH:(\d+)$/) { if (defined($fnh)) { ferr($., $info, "Duplicate FNH: entry\n"); } $fnh = $1; } elsif ($line =~ /^BRDA:(\d+),(\d+),(\d+),(\d+|-)$/) { my ($lineno, $block, $branch, $count) = ($1, $2, $3, $4); if (exists($brda{$lineno}->{$block}->{$branch})) { ferr($., $info, "Duplicate BRDA: entry\n"); } $count = int($count * $multi) if ($count ne "-"); $brda{$lineno}->{$block}->{$branch} = $count; } elsif ($line =~ /^BRF:(\d+)$/) { if (defined($brf)) { ferr($., $info, "Duplicate BRF: entry\n"); } $brf = $1; } elsif ($line =~ /^BRH:(\d+)$/) { if (defined($brh)) { ferr($., $info, "Duplicate BRH: entry\n"); } $brh = $1; } elsif ($line =~ /^DA:(\d+),(\d+)$/) { my ($lineno, $count) = ($1, $2); if (exists($da{$lineno})) { ferr($., $info, "Duplicate FNDA: entry\n"); } $da{$lineno} = int($count * $multi); } elsif ($line =~ /^LF:(\d+)$/) { if (defined($lf)) { ferr($., $info, "Duplicate LF: entry\n"); } $lf = $1; } elsif ($line =~ /^LH:(\d+)$/) { if (defined($lh)) { ferr($., $info, "Duplicate LH: entry\n"); } $lh = $1; } else { ferr($., $info, "Unknown line: $line\n"); } } # FN:, foreach my $lineno (sort({ $a <=> $b } keys(%fn))) { my $fnname = $fn{$lineno}; print("FN:$lineno,$fnname\n"); } # FNDA:, foreach my $fnname (keys(%fnda)) { if (!exists($fns{$fnname})) { ferr(undef, $info, "FNDA entry without FN: $fnname\n"); } } foreach my $fnname (sort({ $fns{$a} <=> $fns{$b} } keys(%fnda))) { my $count = $fnda{$fnname}; print("FNDA:$count,$fnname\n"); } # FNF: print("FNF:$fnf\n") if (defined($fnf)); # FNH: if (defined($fnh)) { $fnh = 0 if ($multi == 0); print("FNH:$fnh\n"); } # BRDA:,,, foreach my $lineno (sort({ $a <=> $b } keys(%brda))) { my $blocks = $brda{$lineno}; foreach my $block (sort({ $a <=> $b } keys(%{$blocks}))) { my $branches = $blocks->{$block}; foreach my $branch (sort({ $a <=> $b } keys(%{$branches}))) { my $count = $branches->{$branch}; $count = "-" if ($multi == 0); print("BRDA:$lineno,$block,$branch,$count\n"); } } } # BRF: print("BRF:$brf\n") if (defined($brf)); # BRH: if (defined($brh)) { $brh = 0 if ($multi == 0); print("BRH:$brh\n"); } # DA:, foreach my $lineno (sort({ $a <=> $b } keys(%da))) { my $count = $da{$lineno}; print("DA:$lineno,$count\n"); } # LF: print("LF:$lf\n") if (defined($lf)); # LH: if (defined($lh)) { $lh = 0 if ($multi == 0); print("LH:$lh\n"); } } sub main() { my $infofile = $ARGV[0]; my $multi = $ARGV[1]; # info: testname -> files # files: infofile -> data # data: [ starting offset, starting line ] my %info; my $fd; my $tn = ""; my %allfiles; $multi = 1 if (!defined($multi)); if (!defined($infofile)) { $infofile = "standard input"; warn("$0: Reading data from standard input\n"); open($fd, "<&STDIN") or die("$0: Could not duplicated stdin: $!\n"); } else { open($fd, "<", $infofile) or die("$0: Could not open $infofile: $!\n"); } # Register starting positions of data sets while (my $line = <$fd>) { if ($line =~ /^TN:(.*)$/) { $tn = $1; } elsif ($line =~ /^SF:(.*)$/) { my $sf = $1; my $pos = tell($fd); die("$0: Could not get file position: $!\n") if ($pos == -1); if (exists($info{$tn}->{$sf})) { ferr($., $infofile, "Duplicate entry for $tn:$sf\n"); } $info{$tn}->{$sf} = [ $pos, $. ]; $allfiles{$sf} = 1; } } # Print data sets in normalized order foreach my $filename (sort(keys(%allfiles))) { foreach my $testname (sort(keys(%info))) { my $pos = $info{$testname}->{$filename}; my ($cpos, $lpos) = @$pos; next if (!defined($pos)); if (seek($fd, $cpos, 0) != 1) { die("$0: Could not seek in $infofile: $!\n"); } printf("TN:$testname\n"); printf("SF:$filename\n"); $. = $lpos; print_sorted($fd, $infofile, $multi); printf("end_of_record\n"); } } foreach my $testname (sort(keys(%info))) { my $files = $info{$testname}; foreach my $filename (sort(keys(%{$files}))) { } } close($fd); } main(); exit(0); lcov-1.14/test/bin/mkinfo0000755000000000000000000005133313435760116014053 0ustar rootroot#!/usr/bin/env perl # # Copyright IBM Corp. 2017 # # Usage: mkinfo [-o ] [--seed ] # [=...] # # Create a fake lcov code coverage data file and optionally the corresponding # source tree. DATA_FILE contains all specifications for creating the data # file. Directives can be overridden using KEY=VALUE specifications with KEY # being in the form SECTION.KEY. SEED specifies the number used to initialize # the pseudo random number generator. # # Example: # mkinfo profiles/small -o src files.numfiles=12 # use strict; use warnings; use Getopt::Long; use Cwd qw(abs_path getcwd); use File::Path qw(make_path); use File::Basename; use Data::Dumper; my $MAX_TAKEN = 1000; my $use_colors = -t STDIN; my $BOLD = $use_colors ? "\033[1m" : ""; my $RESET = $use_colors ? "\033[0m" : ""; sub usage() { print(< [-o ] [--seed ] [=...] Create a fake lcov code coverage data file and optionally the corresponding source tree. DATA_FILE contains all specifications for creating the data file. Directives can be overridden using KEY=VALUE specifications with KEY being in the form SECTION.KEY. SEED specifies the number used to initialize the pseudo random number generator. Example: $0 profiles/small -o src files.numfiles=12 EOF } sub read_config($) { my ($filename) = @_; my $fd; my %config; my $section; open($fd, "<", $filename) or die("Could not open $filename: $!\n"); while (my $line = <$fd>) { my ($key, $value); $line =~ s/(^\s*|\s*$)//g; next if ($line eq "" || $line =~ /^#/); if ($line =~ /^\[\s*(\S+)\s*]$/) { $section = $1; next; } if ($line !~ /^(\S+)\s*=\s*(.*)$/) { die("$filename:$.: Unknown line format: $line\n"); } ($key, $value) = ($1, $2); if (!defined($section)) { die("$filename:$.: Directive outside of section\n"); } $config{$section}->{$1} = $2; } close($fd); return \%config; } sub apply_config($$) { my ($config, $directive) = @_; for my $dir (@$directive) { if ($dir !~ /^([^\.]+)\.([^=]+)=(.*)$/) { die("Unknown directive format: $dir\n"); } $config->{$1}->{$2} = $3; } } sub get_value($$;$) { my ($config, $dir, $default) = @_; my ($section, $key, $value); if ($dir !~ /^([^\.]+)\.([^=]+)$/) { die("$0: Internal error: Unknown key format: $key\n"); } ($section, $key) = ($1, $2); $value = $config->{$section}->{$key}; if (!defined($value)) { if (!defined($default)) { die("$0: Missing config value for $dir\n"); } $value = $default; } return $value; } sub get_int($$;$$$) { my ($config, $dir, $default, $min, $max) = @_; my $value = get_value($config, $dir, $default); if ($value !~ /^\d+$/) { die("$0: Config value $dir must be an integer: $value\n"); } $value = int($value); if (defined($min) && $value < $min) { die("$0: Config value $dir is too low (min $min): $value\n"); } if (defined($max) && $value > $max) { die("$0: Config value $dir is too high (max $max): $value\n"); } return int($value); } sub get_list($$;$) { my ($config, $dir, $default) = @_; my $value = get_value($config, $dir, $default); my @list = split(/\s+/, $value); return \@list; } sub randlist($) { my ($list) = @_; return "" if (!@$list); return $list->[int(rand(scalar(@$list)))]; } sub randbool() { return int(rand(2)); } # Reduce LIST to PERCENTAGE of its former size. sub reduce_list_per($$) { my ($list, $percentage) = @_; my $remove; $remove = int((100 - $percentage) * scalar(@$list) / 100); for (my $i = 0; $i < $remove; $i++) { splice(@$list, int(rand(scalar(@$list))), 1); } } # Reduce LIST to NUM items. sub reduce_list_num($$) { my ($list, $num) = @_; my $remove; $remove = scalar(@$list) - $num; for (my $i = 0; $i < $remove; $i++) { splice(@$list, int(rand(scalar(@$list))), 1); } } sub _gen_filename($$) { my ($c, $root) = @_; my $ltop = get_list($c, "files.top", ""); my $lsub = get_list($c, "files.sub", ""); my $lsubsub = get_list($c, "files.subsub", ""); my $lprefix = get_list($c, "files.prefix"); my $lsuffix = get_list($c, "files.suffix", ""); my $lext = get_list($c, "files.ext"); my ($top, $sub, $subsub, $prefix, $suffix, $ext) = ("", "", "", "", "", ""); my $filename = ""; $top = randlist($ltop) if (randbool()); $sub = randlist($lsub) if (randbool()); $subsub = randlist($lsubsub) if (randbool()); $prefix = randlist($lprefix); $suffix = randlist($lsuffix) if (randbool()); $ext = randlist($lext); $filename = $root; $filename .= "/".$top if ($top ne ""); $filename .= "/".$sub if ($sub ne ""); $filename .= "/".$subsub if ($subsub ne ""); $filename .= "/".$prefix; $filename .= "_".$suffix if ($suffix ne ""); $filename .= $ext; $filename =~ s#^//#/#; return $filename; } sub gen_filename($$$) { my ($c, $root, $filenames) = @_; my $filename; do { $filename = _gen_filename($c, $root); } while ($filenames->{$filename}); $filenames->{$filename} = 1; return $filename; } sub gen_lines($$) { my ($c, $length) = @_; my @lines = 1 .. $length; my $percent = get_int($c, "lines.instrumented", undef, 0, 100); reduce_list_per(\@lines, $percent); return \@lines; } sub gen_fnname($$) { my ($c, $hash) = @_; my $lverb = get_list($c, "functions.verb"); my $ladj = get_list($c, "functions.adj", ""); my $lnoun = get_list($c, "functions.noun", ""); my ($verb, $adj, $noun) = ("", "", ""); my $fnname; $verb = randlist($lverb); $adj = randlist($ladj) if (randbool()); $noun = randlist($lnoun) if (randbool()); $fnname = $verb; $fnname .= "_".$adj if ($adj ne ""); $fnname .= "_".$noun if ($noun ne ""); if (exists($hash->{$fnname})) { my $i = 2; while (exists($hash->{$fnname.$i})) { $i++; } $fnname .= $i; } $hash->{$fnname} = 1; return $fnname; } sub gen_functions($$) { my ($c, $lines) = @_; my @fnlines; my @functions; my %names; my $percent = get_int($c, "functions.perinstrumented", undef, 0, 100); @fnlines = @$lines; reduce_list_per(\@fnlines, $percent); foreach my $fnline (@fnlines) { push(@functions, [ $fnline, gen_fnname($c, \%names) ]); } return \@functions; } # Returns a value distribution object. This object can be used to randomly # choose one element from a list of elements with a given relative distribution. # # dist: [ sumprob, probs] # sumprob: Sum of all probabilities # probs: [ prob1, prob2, ... ] # prob: [ num, x ] # num: Value sub get_dist($$;$) { my ($c, $dir, $default) = @_; my $list = get_list($c, $dir, $default); my $sumprob = 0; my @probs; foreach my $spec (@$list) { my ($n, $p); if ($spec =~ /^(\d+):(\d+)$/) { ($n, $p) = ($1, $2); } elsif ($spec =~ /^(\d+)$/) { $n = $1; $p = 1; } else { die("$0: Config value $dir must be a distribution ". "list (a:p1 b:p2 ...)\n"); } $sumprob += $p; push(@probs, [ $n, $sumprob ]); } return [ $sumprob, \@probs ]; } sub rand_dist($) { my ($dist) = @_; my ($sumprob, $probs) = @$dist; my $r = int(rand($sumprob)); foreach my $prob (@$probs) { my ($num, $x) = @$prob; return $num if ($r < $x); } die("Internal error: Incomplete distribution list\n"); } sub gen_branches($$) { my ($c, $lines) = @_; my $percent = get_int($c, "branches.perinstrumented", undef, 0, 100); my @allblocks = @{get_list($c, "branches.blocks", "0")}; my $branchdist = get_dist($c, "branches.branchdist", "2"); my @brlines; my @branches; @brlines = @$lines; reduce_list_per(\@brlines, $percent); foreach my $brline (@brlines) { my @blocks = @allblocks; my $numblocks = int(rand(scalar(@blocks))) + 1; reduce_list_num(\@blocks, $numblocks); foreach my $block (@blocks) { my $numbranch = rand_dist($branchdist); for (my $branch = 0; $branch < $numbranch; $branch++) { push(@branches, [ $brline, $block, $branch]); } } } return \@branches; } sub gen_filesrc($) { my ($c) = @_; my ($length, $lines, $functions, $branches); my $do_ln = get_int($c, "lines.enabled"); my $do_fn = get_int($c, "functions.enabled"); my $do_br = get_int($c, "branches.enabled"); $length = 1 + int(rand(get_int($c, "lines.maxlines"))); $lines = gen_lines($c, $length); $functions = gen_functions($c, $lines) if ($do_fn); $branches = gen_branches($c, $lines) if ($do_br); return [ $length, $lines, $functions, $branches ]; } # Generate fake source tree. # # returns: [ files, numlns, numfns, numbrs ] # files: filename -> filesrc # filesrc: [ length, lines, functions, branches ] # length: Total number of lines in file # # lines: [ line1, line2, ... ] # # functions: [ fn1, fn2, ... ] # fn: [ fnline, fnname ] # fnline: Starting line of function # fnname: Function name # # branches: [ brdata1, brdata2, ...] # brdata: [ brline, block, branch ] # brline: Line number containing branches # block: Block ID # branch: Branch ID # sub gen_src($$) { my ($c, $root) = @_; my %files; my $numfiles = get_int($c, "files.numfiles"); my %filenames; my ($numlns, $numfns, $numbrs) = (0, 0, 0); for (my $i = 0; $i < $numfiles; $i++) { my $filename = gen_filename($c, $root, \%filenames); my $filesrc = gen_filesrc($c); $files{$filename} = $filesrc; $numlns += scalar(@{$filesrc->[1]}) if (defined($filesrc->[1])); $numfns += scalar(@{$filesrc->[2]}) if (defined($filesrc->[2])); $numbrs += scalar(@{$filesrc->[3]}) if (defined($filesrc->[3])); } return [ \%files, $numlns, $numfns, $numbrs ]; } sub write_src($) { my ($src) = @_; my ($files, $numlns, $numfns, $numbrs) = @$src; foreach my $filename (sort(keys(%{$files}))) { my $filesrc = $files->{$filename}; my $length = $filesrc->[0]; my $dir = dirname($filename); my $fd; if (!-d $dir) { make_path($dir) or die("Could not create directory $dir\n"); } open($fd, ">", $filename) or die("Could not create file $filename: $!\n"); for (my $i = 0; $i < $length; $i++) { print($fd "\n"); } close($fd); } } sub write_branches($$$$) { my ($fd, $branches, $brhits, $iref) = @_; my ($found, $hit) = (0, 0); # Line coverage data foreach my $brdata (@$branches) { my $brhit = $brhits->[$$iref++]; my ($brline, $block, $branch) = @$brdata; $found++; $hit++ if ($brhit ne "-" && $brhit > 0); print($fd "BRDA:$brline,$block,$branch,$brhit\n"); } if ($found > 0) { print($fd "BRF:$found\n"); print($fd "BRH:$hit\n"); } } sub write_lines($$$$) { my ($fd, $lines, $lnhist, $iref) = @_; my ($found, $hit) = (0, 0); # Line coverage data foreach my $line (@$lines) { my $lnhit = $lnhist->[$$iref++]; $found++; $hit++ if ($lnhit > 0); print($fd "DA:$line,$lnhit\n"); } print($fd "LF:$found\n"); print($fd "LH:$hit\n"); } sub write_functions($$$$) { my ($fd, $functions, $fnhits, $iref) = @_; my ($found, $hit) = (0, 0); # Function coverage data foreach my $fn (@$functions) { my ($fnline, $fnname) = @$fn; print($fd "FN:$fnline,$fnname\n"); } foreach my $fn (@$functions) { my ($fnline, $fnname) = @$fn; my $fnhit = $fnhits->[$$iref++]; $found++; $hit++ if ($fnhit > 0); print($fd "FNDA:$fnhit,$fnname\n"); } print($fd "FNF:$found\n"); print($fd "FNH:$hit\n"); } sub write_filesrc($$$$$) { my ($c, $fd, $filesrc, $hits, $iter) = @_; my ($length, $lines, $functions, $branches) = @$filesrc; my $do_ln = get_int($c, "lines.enabled"); my $do_fn = get_int($c, "functions.enabled"); my $do_br = get_int($c, "branches.enabled"); write_functions($fd, $functions, $hits->[1], \$iter->[1]) if ($do_fn); write_branches($fd, $branches, $hits->[2], \$iter->[2]) if ($do_br); write_lines($fd, $lines, $hits->[0], \$iter->[0]) if ($do_ln); } sub write_info($$$$) { my ($c, $filename, $src, $hits) = @_; my $files = $src->[0]; my $fd; my %iters; foreach my $testname (keys(%{$hits})) { $iters{$testname} = [ 0, 0, 0 ]; } open($fd, ">", $filename) or die("Could not create $filename: $!\n"); foreach my $filename (sort(keys(%{$files}))) { my $filesrc = $files->{$filename}; foreach my $testname (sort(keys(%{$hits}))) { my $testhits = $hits->{$testname}; my $iter = $iters{$testname}; print($fd "TN:$testname\n"); print($fd "SF:$filename\n"); write_filesrc($c, $fd, $filesrc, $testhits, $iter); print($fd "end_of_record\n"); } } close($fd); } sub get_hit_found($) { my ($list) = @_; my ($hit, $found) = (0, 0); foreach my $e (@$list) { $hit++ if ($e ne "-" && $e > 0); $found++; } return ($hit, $found); } sub write_counts($$) { my ($filename, $hits) = @_; my $fd; my (@tlnhits, @tfnhits, @tbrhits); foreach my $testname (keys(%{$hits})) { my $testhits = $hits->{$testname}; my ($lnhits, $fnhits, $brhits) = @$testhits; for (my $i = 0; $i < scalar(@$lnhits); $i++) { $tlnhits[$i] += $lnhits->[$i]; } for (my $i = 0; $i < scalar(@$fnhits); $i++) { $tfnhits[$i] += $fnhits->[$i]; } for (my $i = 0; $i < scalar(@$brhits); $i++) { my $h = $brhits->[$i]; $h = 0 if ($h eq "-"); $tbrhits[$i] += $h; } } open($fd, ">", $filename) or die("Could not create $filename: $!\n"); print($fd join(" ", get_hit_found(\@tlnhits), get_hit_found(\@tfnhits), get_hit_found(\@tbrhits))."\n"); close($fd); } # A branch hit value for a block that was not hit must be "-". A branch hit # value for a block that was hit cannot be "-", but must be "0" if not hit. sub sanitize_brhits($) { my ($brhits) = @_; my $block_hit = 0; foreach my $brhit_ref (@$brhits) { if ($$brhit_ref ne "-" && $$brhit_ref > 0) { $block_hit = 1; last; } } foreach my $brhit_ref (@$brhits) { if (!$block_hit) { $$brhit_ref = "-"; } elsif ($$brhit_ref eq "-") { $$brhit_ref = 0; } } } # Ensure coverage rate interdependencies are met sub sanitize_hits($$) { my ($src, $hits) = @_; my $files = $src->[0]; foreach my $hits (values(%{$hits})) { my $brhits = $hits->[2]; my $i = 0; foreach my $filename (sort(keys(%{$files}))) { my $filesrc = $files->{$filename}; my $branches = $filesrc->[3]; my $lastblock; my $lastline; my @blist; foreach my $brdata (@$branches) { my ($brline, $block, $branch) = @$brdata; if (!defined($lastblock) || $block != $lastblock || $brline != $lastline) { sanitize_brhits(\@blist); @blist = (); $lastblock = $block; $lastline = $brline; } push(@blist, \$brhits->[$i++]); } sanitize_brhits(\@blist); } } } # Generate random coverage data # # returns: testname -> testhits # testhits: [ lnhits, fnhits, brhits ] # lnhits: [ ln1hit, ln2hit, ... ] # lnhit: Number of times a line was hit by a specific test # fnhits: [ fn1hit, fn2hit, ... ] # fnhit: Number of times a function was hit by a specific test # brhits: [ br1hit, br2hit, ... ] # brhit: Number of times a branch was hit by a specific test sub gen_hits($$) { my ($c, $src) = @_; my (@lnhits, @fnhits, @brhits); my ($files, $numlns, $numfns, $numbrs) = @$src; my $testnames = get_list($c, "tests.names", ""); my %hits; $testnames = [ "" ] if (!@$testnames); foreach my $testname (@$testnames) { my (@lnhits, @fnhits, @brhits); for (my $i = 0; $i < $numlns; $i++) { push(@lnhits, 1 + int(rand($MAX_TAKEN))); } for (my $i = 0; $i < $numfns; $i++) { push(@fnhits, 1 + int(rand($MAX_TAKEN))); } for (my $i = 0; $i < $numbrs; $i++) { push(@brhits, 1 + int(rand($MAX_TAKEN))); } $hits{$testname} = [ \@lnhits, \@fnhits, \@brhits ]; } sanitize_hits($src, \%hits); return \%hits; } # Return a hash containing RATE percent of indices [0..NUM-1]. sub gen_filter($$) { my ($num, $rate) = @_; my @list = (0 .. ($num - 1)); my %hash; reduce_list_per(\@list, $rate); foreach my $i (@list) { $hash{$i} = 1; } return \%hash; } # Zero all entries in LIST identified by the indices in FILTER. sub zero_by_filter($$) { my ($list, $filter) = @_; foreach my $i (keys(%{$filter})) { $list->[$i] = 0; } } # Add a random number of indices between [0..NUM-1] to FILTER. sub widen_filter($$) { my ($filter, $num) = @_; my @list; for (my $i = 0; $i < $num; $i++) { push(@list, $i) if (!exists($filter->{$i})); } reduce_list_per(\@list, int(rand(101))); foreach my $i (@list) { $filter->{$i} = 1; } } # Zero coverage data in HITS until the combined coverage rates reach the # specified RATEs. sub reduce_hits($$$$$) { my ($src, $hits, $lnrate, $fnrate, $brrate) = @_; my ($files, $numlns, $numfns, $numbrs) = @$src; my ($lnfilter, $fnfilter, $brfilter); $lnfilter = gen_filter($numlns, 100 - $lnrate); $fnfilter = gen_filter($numfns, 100 - $fnrate); $brfilter = gen_filter($numbrs, 100 - $brrate); foreach my $testhits (values(%{$hits})) { my ($lnhits, $fnhits, $brhits) = @$testhits; zero_by_filter($lnhits, $lnfilter); zero_by_filter($fnhits, $fnfilter); zero_by_filter($brhits, $brfilter); # Provide some variation between tests widen_filter($lnfilter, $numlns); widen_filter($fnfilter, $numfns); widen_filter($brfilter, $numbrs); } sanitize_hits($src, $hits); } sub zero_list($) { my ($list) = @_; foreach my $i (@$list) { $i = 0; } } # Zero all coverage in HITS. sub zero_hits($$) { my ($src, $hits) = @_; foreach my $testhits (values(%{$hits})) { my ($lnhits, $fnhits, $brhits) = @$testhits; zero_list($lnhits); zero_list($fnhits); zero_list($brhits); } sanitize_hits($src, $hits); } # Distribute items from LIST to A and B depending on whether the index for # an item is found in FILTER. sub split_by_filter($$$$) { my ($list, $filter, $a, $b) = @_; for (my $i = 0; $i < scalar(@$list); $i++) { if (exists($filter->{$i})) { push(@$a, $list->[$i]); push(@$b, 0); } else { push(@$a, 0); push(@$b, $list->[$i]); } } } sub split_hits($$$) { my ($c, $src, $hits) = @_; my ($files, $numlns, $numfns, $numbrs) = @$src; my ($lnsplit, $fnsplit, $brsplit); my (%a, %b); $lnsplit = gen_filter($numlns, int(rand(101))); $fnsplit = gen_filter($numfns, int(rand(101))); $brsplit = gen_filter($numbrs, int(rand(101))); foreach my $testname (keys(%{$hits})) { my $testhits = $hits->{$testname}; my ($lnhits, $fnhits, $brhits) = @$testhits; my (@lnhitsa, @fnhitsa, @brhitsa); my (@lnhitsb, @fnhitsb, @brhitsb); split_by_filter($lnhits, $lnsplit, \@lnhitsa, \@lnhitsb); split_by_filter($fnhits, $fnsplit, \@fnhitsa, \@fnhitsb); split_by_filter($brhits, $brsplit, \@brhitsa, \@brhitsb); $a{$testname} = [ \@lnhitsa, \@fnhitsa, \@brhitsa ]; $b{$testname} = [ \@lnhitsb, \@fnhitsb, \@brhitsb ]; } sanitize_hits($src, \%a); sanitize_hits($src, \%b); return (\%a, \%b); } sub plural($$$) { my ($num, $sing, $plur) = @_; return $num <= 1 ? $sing : $plur; } sub print_intro($) { my ($c) = @_; my $numtests = scalar(@{get_list($c, "tests.names")}); my $numfiles = get_int($c, "files.numfiles"); $numtests = 1 if ($numtests < 1); print($BOLD."Creating coverage files ($numtests ". plural($numtests, "test", "tests").", $numfiles ". plural($numfiles, "source file", "source files").")\n".$RESET); } sub main() { my $opt_help; my $opt_output; my $opt_configfile; my $opt_seed = 0; my $c; my $src; my $hits; my $root; my $enum; my ($a, $b); # Parse options if (!GetOptions("output|o=s" => \$opt_output, "seed=s" => \$opt_seed, "help|h" => \$opt_help, )) { print(STDERR "Use $0 --help to get usage information\n"); exit(2); } if ($opt_help) { usage(); exit(0); } $opt_configfile = shift(@ARGV); if (!defined($opt_configfile)) { print(STDERR "Please specify a config file\n"); exit(2); } if (defined($opt_output)) { if (! -d $opt_output) { mkdir($opt_output) or die("$0: Could not create directory ". "$opt_output: $!\n"); } $root = abs_path($opt_output) } else { $root = "/"; } srand($opt_seed); # Get config $c = read_config($opt_configfile); apply_config($c, \@ARGV) if (@ARGV); print_intro($c); # Show lines on STDOUT without newline $| = 1; # Create source tree print(" Source tree ......... "); $src = gen_src($c, $root); # Write out source code if requested write_src($src) if (defined($opt_output)); print("done ("); print($src->[1]." lines, "); print($src->[2]." functions, "); print($src->[3]." branches)\n"); # Write out full-coverage data files print(" Full coverage ....... "); $hits = gen_hits($c, $src); write_info($c, "full.info", $src, $hits); write_counts("full.counts", $hits); print("done\n"); # Write out data files with target coverage rates print(" Target coverage ..... "); reduce_hits($src, $hits, get_int($c, "lines.covered"), get_int($c, "functions.covered"), get_int($c, "branches.covered")); write_info($c, "target.info", $src, $hits); write_counts("target.counts", $hits); print("done\n"); # Write out partial data files print(" Partial coverage .... "); ($a, $b) = split_hits($c, $src, $hits); write_info($c, "part1.info", $src, $a); write_counts("part1.counts", $a); write_info($c, "part2.info", $src, $b); write_counts("part2.counts", $b); print("done\n"); # Write out zero-coverage data files print(" Zero coverage ....... "); zero_hits($src, $hits); write_info($c, "zero.info", $src, $hits); write_counts("zero.counts", $hits); print("done\n"); } main(); exit(0); lcov-1.14/test/bin/common0000644000000000000000000000300213067205540014037 0ustar rootrootfunction elapsed_to_ms() { local ELAPSED=$1 local IFS=:. local MS set -- $ELAPSED if [ $# -eq 3 ] ; then let MS=${3#0}*10+${2#0}*1000+$1*60000 else let MS=${4#0}*10+${3#0}*1000+${2#0}*60000+$1*3600000 fi echo $MS } function t_timestamp() { date +"%Y-%m-%d %H:%M:%S %z" } function t_marker() { echo echo "======================================================================" } function t_detail() { local KEY=$1 local VALUE=$2 local DOTS=" ............" printf "%-.12s: %s\n" "$KEY$DOTS" "$VALUE" } function t_announce() { local TESTNAME="$1" printf "$BOLD%-.30s$RESET " "$TESTNAME .............................." t_marker >> "$LOGFILE" t_detail "DATE" "$(t_timestamp)" >> "$LOGFILE" t_detail "TESTNAME" "$TESTNAME" >> "$LOGFILE" } function t_result() { local COLOR="$1" local TEXT="$2" printf "[$COLOR$TEXT$RESET]" } function t_pass() { local TESTNAME="$1" t_result "$GREEN" "pass" echo "pass $TESTNAME" >> "$COUNTFILE" } function t_fail() { local TESTNAME="$1" t_result "$RED" "fail" echo "fail $TESTNAME" >> "$COUNTFILE" } function t_kill() { local TESTNAME="$1" t_result "$RED" "kill" echo "fail $TESTNAME" >> "$COUNTFILE" } function t_skip() { local TESTNAME="$1" t_result "$BLUE" "skip" echo "skip $TESTNAME" >> "$COUNTFILE" } function t_indent() { sed -e 's/^/ /' } LOGFILE="$TOPDIR/test.log" COUNTFILE="$TOPDIR/test.counts" TIMEFILE="$TOPDIR/test.time" if [ -t 1 ] ; then RED="\e[31m" GREEN="\e[32m" BLUE="\e[34m" BOLD="\e[1m" DEFAULT="\e[39m" RESET="\e[0m" fi lcov-1.14/test/bin/testsuite_init0000755000000000000000000000103313435760116015634 0ustar rootroot#!/usr/bin/env bash # # Copyright IBM Corp. 2017 # # Usage: testsuite_init # # Announce start of test suite and prepare log files. # TOPDIR=$(realpath $(dirname $0)/..) && source "$TOPDIR/bin/common" echo -e $BOLD"Starting tests"$RESET echo "start_time $(date +%s.%N)" >"$COUNTFILE" exec >"$LOGFILE" 2>&1 t_detail "DATE" "$(t_timestamp)" t_detail "LCOV" "" lcov --version 2>&1 | t_indent t_detail "GCOV" "" gcov --version 2>&1 | t_indent t_detail "CPUINFO" "" t_indent < /proc/cpuinfo t_detail "MEMINFO" "" t_indent < /proc/meminfo lcov-1.14/test/bin/testsuite_exit0000755000000000000000000000316113435760116015646 0ustar rootroot#!/usr/bin/env bash # # Copyright IBM Corp. 2017 # # Usage: testsuite_exit # # Announce end of test suite and show aggregate results. # TOPDIR=$(realpath $(dirname $0)/..) && source "$TOPDIR/bin/common" echo "end_time $(date +%s.%N)" >>"$COUNTFILE" SUCCESS=0 FAILED=0 SKIPPED=0 TOTAL_TIME=0 TOTAL_MEM=0 HAVE_EXT=0 # Get results while read LINE ; do set -- $LINE case "$1" in start_time) START_TIME=$2 ;; end_time) END_TIME=$2 ;; pass) let SUCCESS=$SUCCESS+1 ;; fail) let FAILED=$FAILED+1 ;; skip) let SKIPPED=$SKIPPED+1 ;; elapsed) let TOTAL_TIME=$TOTAL_TIME+$3 ; HAVE_EXT=1 ;; resident) let TOTAL_MEM=$TOTAL_MEM+$3 ; HAVE_EXT=1 ;; esac done < "$COUNTFILE" exec 3>&1 exec >>"$LOGFILE" 2>&1 t_marker t_detail "DATE" "$(t_timestamp)" let TOTAL=$SUCCESS+$SKIPPED+$FAILED t_detail "EXECUTED" "$TOTAL" t_detail "PASSED" "$SUCCESS" t_detail "FAILED" "$FAILED" t_detail "SKIPPED" "$SKIPPED" [ $HAVE_EXT -eq 1 ] && t_detail "TIME" "${TOTAL_TIME}ms" [ $HAVE_EXT -eq 1 ] && t_detail "MEM" "${TOTAL_MEM}kB" TOTAL_TIME=$(($TOTAL_TIME/1000)).$(($TOTAL_TIME%1000/100)) TOTAL_MEM=$(($TOTAL_MEM/1024)).$((($TOTAL_MEM%1024)/100)) TOTAL="$BOLD$TOTAL tests executed$RESET" PASS="$SUCCESS passed" FAIL="$FAILED failed" SKIP="$SKIPPED skipped" TIME="time ${TOTAL_TIME}s" MEM="mem ${TOTAL_MEM}MB" [ "$SUCCESS" -gt 0 ] && PASS="$GREEN$PASS$DEFAULT" [ "$FAILED" -gt 0 ] && FAIL="$RED$FAIL$DEFAULT" [ "$SKIPPED" -gt 0 ] && SKIP="$BLUE$SKIP$DEFAULT" echo -en "$TOTAL, $PASS, $FAIL, $SKIP$RESET" >&3 [ $HAVE_EXT -eq 1 ] && echo -n " ($TIME, $MEM)" >&3 echo >&3 echo "Result log stored in $LOGFILE" >&3 if [ "$FAILED" -gt 0 ] ; then exit 1 fi exit 0 lcov-1.14/test/bin/test_skip0000755000000000000000000000076313435760116014576 0ustar rootroot#!/usr/bin/env bash # # Copyright IBM Corp. 2017 # # Usage: test_skip # # Announce and record that a single test case was skipped, including an # optional reason text. Must be run after testsuite_init. # TOPDIR=$(realpath $(dirname $0)/..) && source "$TOPDIR/bin/common" TESTNAME="$1" REASON="${*:2}" ; [ -z "$REASON" ] && REASON="" t_announce "$TESTNAME" t_skip "$TESTNAME" echo t_detail "REASON" "$REASON" >>"$LOGFILE" t_detail "REASON" "$REASON" | t_indent lcov-1.14/test/lcovrc0000644000000000000000000000012513067205540013272 0ustar rootroot# lcovrc file used during tests lcov_function_coverage = 1 lcov_branch_coverage = 1 lcov-1.14/test/common.mak0000644000000000000000000000253613071704076014054 0ustar rootrootTOPDIR := $(dir $(realpath $(lastword $(MAKEFILE_LIST)))) TESTDIR := $(dir $(realpath $(firstword $(MAKEFILE_LIST)))) PARENTDIR := $(dir $(patsubst %/,%,$(TOPDIR))) RELDIR := $(TESTDIR:$(PARENTDIR)%=%) ZEROINFO := $(TOPDIR)zero.info ZEROCOUNTS := $(TOPDIR)zero.counts FULLINFO := $(TOPDIR)full.info FULLCOUNTS := $(TOPDIR)full.counts TARGETINFO := $(TOPDIR)target.info TARGETCOUNTS := $(TOPDIR)target.counts PART1INFO := $(TOPDIR)part1.info PART1COUNTS := $(TOPDIR)part1.counts PART2INFO := $(TOPDIR)part2.info PART2COUNTS := $(TOPDIR)part2.counts INFOFILES := $(ZEROINFO) $(FULLINFO) $(TARGETINFO) $(PART1INFO) $(PART2INFO) COUNTFILES := $(ZEROCOUNTS) $(FULLCOUNTS) $(TARGETCOUNTS) $(PART1COUNTS) \ $(PART2COUNTS) LCOVRC := $(TOPDIR)lcovrc LCOVFLAGS := --config-file $(LCOVRC) SIZE := small CC := gcc export LCOV := lcov $(LCOVFLAGS) export GENHTML := genhtml $(LCOVFLAGS) export PATH := $(TOPDIR)/../bin:$(TOPDIR)/bin:$(PATH) export LANG := C all: prepare init test exit init: testsuite_init exit: testsuite_exit prepare: $(INFOFILES) $(COUNTFILES) clean: clean_common clean_common: echo " CLEAN $(patsubst %/,%,$(RELDIR))" $(INFOFILES) $(COUNTFILES): cd $(TOPDIR) && mkinfo profiles/$(SIZE) -o src/ ifneq ($(V),2) .SILENT: endif .PHONY: all init exit prepare clean clean_common lcov-1.14/test/lcov_diff/0000755000000000000000000000000013436014706014016 5ustar rootrootlcov-1.14/test/lcov_diff/diff_test0000755000000000000000000000202713435760116015716 0ustar rootroot#!/usr/bin/env bash # # Copyright IBM Corp. 2017 # # Usage: diff_test # # Check lcov's diff function: # - Compile two slightly different test programs # - Run the programs and collect coverage data # - Generate a patch containing the difference between the source code # - Apply the patch to the coverage data # - Compare the resulting patched coverage data file with the data from the # patched source file # function die() { echo "Error: $@" >&2 exit 1 } make -C old || die "Failed to compile old source" make -C new || die "Failed to compile new source" diff -u $PWD/old/prog.c $PWD/new/prog.c > diff $LCOV --diff old/prog.info diff --convert-filenames -o patched.info -t bla || \ die "Failed to apply patch to coverage data file" norminfo new/prog.info > new_normalized.info norminfo patched.info > patched_normalized.info sed -i -e 's/^TN:.*$/TN:/' patched_normalized.info diff -u patched_normalized.info new_normalized.info || \ die "Mismatch in patched coverage data file" echo "Patched coverage data file matches expected file" lcov-1.14/test/lcov_diff/new/0000755000000000000000000000000013436014706014607 5ustar rootrootlcov-1.14/test/lcov_diff/new/prog.c0000644000000000000000000000041513071363720015720 0ustar rootroot int fn(int x) { switch (x) { case -1: return 0; case 0: return 2; case 2: return 3; case 12: return 7; default: return 255; } } int fn2() { return 7; } int main(int argc, char *argv[]) { if (argc > 1) return fn(argc); return fn2(); } lcov-1.14/test/lcov_diff/new/Makefile0000644000000000000000000000035613071363720016251 0ustar rootrootprog.info: include ../../common.mak prog.info: prog.gcda $(LCOV) -c -d . -o prog.info prog.gcda: prog ./prog || true prog: prog.c $(CC) prog.c -o prog --coverage clean: rm -f prog prog.gcda prog.gcno prog.info .PHONY: all clean lcov-1.14/test/lcov_diff/old/0000755000000000000000000000000013436014706014574 5ustar rootrootlcov-1.14/test/lcov_diff/old/prog.c0000644000000000000000000000037213071363720015707 0ustar rootrootint fn(int x) { switch (x) { case -1: return 0; case 0: return 2; case 2: return 3; case 12: return 7; default: return 255; } } int fn2() { return 7; } int main(int argc, char *argv[]) { if (argc > 1) return fn(argc); return fn2(); } lcov-1.14/test/lcov_diff/old/Makefile0000644000000000000000000000035613071363720016236 0ustar rootrootprog.info: include ../../common.mak prog.info: prog.gcda $(LCOV) -c -d . -o prog.info prog.gcda: prog ./prog || true prog: prog.c $(CC) prog.c -o prog --coverage clean: rm -f prog prog.gcda prog.gcno prog.info .PHONY: all clean lcov-1.14/test/lcov_diff/Makefile0000644000000000000000000000020413071363720015450 0ustar rootrootinclude ../common.mak test: test_run lcov_diff_apply ./diff_test clean: make -C old clean make -C new clean rm -f *.info diff lcov-1.14/test/lcov_add_files/0000755000000000000000000000000013436014706015020 5ustar rootrootlcov-1.14/test/lcov_add_files/add_test0000755000000000000000000000177713435760116016553 0ustar rootroot#!/usr/bin/env bash # # Copyright IBM Corp. 2017 # # Usage: add_test [...] # # Add multiple coverage data files, normalize the output and multiply counts # with multiplier. Compare against reference file. Report deviations. # MULTI=$1 REFFILE=$2 shift 2 ADD= for INFO in $* ; do ADD="$ADD -a $INFO" done if [ -z "$MULTI" -o -z "$REFFILE" -o -z "$ADD" ] ; then echo "Usage: $0 [...]" >&2 exit 1 fi OUTFILE="add_"$(basename "$REFFILE") SORTFILE="norm_$OUTFILE" set -x echo "Adding files..." if ! $LCOV $ADD -o "$OUTFILE" ; then echo "Error: lcov returned with non-zero exit code $?" >&2 exit 1 fi echo "Normalizing result..." if ! norminfo "$OUTFILE" "$MULTI" > "$SORTFILE" ; then echo "Error: Normalization of lcov result file failed" >&2 exit 1 fi echo "Comparing with reference..." if ! diff -u "$REFFILE" "$SORTFILE" ; then echo "Error: Result of combination differs from reference file" >&2 exit 1 fi lcov-1.14/test/lcov_add_files/Makefile0000644000000000000000000000305313071363720016457 0ustar rootrootinclude ../common.mak ADDTEST := ./add_test TESTS := lcov_add_zero lcov_add_zero2 lcov_add_full lcov_add_full2 \ lcov_add_part lcov_add_part2 lcov_add_concatenated4 test: $(TESTS) lcov_add_zero: # Add single zero coverage file - output should be same as input test_run lcov_add_zero $(ADDTEST) 1 "$(ZEROINFO)" "$(ZEROINFO)" lcov_add_zero2: # Add two zero coverage files - output should be same as input test_run lcov_add_zero2 $(ADDTEST) 1 "$(ZEROINFO)" "$(ZEROINFO)" "$(ZEROINFO)" lcov_add_full: # Add single 100% coverage file - output should be same as input test_run lcov_add_full $(ADDTEST) 1 "$(FULLINFO)" "$(FULLINFO)" lcov_add_full2: # Add two 100% coverage file and reduce counts to 1/2 - output should # be same as input test_run lcov_add_full2 $(ADDTEST) 0.5 "$(FULLINFO)" "$(FULLINFO)" "$(FULLINFO)" lcov_add_part: # Add single coverage file with random coverage rate - output should # be same as input test_run lcov_add_part $(ADDTEST) 1 "$(PART1INFO)" "$(PART1INFO)" lcov_add_part2: # Add two coverage files that were split from target file - output # should be same as target file test_run lcov_add_part2 $(ADDTEST) 1 "$(TARGETINFO)" "$(PART1INFO)" "$(PART2INFO)" lcov_add_concatenated4: # Add coverage file that consists of 4 concatenation of target files # and reduce counts to 1/4 - output should be the same as input cat $(TARGETINFO) $(TARGETINFO) $(TARGETINFO) $(TARGETINFO) >concatenated.info test_run lcov_add_concatenated4 $(ADDTEST) 0.25 $(TARGETINFO) concatenated.info clean: rm -f *.info .PHONY: test $(TESTS) clean lcov-1.14/test/lcov_summary/0000755000000000000000000000000013436014706014603 5ustar rootrootlcov-1.14/test/lcov_summary/check_counts0000755000000000000000000000374113435760116017210 0ustar rootroot#!/usr/bin/env perl # # Copyright IBM Corp. 2017 # # Usage: check_counts # # Compare the output of "lcov --summary" for with the # coverage data counts specified in . This file has the following # format (all in a single line): # # lnhit lnfound fnhit fnfound brhit brfound2 # use strict; use warnings; sub do_cmp($$$) { my ($title, $a, $b) = @_; if ($a == $b) { print("$title: $a == $b\n"); return 0; } else { print("$title: $a != $b => mismatch!\n"); return 1; } } my $lcov = $ENV{"LCOV"}; my ($counts, $info) = @ARGV; my $fd; my $cmdline; my ($lnhit, $lnfound, $fnhit, $fnfound, $brhit, $brfound) = (0, 0, 0, 0, 0, 0); my ($lnhit2, $lnfound2, $fnhit2, $fnfound2, $brhit2, $brfound2); my $rc = 0; die("$0: LCOV environment variable not defined\n") if (!defined($lcov)); if (!defined($counts) || !defined($info)) { die("Usage: $0 \n"); } $cmdline = "$lcov --summary $info"; open($fd, "-|", $cmdline) or die("$0: Could not run $cmdline: $!\n"); while (<$fd>) { ($lnhit, $lnfound) = ($1, $2) if (/(\d+) of (\d+) lines/); ($fnhit, $fnfound) = ($1, $2) if (/(\d+) of (\d+) functions/); ($brhit, $brfound) = ($1, $2) if (/(\d+) of (\d+) branches/); } close($fd); die("$0: Non-zero result code ($?) of command: $cmdline\n") if ($? != 0); open($fd, "<", $counts) or die("$0: Could not open $counts: $!\n"); if (<$fd> !~ /^(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\s+(\d+)\s+(\d+)/) { die("$0: Invalid count file: $counts\n"); } ($lnhit2, $lnfound2, $fnhit2, $fnfound2, $brhit2, $brfound2) = ($1, $2, $3, $4, $5, $6); close($fd); print("Comparing --summary output for $info and $counts:\n"); $rc |= do_cmp("line hit", $lnhit, $lnhit2); $rc |= do_cmp("line found", $lnfound, $lnfound2); $rc |= do_cmp("functions hit", $fnhit, $fnhit2); $rc |= do_cmp("functions found", $fnfound, $fnfound2); $rc |= do_cmp("branches hit", $brhit, $brhit2); $rc |= do_cmp("branches found", $brfound, $brfound2); exit($rc); lcov-1.14/test/lcov_summary/Makefile0000644000000000000000000000300013067205540016232 0ustar rootrootinclude ../common.mak CHECK := ./check_counts TESTS := lcov_summary_zero lcov_summary_full lcov_summary_target \ lcov_summary_part1 lcov_summary_part2 lcov_summary_concatenated \ lcov_summary_concatenated2 test: $(TESTS) lcov_summary_zero: # Compare output of lcov --summary with generated counts test_run lcov_summary_zero $(CHECK) $(ZEROCOUNTS) $(ZEROINFO) lcov_summary_full: # Compare output of lcov --summary with generated counts test_run lcov_summary_full $(CHECK) $(FULLCOUNTS) $(FULLINFO) lcov_summary_target: # Compare output of lcov --summary with generated counts test_run lcov_summary_target $(CHECK) $(TARGETCOUNTS) $(TARGETINFO) lcov_summary_part1: # Compare output of lcov --summary with generated counts test_run lcov_summary_part1 $(CHECK) $(PART1COUNTS) $(PART1INFO) lcov_summary_part2: # Compare output of lcov --summary with generated counts test_run lcov_summary_part2 $(CHECK) $(PART2COUNTS) $(PART2INFO) lcov_summary_concatenated: # Compare output of lcov --summary with generated counts for a # concatenated coverage data file cat $(TARGETINFO) $(TARGETINFO) > concatenated.info test_run lcov_summary_concatenated $(CHECK) $(TARGETCOUNTS) concatenated.info lcov_summary_concatenated2: # Compare output of lcov --summary with generated counts for a # concatenated coverage data file (part1+part2=target) cat $(PART1INFO) $(PART2INFO) > concatenated2.info test_run lcov_summary_concatenated2 $(CHECK) $(TARGETCOUNTS) concatenated2.info clean: rm -f *.info .PHONY: test $(TESTS) clean lcov-1.14/test/genhtml_output/0000755000000000000000000000000013436014706015141 5ustar rootrootlcov-1.14/test/genhtml_output/Makefile0000644000000000000000000000154013071704076016602 0ustar rootrootinclude ../common.mak GENHTML_TEST := ./genhtml_test TESTS := genhtml_output_zero genhtml_output_full genhtml_output_target \ genhtml_output_part1 genhtml_output_part2 genhtml_output_combined test: $(TESTS) genhtml_output_zero: @test_run genhtml_output_zero $(GENHTML) $(ZEROINFO) -o out_zero/ genhtml_output_full: @test_run genhtml_output_full $(GENHTML) $(FULLINFO) -o out_full/ genhtml_output_target: @test_run genhtml_output_target $(GENHTML) $(TARGETINFO) -o out_target/ genhtml_output_part1: @test_run genhtml_output_part1 $(GENHTML) $(PART1INFO) -o out_part1/ genhtml_output_part2: @test_run genhtml_output_part2 $(GENHTML) $(PART2INFO) -o out_part2/ genhtml_output_combined: genhtml_output_target @test_run genhtml_output_combined $(GENHTML_TEST) $(TARGETINFO) $(PART1INFO) $(PART2INFO) clean: rm -rf out_*/ .PHONY: test $(TESTS) clean lcov-1.14/test/genhtml_output/genhtml_test0000755000000000000000000000141313435760116017565 0ustar rootroot#!/usr/bin/env bash # # Copyright IBM Corp. 2017 # # Usage: genhtml_test [...] # # Compare genhtml output of a reference coverage data file with that of # a combination of multiple files. # function die() { echo "Error: $@" >&2 exit 1 } GENHTMLFLAGS="-t title" REFFILE=$1 shift if [ -z "$REFFILE" -o -z "$*" ] ; then echo "Usage: $0 [...]" >&2 exit 2 fi OUTREF="out_$(basename $REFFILE .info)" OUTCOMBINED="out_combined" $GENHTML $GENHTMLFLAGS "$REFFILE" -o "$OUTREF" || \ die "Could not generate HTML for reference file" $GENHTML $GENHTMLFLAGS "$@" -o "$OUTCOMBINED" || \ die "Could not generate HTML for combined files" diff -ur "$OUTREF" "$OUTCOMBINED" -I "headerValue" || \ die "Mismatch in generated output" lcov-1.14/test/Makefile0000644000000000000000000000113313071704076013522 0ustar rootrootinclude common.mak TESTDIRS := $(sort $(patsubst %/,%,$(dir $(wildcard */Makefile)))) help: info info: echo "Available make targets:" echo " test : perform self-tests" echo " clean : remove all temporary files" echo "" echo "Available make variables:" echo " SIZE : specify size of test data (small, medium, large)" echo " V : specify level of verbosity (0, 1, 2)" test: for TEST in $(TESTDIRS) ; do \ make -C $$TEST test ; \ done clean: rm -rf *.info *.counts test.log src/ for TEST in $(TESTDIRS) ; do \ make -C $$TEST clean ; \ done .PHONY: help info test clean lcov-1.14/rpm/0000755000000000000000000000000013436014710011675 5ustar rootrootlcov-1.14/rpm/lcov.spec0000644000000000000000000000334413436002743013523 0ustar rootrootSummary: A graphical GCOV front-end Name: lcov Version: 1.14 Release: 1 License: GPLv2+ Group: Development/Tools URL: http://ltp.sourceforge.net/coverage/lcov.php Source0: http://downloads.sourceforge.net/ltp/%{name}-%{version}.tar.gz BuildRoot: %{_tmppath}/%{name}-%{version}-root BuildArch: noarch Requires: perl >= 5.8.8 %description LCOV is a graphical front-end for GCC's coverage testing tool gcov. It collects gcov data for multiple source files and creates HTML pages containing the source code annotated with coverage information. It also adds overview pages for easy navigation within the file structure. %prep %setup -q -n %{name}-%{version} %build exit 0 %install rm -rf $RPM_BUILD_ROOT make install DESTDIR=$RPM_BUILD_ROOT PREFIX=/usr CFG_DIR=/etc %clean rm -rf $RPM_BUILD_ROOT %files %defattr(-,root,root) /usr/bin/* /usr/share/man/man*/* %config /etc/* %changelog * Mon Aug 22 2016 Peter Oberparleiter (Peter.Oberparleiter@de.ibm.com) - updated "make install" call to work with PREFIX Makefile changes * Mon May 07 2012 Peter Oberparleiter (Peter.Oberparleiter@de.ibm.com) - added dependency on perl 5.8.8 for >>& open mode support * Wed Aug 13 2008 Peter Oberparleiter (Peter.Oberparleiter@de.ibm.com) - changed description + summary text * Mon Aug 20 2007 Peter Oberparleiter (Peter.Oberparleiter@de.ibm.com) - fixed "Copyright" tag * Mon Jul 14 2003 Peter Oberparleiter (Peter.Oberparleiter@de.ibm.com) - removed variables for version/release to support source rpm building - added initial rm command in install section * Mon Apr 7 2003 Peter Oberparleiter (Peter.Oberparleiter@de.ibm.com) - implemented variables for version/release * Fri Oct 18 2002 Peter Oberparleiter (Peter.Oberparleiter@de.ibm.com) - created initial spec file lcov-1.14/man/0000755000000000000000000000000013436014710011652 5ustar rootrootlcov-1.14/man/lcovrc.50000644000000000000000000004316213436002743013241 0ustar rootroot.TH lcovrc 5 "LCOV 1.14" 2019\-02\-28 "User Manuals" .SH NAME lcovrc \- lcov configuration file .SH DESCRIPTION The .I lcovrc file contains configuration information for the .B lcov code coverage tool (see .BR lcov (1)). .br The system\-wide configuration file is located at .IR /etc/lcovrc . To change settings for a single user, place a customized copy of this file at location .IR ~/.lcovrc . Where available, command\-line options override configuration file settings. Lines in a configuration file can either be: .IP " *" empty lines or lines consisting only of white space characters. These lines are ignored. .IP " *" comment lines which start with a hash sign ('#'). These are treated like empty lines and will be ignored. .IP " *" statements in the form .RI ' key " = " value '. A list of valid statements and their description can be found in section 'OPTIONS' below. .PP .B Example configuration: .IP # .br # Example LCOV configuration file .br # .br # External style sheet file .br #genhtml_css_file = gcov.css .br # Coverage rate limits .br genhtml_hi_limit = 90 .br genhtml_med_limit = 75 .br # Width of line coverage field in source code view .br genhtml_line_field_width = 12 .br # Width of branch coverage field in source code view .br genhtml_branch_field_width = 16 .br # Width of overview image .br genhtml_overview_width = 80 .br # Resolution of overview navigation .br genhtml_nav_resolution = 4 .br # Offset for source code navigation .br genhtml_nav_offset = 10 .br # Do not remove unused test descriptions if non\-zero .br genhtml_keep_descriptions = 0 .br # Do not remove prefix from directory names if non\-zero .br genhtml_no_prefix = 0 .br # Do not create source code view if non\-zero .br genhtml_no_source = 0 .br # Specify size of tabs .br genhtml_num_spaces = 8 .br # Highlight lines with converted\-only data if non\-zero .br genhtml_highlight = 0 .br # Include color legend in HTML output if non\-zero .br genhtml_legend = 0 .br # Include HTML file at start of HTML output .br #genhtml_html_prolog = prolog.html .br # Include HTML file at end of HTML output .br #genhtml_html_epilog = epilog.html .br # Use custom HTML file extension .br #genhtml_html_extension = html .br # Compress all generated html files with gzip. .br #genhtml_html_gzip = 1 .br # Include sorted overview pages .br genhtml_sort = 1 .br # Include function coverage data display .br #genhtml_function_coverage = 1 .br # Include branch coverage data display .br #genhtml_branch_coverage = 1 .br # Specify the character set of all generated HTML pages .br genhtml_charset=UTF\-8 .br # Allow HTML markup in test case description text if non\-zero .br genhtml_desc_html=0 .br # Specify the precision for coverage rates .br #genhtml_precision=1 .br # Show missed counts instead of hit counts .br #genhtml_missed=1 .br # Demangle C++ symbols .br #genhtml_demangle_cpp=1 .br # Location of the gcov tool .br #geninfo_gcov_tool = gcov .br # Adjust test names if non\-zero .br #geninfo_adjust_testname = 0 .br # Calculate a checksum for each line if non\-zero .br geninfo_checksum = 0 .br # Enable libtool compatibility mode if non\-zero .br geninfo_compat_libtool = 0 .br # Specify whether to capture coverage data for external source .br # files .br #geninfo_external = 1 .br # Use gcov's --all-blocks option if non-zero .br #geninfo_gcov_all_blocks = 1 .br # Specify compatiblity modes (same as \-\-compat option .br # of geninfo) .br #geninfo_compat = libtool=on, hammer=auto, split_crc=auto .br # Adjust path to source files by removing or changing path .br # components that match the specified pattern (Perl regular .br # expression format) .br #geninfo_adjust_src_path = /tmp/build => /usr/src # Specify if geninfo should try to automatically determine .br # the base-directory when collecting coverage data. .br geninfo_auto_base = 1 .br # Directory containing gcov kernel files .br lcov_gcov_dir = /proc/gcov .br # Location for temporary directories .br lcov_tmp_dir = /tmp .br # Show full paths during list operation if non\-zero .br lcov_list_full_path = 0 .br # Specify the maximum width for list output. This value is .br # ignored when lcov_list_full_path is non\-zero. .br lcov_list_width = 80 .br # Specify the maximum percentage of file names which may be .br # truncated when choosing a directory prefix in list output. .br # This value is ignored when lcov_list_full_path is non\-zero. .br lcov_list_truncate_max = 20 # Specify if function coverage data should be collected and .br # processed. .br lcov_function_coverage = 1 .br # Specify if branch coverage data should be collected and .br # processed. .br lcov_branch_coverage = 0 .br .PP .SH OPTIONS .BR genhtml_css_file " =" .I filename .IP Specify an external style sheet file. Use this option to modify the appearance of the HTML output as generated by .BR genhtml . During output generation, a copy of this file will be placed in the output directory. .br This option corresponds to the \-\-css\-file command line option of .BR genhtml . .br By default, a standard CSS file is generated. .PP .BR genhtml_hi_limit " =" .I hi_limit .br .BR genhtml_med_limit " =" .I med_limit .br .IP Specify coverage rate limits for classifying file entries. Use this option to modify the coverage rates (in percent) for line, function and branch coverage at which a result is classified as high, medium or low coverage. This classification affects the color of the corresponding entries on the overview pages of the HTML output: .br High: hi_limit <= rate <= 100 default color: green .br Medium: med_limit <= rate < hi_limit default color: orange .br Low: 0 <= rate < med_limit default color: red .br Defaults are 90 and 75 percent. .PP .BR genhtml_line_field_width " =" .I number_of_characters .IP Specify the width (in characters) of the source code view column containing line coverage information. .br Default is 12. .PP .BR genhtml_branch_field_width " =" .I number_of_characters .IP Specify the width (in characters) of the source code view column containing branch coverage information. .br Default is 16. .PP .BR genhtml_overview_width " =" .I pixel_size .IP Specify the width (in pixel) of the overview image created when generating HTML output using the \-\-frames option of .BR genhtml . .br Default is 80. .PP .BR genhtml_nav_resolution " =" .I lines .IP Specify the resolution of overview navigation when generating HTML output using the \-\-frames option of .BR genhtml . This number specifies the maximum difference in lines between the position a user selected from the overview and the position the source code window is scrolled to. .br Default is 4. .PP .BR genhtml_nav_offset " =" .I lines .IP Specify the overview navigation line offset as applied when generating HTML output using the \-\-frames option of .BR genhtml. .br Clicking a line in the overview image should show the source code view at a position a bit further up, so that the requested line is not the first line in the window. This number specifies that offset. .br Default is 10. .PP .BR genhtml_keep_descriptions " =" .IR 0 | 1 .IP If non\-zero, keep unused test descriptions when generating HTML output using .BR genhtml . .br This option corresponds to the \-\-keep\-descriptions option of .BR genhtml . .br Default is 0. .PP .BR genhtml_no_prefix " =" .IR 0 | 1 .IP If non\-zero, do not try to find and remove a common prefix from directory names. .br This option corresponds to the \-\-no\-prefix option of .BR genhtml . .br Default is 0. .PP .BR genhtml_no_source " =" .IR 0 | 1 .IP If non\-zero, do not create a source code view when generating HTML output using .BR genhtml . .br This option corresponds to the \-\-no\-source option of .BR genhtml . .br Default is 0. .PP .BR genhtml_num_spaces " =" .I num .IP Specify the number of spaces to use as replacement for tab characters in the HTML source code view as generated by .BR genhtml . .br This option corresponds to the \-\-num\-spaces option of .BR genthml . .br Default is 8. .PP .BR genhtml_highlight " =" .IR 0 | 1 .IP If non\-zero, highlight lines with converted\-only data in HTML output as generated by .BR genhtml . .br This option corresponds to the \-\-highlight option of .BR genhtml . .br Default is 0. .PP .BR genhtml_legend " =" .IR 0 | 1 .IP If non\-zero, include a legend explaining the meaning of color coding in the HTML output as generated by .BR genhtml . .br This option corresponds to the \-\-legend option of .BR genhtml . .br Default is 0. .PP .BR genhtml_html_prolog " =" .I filename .IP If set, include the contents of the specified file at the beginning of HTML output. This option corresponds to the \-\-html\-prolog option of .BR genhtml . .br Default is to use no extra prolog. .PP .BR genhtml_html_epilog " =" .I filename .IP If set, include the contents of the specified file at the end of HTML output. This option corresponds to the \-\-html\-epilog option of .BR genhtml . .br Default is to use no extra epilog. .PP .BR genhtml_html_extension " =" .I extension .IP If set, use the specified string as filename extension for generated HTML files. This option corresponds to the \-\-html\-extension option of .BR genhtml . .br Default extension is "html". .PP .BR genhtml_html_gzip " =" .IR 0 | 1 .IP If set, compress all html files using gzip. This option corresponds to the \-\-html\-gzip option of .BR genhtml . .br Default extension is 0. .PP .BR genhtml_sort " =" .IR 0 | 1 .IP If non\-zero, create overview pages sorted by coverage rates when generating HTML output using .BR genhtml . .br This option can be set to 0 by using the \-\-no\-sort option of .BR genhtml . .br Default is 1. .PP .BR genhtml_function_coverage " =" .IR 0 | 1 .IP If non\-zero, include function coverage data when generating HTML output using .BR genhtml . .br This option can be set to 0 by using the \-\-no\-function\-coverage option of .BR genhtml . .br Default is 1. .PP .BR genhtml_branch_coverage " =" .IR 0 | 1 .IP If non\-zero, include branch coverage data when generating HTML output using .BR genhtml . .br This option can be set to 0 by using the \-\-no\-branch\-coverage option of .BR genhtml . .br Default is 1. .PP .BR genhtml_charset " =" .I charset .IP Specify the character set of all generated HTML pages. .br Use this option if the source code contains characters which are not part of the default character set. Note that this option is ignored when a custom HTML prolog is specified (see also .BR genhtml_html_prolog ). .br Default is UTF-8. .PP .BR genhtml_demangle_cpp " =" .IR 0 | 1 .IP If non-zero, demangle C++ function names in function overviews. Set this option to one if you want to convert C++ internal function names to human readable format for display on the HTML function overview page. This option requires that the c++filt tool is installed (see .BR c++filt(1) ). .br This option corresponds to the \-\-demangle\-cpp command line option of .BR genhtml . .br Default is 0. .PP .BR genhtml_desc_html " =" .IR 0 | 1 .IP If non-zero, test case descriptions may contain HTML markup. Set this option to one if you want to embed HTML markup (for example to include links) in test case descriptions. When set to zero, HTML markup characters will be escaped to show up as plain text on the test case description page. .br Default is 0. .PP .BR genhtml_precision " =" .IR 1 | 2 | 3 | 4 .IP Specify how many digits after the decimal-point should be used for displaying coverage rates. .br Default is 1. .PP .BR genhtml_missed " =" .IR 0 | 1 .IP If non-zero, the count of missed lines, functions, or branches is shown as negative numbers in overview pages. .br Default is 0. .PP . .BR geninfo_gcov_tool " =" .I path_to_gcov .IP Specify the location of the gcov tool (see .BR gcov (1)) which is used to generate coverage information from data files. .br Default is 'gcov'. .PP .BR geninfo_adjust_testname " =" .IR 0 | 1 .IP If non\-zero, adjust test names to include operating system information when capturing coverage data. .br Default is 0. .PP .BR geninfo_checksum " =" .IR 0 | 1 .IP If non\-zero, generate source code checksums when capturing coverage data. Checksums are useful to prevent merging coverage data from incompatible source code versions but checksum generation increases the size of coverage files and the time used to generate those files. .br This option corresponds to the \-\-checksum and \-\-no\-checksum command line option of .BR geninfo . .br Default is 0. .PP .BR geninfo_compat_libtool " =" .IR 0 | 1 .IP If non\-zero, enable libtool compatibility mode. When libtool compatibility mode is enabled, lcov will assume that the source code relating to a .da file located in a directory named ".libs" can be found in its parent directory. .br This option corresponds to the \-\-compat\-libtool and \-\-no\-compat\-libtool command line option of .BR geninfo . .br Default is 1. .PP .BR geninfo_external " =" .IR 0 | 1 .IP If non\-zero, capture coverage data for external source files. External source files are files which are not located in one of the directories (including sub-directories) specified by the \-\-directory or \-\-base\-directory options of .BR lcov / geninfo . Default is 1. .PP .BR geninfo_gcov_all_blocks " =" .IR 0 | 1 .IP If non\-zero, call the gcov tool with option --all-blocks. Using --all-blocks will produce more detailed branch coverage information for each line. Set this option to zero if you do not need detailed branch coverage information to speed up the process of capturing code coverage or to work around a bug in some versions of gcov which will cause it to endlessly loop when analysing some files. Default is 1. .PP .BR geninfo_compat " =" .IR mode = value [, mode = value ,...] .IP Specify that geninfo should enable one or more compatibility modes when capturing coverage data. This option corresponds to the \-\-compat command line option of .BR geninfo . Default is 'libtool=on, hammer=auto, split_crc=auto'. .PP .BR geninfo_adjust_src_path " =" .IR pattern " => " replacement .br .BR geninfo_adjust_src_path " =" .I pattern .IP Adjust source paths when capturing coverage data. Use this option in situations where geninfo cannot find the correct path to source code files of a project. By providing a .I pattern in Perl regular expression format (see .BR perlre (1)) and an optional replacement string, you can instruct geninfo to remove or change parts of the incorrect source path. .B Example: .br 1. When geninfo reports that it cannot find source file .br /path/to/src/.libs/file.c .br while the file is actually located in .br /path/to/src/file.c .br use the following parameter: .br geninfo_adjust_src_path = /.libs This will remove all "/.libs" strings from the path. 2. When geninfo reports that it cannot find source file .br /tmp/build/file.c .br while the file is actually located in .br /usr/src/file.c .br use the following parameter: .br geninfo_adjust_src_path = /tmp/build => /usr/src .br This will change all "/tmp/build" strings in the path to "/usr/src". .PP .BR geninfo_auto_base " =" .IR 0 | 1 .IP If non\-zero, apply a heuristic to determine the base directory when collecting coverage data. .br Use this option when using geninfo on projects built with libtool or similar build environments that work with multiple base directories, i.e. environments, where the current working directory when invoking the compiler ist not the same directory in which the source code file is located, and in addition, is different between files of the same project. .br Default is 1. .PP .BR lcov_gcov_dir " =" .I path_to_kernel_coverage_data .IP Specify the path to the directory where kernel coverage data can be found or leave undefined for auto-detection. .br Default is auto-detection. .PP .BR lcov_tmp_dir " =" .I temp .IP Specify the location of a directory used for temporary files. .br Default is '/tmp'. .PP .BR lcov_list_full_path " =" .IR 0 | 1 .IP If non-zero, print the full path to source code files during a list operation. .br This option corresponds to the \-\-list\-full\-path option of .BR lcov . .br Default is 0. .PP .BR lcov_list_max_width " =" .IR width .IP Specify the maximum width for list output. This value is ignored when lcov_list_full_path is non\-zero. .br Default is 80. .PP .BR lcov_list_truncate_max .B " =" .IR percentage .IP Specify the maximum percentage of file names which may be truncated when choosing a directory prefix in list output. This value is ignored when lcov_list_full_path is non\-zero. .br Default is 20. .PP .BR lcov_function_coverage " =" .IR 0 | 1 .IP Specify whether lcov should handle function coverage data. .br Setting this option to 0 can reduce memory and CPU time consumption when lcov is collecting and processing coverage data, as well as reduce the size of the resulting data files. Note that setting .B genhtml_function_coverage will override this option for HTML generation. .br Default is 1. .PP .BR lcov_branch_coverage " =" .IR 0 | 1 .IP Specify whether lcov should handle branch coverage data. .br Setting this option to 0 can reduce memory and CPU time consumption when lcov is collecting and processing coverage data, as well as reduce the size of the resulting data files. Note that setting .B genhtml_branch_coverage will override this option for HTML generation. .br Default is 0. .PP .BR lcov_excl_line " =" .I expression .IP Specify the regular expression of lines to exclude. .br Default is 'LCOV_EXCL_LINE'. .PP .BR lcov_excl_br_line " =" .I expression .IP Specify the regular expression of lines to exclude from branch coverage. .br Default is 'LCOV_EXCL_BR_LINE'. .PP .SH FILES .TP .I /etc/lcovrc The system\-wide .B lcov configuration file. .TP .I ~/.lcovrc The individual per\-user configuration file. .PP .SH SEE ALSO .BR lcov (1), .BR genhtml (1), .BR geninfo (1), .BR gcov (1) lcov-1.14/man/lcov.10000644000000000000000000005002713436002743012706 0ustar rootroot.TH lcov 1 "LCOV 1.14" 2019\-02\-28 "User Manuals" .SH NAME lcov \- a graphical GCOV front\-end .SH SYNOPSIS .B lcov .BR \-c | \-\-capture .RS 5 .br .RB [ \-d | \-\-directory .IR directory ] .RB [ \-k | \-\-kernel\-directory .IR directory ] .br .RB [ \-o | \-\-output\-file .IR tracefile ] .RB [ \-t | \-\-test\-name .IR testname ] .br .RB [ \-b | \-\-base\-directory .IR directory ] .RB [ \-i | \-\-initial ] .RB [ \-\-gcov\-tool .IR tool ] .br .RB [ \-\-checksum ] .RB [ \-\-no\-checksum ] .RB [ \-\-no\-recursion ] .RB [ \-f | \-\-follow ] .br .RB [ \-\-compat\-libtool ] .RB [ \-\-no\-compat\-libtool ] .RB [ \-\-ignore\-errors .IR errors ] .br .RB [ \-\-to\-package .IR package ] .RB [ \-\-from\-package .IR package ] .RB [ \-q | \-\-quiet ] .br .RB [ \-\-no\-markers ] .RB [ \-\-external ] .RB [ \-\-no\-external ] .br .RB [ \-\-config\-file .IR config\-file ] .RB [ \-\-rc .IR keyword = value ] .br .RB [ \-\-compat .IR mode =on|off|auto] .br .RB [ \-\-include .IR pattern ] .RB [ \-\-exclude .IR pattern ] .br .RE .B lcov .BR \-z | \-\-zerocounters .RS 5 .br .RB [ \-d | \-\-directory .IR directory ] .RB [ \-\-no\-recursion ] .RB [ \-f | \-\-follow ] .br .RB [ \-q | \-\-quiet ] .br .RE .B lcov .BR \-l | \-\-list .I tracefile .RS 5 .br .RB [ \-q | \-\-quiet ] .RB [ \-\-list\-full\-path ] .RB [ \-\-no\-list\-full\-path ] .br .RB [ \-\-config\-file .IR config\-file ] .RB [ \-\-rc .IR keyword = value ] .br .RE .B lcov .BR \-a | \-\-add\-tracefile .I tracefile .RS 5 .br .RB [ \-o | \-\-output\-file .IR tracefile ] .RB [ \-\-checksum ] .RB [ \-\-no\-checksum ] .br .RB [ \-q | \-\-quiet ] .RB [ \-\-config\-file .IR config\-file ] .RB [ \-\-rc .IR keyword = value ] .br .RE .B lcov .BR \-e | \-\-extract .I tracefile pattern .RS 5 .br .RB [ \-o | \-\-output\-file .IR tracefile ] .RB [ \-\-checksum ] .RB [ \-\-no\-checksum ] .br .RB [ \-q | \-\-quiet ] .RB [ \-\-config\-file .IR config\-file ] .RB [ \-\-rc .IR keyword = value ] .br .RE .B lcov .BR \-r | \-\-remove .I tracefile pattern .RS 5 .br .RB [ \-o | \-\-output\-file .IR tracefile ] .RB [ \-\-checksum ] .RB [ \-\-no\-checksum ] .br .RB [ \-q | \-\-quiet ] .RB [ \-\-config\-file .IR config\-file ] .RB [ \-\-rc .IR keyword = value ] .br .RE .B lcov .BR \-\-diff .IR "tracefile diff" .RS 5 .br .RB [ \-o | \-\-output\-file .IR tracefile ] .RB [ \-\-checksum ] .RB [ \-\-no\-checksum ] .br .RB [ \-\-convert\-filenames ] .RB [ \-\-strip .IR depth ] .RB [ \-\-path .IR path ] .RB [ \-q | \-\-quiet ] .br .RB [ \-\-config\-file .IR config\-file ] .RB [ \-\-rc .IR keyword = value ] .br .RE .B lcov .BR \-\-summary .I tracefile .RS 5 .br .RB [ \-q | \-\-quiet ] .br .RE .B lcov .RB [ \-h | \-\-help ] .RB [ \-v | \-\-version ] .RS 5 .br .RE .SH DESCRIPTION .B lcov is a graphical front\-end for GCC's coverage testing tool gcov. It collects line, function and branch coverage data for multiple source files and creates HTML pages containing the source code annotated with coverage information. It also adds overview pages for easy navigation within the file structure. Use .B lcov to collect coverage data and .B genhtml to create HTML pages. Coverage data can either be collected from the currently running Linux kernel or from a user space application. To do this, you have to complete the following preparation steps: For Linux kernel coverage: .RS Follow the setup instructions for the gcov\-kernel infrastructure: .I http://ltp.sourceforge.net/coverage/gcov.php .br .RE For user space application coverage: .RS Compile the application with GCC using the options "\-fprofile\-arcs" and "\-ftest\-coverage". .RE Please note that this man page refers to the output format of .B lcov as ".info file" or "tracefile" and that the output of GCOV is called ".da file". Also note that when printing percentages, 0% and 100% are only printed when the values are exactly 0% and 100% respectively. Other values which would conventionally be rounded to 0% or 100% are instead printed as nearest non-boundary value. This behavior is in accordance with that of the .BR gcov (1) tool. .SH OPTIONS .B \-a .I tracefile .br .B \-\-add\-tracefile .I tracefile .br .RS Add contents of .IR tracefile . Specify several tracefiles using the \-a switch to combine the coverage data contained in these files by adding up execution counts for matching test and filename combinations. The result of the add operation will be written to stdout or the tracefile specified with \-o. Only one of \-z, \-c, \-a, \-e, \-r, \-l, \-\-diff or \-\-summary may be specified at a time. .RE .B \-b .I directory .br .B \-\-base\-directory .I directory .br .RS .RI "Use " directory as base directory for relative paths. Use this option to specify the base directory of a build\-environment when lcov produces error messages like: .RS ERROR: could not read source file /home/user/project/subdir1/subdir2/subdir1/subdir2/file.c .RE In this example, use /home/user/project as base directory. This option is required when using lcov on projects built with libtool or similar build environments that work with a base directory, i.e. environments, where the current working directory when invoking the compiler is not the same directory in which the source code file is located. Note that this option will not work in environments where multiple base directories are used. In that case use configuration file setting .B geninfo_auto_base=1 (see .BR lcovrc (5)). .RE .B \-c .br .B \-\-capture .br .RS Capture coverage data. By default captures the current kernel execution counts and writes the resulting coverage data to the standard output. Use the \-\-directory option to capture counts for a user space program. The result of the capture operation will be written to stdout or the tracefile specified with \-o. Only one of \-z, \-c, \-a, \-e, \-r, \-l, \-\-diff or \-\-summary may be specified at a time. .RE .B \-\-checksum .br .B \-\-no\-checksum .br .RS Specify whether to generate checksum data when writing tracefiles. Use \-\-checksum to enable checksum generation or \-\-no\-checksum to disable it. Checksum generation is .B disabled by default. When checksum generation is enabled, a checksum will be generated for each source code line and stored along with the coverage data. This checksum will be used to prevent attempts to combine coverage data from different source code versions. If you don't work with different source code versions, disable this option to speed up coverage data processing and to reduce the size of tracefiles. .RE .B \-\-compat .IR mode = value [, mode = value ,...] .br .RS Set compatibility mode. Use \-\-compat to specify that lcov should enable one or more compatibility modes when capturing coverage data. You can provide a comma-separated list of mode=value pairs to specify the values for multiple modes. Valid .I values are: .B on .RS Enable compatibility mode. .RE .B off .RS Disable compatibility mode. .RE .B auto .RS Apply auto-detection to determine if compatibility mode is required. Note that auto-detection is not available for all compatibility modes. .RE If no value is specified, 'on' is assumed as default value. Valid .I modes are: .B libtool .RS Enable this mode if you are capturing coverage data for a project that was built using the libtool mechanism. See also \-\-compat\-libtool. The default value for this setting is 'on'. .RE .B hammer .RS Enable this mode if you are capturing coverage data for a project that was built using a version of GCC 3.3 that contains a modification (hammer patch) of later GCC versions. You can identify a modified GCC 3.3 by checking the build directory of your project for files ending in the extension '.bbg'. Unmodified versions of GCC 3.3 name these files '.bb'. The default value for this setting is 'auto'. .RE .B split_crc .RS Enable this mode if you are capturing coverage data for a project that was built using a version of GCC 4.6 that contains a modification (split function checksums) of later GCC versions. Typical error messages when running lcov on coverage data produced by such GCC versions are \'out of memory' and 'reached unexpected end of file'. The default value for this setting is 'auto' .RE .RE .B \-\-compat\-libtool .br .B \-\-no\-compat\-libtool .br .RS Specify whether to enable libtool compatibility mode. Use \-\-compat\-libtool to enable libtool compatibility mode or \-\-no\-compat\-libtool to disable it. The libtool compatibility mode is .B enabled by default. When libtool compatibility mode is enabled, lcov will assume that the source code relating to a .da file located in a directory named ".libs" can be found in its parent directory. If you have directories named ".libs" in your build environment but don't use libtool, disable this option to prevent problems when capturing coverage data. .RE .B \-\-config\-file .I config\-file .br .RS Specify a configuration file to use. When this option is specified, neither the system\-wide configuration file /etc/lcovrc, nor the per\-user configuration file ~/.lcovrc is read. This option may be useful when there is a need to run several instances of .B lcov with different configuration file options in parallel. .RE .B \-\-convert\-filenames .br .RS Convert filenames when applying diff. Use this option together with \-\-diff to rename the file names of processed data sets according to the data provided by the diff. .RE .B \-\-diff .I tracefile .I difffile .br .RS Convert coverage data in .I tracefile using source code diff file .IR difffile . Use this option if you want to merge coverage data from different source code levels of a program, e.g. when you have data taken from an older version and want to combine it with data from a more current version. .B lcov will try to map source code lines between those versions and adjust the coverage data respectively. .I difffile needs to be in unified format, i.e. it has to be created using the "\-u" option of the .B diff tool. Note that lines which are not present in the old version will not be counted as instrumented, therefore tracefiles resulting from this operation should not be interpreted individually but together with other tracefiles taken from the newer version. Also keep in mind that converted coverage data should only be used for overview purposes as the process itself introduces a loss of accuracy. The result of the diff operation will be written to stdout or the tracefile specified with \-o. Only one of \-z, \-c, \-a, \-e, \-r, \-l, \-\-diff or \-\-summary may be specified at a time. .RE .B \-d .I directory .br .B \-\-directory .I directory .br .RS Use .da files in .I directory instead of kernel. If you want to work on coverage data for a user space program, use this option to specify the location where the program was compiled (that's where the counter files ending with .da will be stored). Note that you may specify this option more than once. .RE .B \-\-exclude .I pattern .br .RS Exclude source files matching .IR pattern . Use this switch if you want to exclude coverage data for a particular set of source files matching any of the given patterns. Multiple patterns can be specified by using multiple .B --exclude command line switches. The .I patterns will be interpreted as shell wildcard patterns (note that they may need to be escaped accordingly to prevent the shell from expanding them first). Can be combined with the .B --include command line switch. If a given file matches both the include pattern and the exclude pattern, the exclude pattern will take precedence. .RE .B \-\-external .br .B \-\-no\-external .br .RS Specify whether to capture coverage data for external source files. External source files are files which are not located in one of the directories specified by \-\-directory or \-\-base\-directory. Use \-\-external to include external source files while capturing coverage data or \-\-no\-external to ignore this data. Data for external source files is .B included by default. .RE .B \-e .I tracefile .I pattern .br .B \-\-extract .I tracefile .I pattern .br .RS Extract data from .IR tracefile . Use this switch if you want to extract coverage data for only a particular set of files from a tracefile. Additional command line parameters will be interpreted as shell wildcard patterns (note that they may need to be escaped accordingly to prevent the shell from expanding them first). Every file entry in .I tracefile which matches at least one of those patterns will be extracted. The result of the extract operation will be written to stdout or the tracefile specified with \-o. Only one of \-z, \-c, \-a, \-e, \-r, \-l, \-\-diff or \-\-summary may be specified at a time. .RE .B \-f .br .B \-\-follow .br .RS Follow links when searching for .da files. .RE .B \-\-from\-package .I package .br .RS Use .da files in .I package instead of kernel or directory. Use this option if you have separate machines for build and test and want to perform the .info file creation on the build machine. See \-\-to\-package for more information. .RE .B \-\-gcov\-tool .I tool .br .RS Specify the location of the gcov tool. .RE .B \-h .br .B \-\-help .br .RS Print a short help text, then exit. .RE .B \-\-include .I pattern .br .RS Include source files matching .IR pattern . Use this switch if you want to include coverage data for only a particular set of source files matching any of the given patterns. Multiple patterns can be specified by using multiple .B --include command line switches. The .I patterns will be interpreted as shell wildcard patterns (note that they may need to be escaped accordingly to prevent the shell from expanding them first). .RE .B \-\-ignore\-errors .I errors .br .RS Specify a list of errors after which to continue processing. Use this option to specify a list of one or more classes of errors after which lcov should continue processing instead of aborting. .I errors can be a comma\-separated list of the following keywords: .B gcov: the gcov tool returned with a non\-zero return code. .B source: the source code file for a data set could not be found. .B graph: the graph file could not be found or is corrupted. .RE .B \-i .br .B \-\-initial .RS Capture initial zero coverage data. Run lcov with \-c and this option on the directories containing .bb, .bbg or .gcno files before running any test case. The result is a "baseline" coverage data file that contains zero coverage for every instrumented line. Combine this data file (using lcov \-a) with coverage data files captured after a test run to ensure that the percentage of total lines covered is correct even when not all source code files were loaded during the test. Recommended procedure when capturing data for a test case: 1. create baseline coverage data file .RS # lcov \-c \-i \-d appdir \-o app_base.info .br .RE 2. perform test .RS # appdir/test .br .RE 3. create test coverage data file .RS # lcov \-c \-d appdir \-o app_test.info .br .RE 4. combine baseline and test coverage data .RS # lcov \-a app_base.info \-a app_test.info \-o app_total.info .br .RE .RE .B \-k .I subdirectory .br .B \-\-kernel\-directory .I subdirectory .br .RS Capture kernel coverage data only from .IR subdirectory . Use this option if you don't want to get coverage data for all of the kernel, but only for specific subdirectories. This option may be specified more than once. Note that you may need to specify the full path to the kernel subdirectory depending on the version of the kernel gcov support. .RE .B \-l .I tracefile .br .B \-\-list .I tracefile .br .RS List the contents of the .IR tracefile . Only one of \-z, \-c, \-a, \-e, \-r, \-l, \-\-diff or \-\-summary may be specified at a time. .RE .B \-\-list\-full\-path .br .B \-\-no\-list\-full\-path .br .RS Specify whether to show full paths during list operation. Use \-\-list\-full\-path to show full paths during list operation or \-\-no\-list\-full\-path to show shortened paths. Paths are .B shortened by default. .RE .B \-\-no\-markers .br .RS Use this option if you want to get coverage data without regard to exclusion markers in the source code file. See .BR "geninfo " (1) for details on exclusion markers. .RE .B \-\-no\-recursion .br .RS Use this option if you want to get coverage data for the specified directory only without processing subdirectories. .RE .B \-o .I tracefile .br .B \-\-output\-file .I tracefile .br .RS Write data to .I tracefile instead of stdout. Specify "\-" as a filename to use the standard output. By convention, lcov\-generated coverage data files are called "tracefiles" and should have the filename extension ".info". .RE .B \-\-path .I path .br .RS Strip path from filenames when applying diff. Use this option together with \-\-diff to tell lcov to disregard the specified initial path component when matching between tracefile and diff filenames. .RE .B \-q .br .B \-\-quiet .br .RS Do not print progress messages. This option is implied when no output filename is specified to prevent progress messages to mess with coverage data which is also printed to the standard output. .RE .B \-\-rc .IR keyword = value .br .RS Override a configuration directive. Use this option to specify a .IR keyword = value statement which overrides the corresponding configuration statement in the lcovrc configuration file. You can specify this option more than once to override multiple configuration statements. See .BR lcovrc (5) for a list of available keywords and their meaning. .RE .B \-r .I tracefile .I pattern .br .B \-\-remove .I tracefile .I pattern .br .RS Remove data from .IR tracefile . Use this switch if you want to remove coverage data for a particular set of files from a tracefile. Additional command line parameters will be interpreted as shell wildcard patterns (note that they may need to be escaped accordingly to prevent the shell from expanding them first). Every file entry in .I tracefile which matches at least one of those patterns will be removed. The result of the remove operation will be written to stdout or the tracefile specified with \-o. Only one of \-z, \-c, \-a, \-e, \-r, \-l, \-\-diff or \-\-summary may be specified at a time. .RE .B \-\-strip .I depth .br .RS Strip path components when applying diff. Use this option together with \-\-diff to tell lcov to disregard the specified number of initial directories when matching tracefile and diff filenames. .RE .B \-\-summary .I tracefile .br .RS Show summary coverage information for the specified tracefile. Note that you may specify this option more than once. Only one of \-z, \-c, \-a, \-e, \-r, \-l, \-\-diff or \-\-summary may be specified at a time. .RE .B \-t .I testname .br .B \-\-test\-name .I testname .br .RS Specify test name to be stored in the tracefile. This name identifies a coverage data set when more than one data set is merged into a combined tracefile (see option \-a). Valid test names can consist of letters, decimal digits and the underscore character ("_"). .RE .B \-\-to\-package .I package .br .RS Store .da files for later processing. Use this option if you have separate machines for build and test and want to perform the .info file creation on the build machine. To do this, follow these steps: On the test machine: .RS .br \- run the test .br \- run lcov \-c [\-d directory] \-\-to-package .I file .br \- copy .I file to the build machine .RE .br On the build machine: .RS .br \- run lcov \-c \-\-from-package .I file [\-o and other options] .RE .br This works for both kernel and user space coverage data. Note that you might have to specify the path to the build directory using \-b with either \-\-to\-package or \-\-from-package. Note also that the package data must be converted to a .info file before recompiling the program or it will become invalid. .RE .B \-v .br .B \-\-version .br .RS Print version number, then exit. .RE .B \-z .br .B \-\-zerocounters .br .RS Reset all execution counts to zero. By default tries to reset kernel execution counts. Use the \-\-directory option to reset all counters of a user space program. Only one of \-z, \-c, \-a, \-e, \-r, \-l, \-\-diff or \-\-summary may be specified at a time. .RE .SH FILES .I /etc/lcovrc .RS The system\-wide configuration file. .RE .I ~/.lcovrc .RS The per\-user configuration file. .RE .SH AUTHOR Peter Oberparleiter .SH SEE ALSO .BR lcovrc (5), .BR genhtml (1), .BR geninfo (1), .BR genpng (1), .BR gendesc (1), .BR gcov (1) lcov-1.14/man/genpng.10000644000000000000000000000325713436002743013224 0ustar rootroot.TH genpng 1 "LCOV 1.14" 2019\-02\-28 "User Manuals" .SH NAME genpng \- Generate an overview image from a source file .SH SYNOPSIS .B genpng .RB [ \-h | \-\-help ] .RB [ \-v | \-\-version ] .RS 7 .br .RB [ \-t | \-\-tab\-size .IR tabsize ] .RB [ \-w | \-\-width .IR width ] .br .RB [ \-o | \-\-output\-filename .IR output\-filename ] .br .IR source\-file .SH DESCRIPTION .B genpng creates an overview image for a given source code file of either plain text or .gcov file format. Note that the .I GD.pm Perl module has to be installed for this script to work (it may be obtained from .IR http://www.cpan.org ). Note also that .B genpng is called from within .B genhtml so that there is usually no need to call it directly. .SH OPTIONS .B \-h .br .B \-\-help .RS Print a short help text, then exit. .RE .B \-v .br .B \-\-version .RS Print version number, then exit. .RE .BI "\-t " tab\-size .br .BI "\-\-tab\-size " tab\-size .RS Use .I tab\-size spaces in place of tab. All occurrences of tabulator signs in the source code file will be replaced by the number of spaces defined by .I tab\-size (default is 4). .RE .BI "\-w " width .br .BI "\-\-width " width .RS Set width of output image to .I width pixel. The resulting image will be exactly .I width pixel wide (default is 80). Note that source code lines which are longer than .I width will be truncated. .RE .BI "\-o " filename .br .BI "\-\-output\-filename " filename .RS Write image to .IR filename . Specify a name for the resulting image file (default is .IR source\-file .png). .RE .SH AUTHOR Peter Oberparleiter .SH SEE ALSO .BR lcov (1), .BR genhtml (1), .BR geninfo (1), .BR gendesc (1), .BR gcov (1) lcov-1.14/man/geninfo.10000644000000000000000000003450613436002743013374 0ustar rootroot.TH geninfo 1 "LCOV 1.14" 2019\-02\-28 "User Manuals" .SH NAME geninfo \- Generate tracefiles from .da files .SH SYNOPSIS .B geninfo .RB [ \-h | \-\-help ] .RB [ \-v | \-\-version ] .RB [ \-q | \-\-quiet ] .br .RS 8 .RB [ \-i | \-\-initial ] .RB [ \-t | \-\-test\-name .IR test\-name ] .br .RB [ \-o | \-\-output\-filename .IR filename ] .RB [ \-f | \-\-follow ] .br .RB [ \-b | \-\-base\-directory .IR directory ] .br .RB [ \-\-checksum ] .RB [ \-\-no\-checksum ] .br .RB [ \-\-compat\-libtool ] .RB [ \-\-no\-compat\-libtool ] .br .RB [ \-\-gcov\-tool .IR tool ] .RB [ \-\-ignore\-errors .IR errors ] .br .RB [ \-\-no\-recursion ] .I directory .RB [ \-\-external ] .RB [ \-\-no\-external ] .br .RB [ \-\-config\-file .IR config\-file ] .RB [ \-\-no\-markers ] .br .RB [ \-\-derive\-func\-data ] .RB [ \-\-compat .IR mode =on|off|auto] .br .RB [ \-\-rc .IR keyword = value ] .br .RB [ \-\-include .IR pattern ] .RB [ \-\-exclude .IR pattern ] .RE .SH DESCRIPTION .B geninfo converts all GCOV coverage data files found in .I directory into tracefiles, which the .B genhtml tool can convert to HTML output. Unless the \-\-output\-filename option is specified, .B geninfo writes its output to one file per .da file, the name of which is generated by simply appending ".info" to the respective .da file name. Note that the current user needs write access to both .I directory as well as to the original source code location. This is necessary because some temporary files have to be created there during the conversion process. Note also that .B geninfo is called from within .BR lcov , so that there is usually no need to call it directly. .B Exclusion markers To exclude specific lines of code from a tracefile, you can add exclusion markers to the source code. Additionally you can exclude specific branches from branch coverage without excluding the involved lines from line and function coverage. Exclusion markers are keywords which can for example be added in the form of a comment. See .BR lcovrc (5) how to override some of them. The following markers are recognized by geninfo: LCOV_EXCL_LINE .RS Lines containing this marker will be excluded. .br .RE LCOV_EXCL_START .RS Marks the beginning of an excluded section. The current line is part of this section. .br .RE LCOV_EXCL_STOP .RS Marks the end of an excluded section. The current line not part of this section. .RE .br LCOV_EXCL_BR_LINE .RS Lines containing this marker will be excluded from branch coverage. .br .RE LCOV_EXCL_BR_START .RS Marks the beginning of a section which is excluded from branch coverage. The current line is part of this section. .br .RE LCOV_EXCL_BR_STOP .RS Marks the end of a section which is excluded from branch coverage. The current line not part of this section. .RE .br .SH OPTIONS .B \-b .I directory .br .B \-\-base\-directory .I directory .br .RS .RI "Use " directory as base directory for relative paths. Use this option to specify the base directory of a build\-environment when geninfo produces error messages like: .RS ERROR: could not read source file /home/user/project/subdir1/subdir2/subdir1/subdir2/file.c .RE In this example, use /home/user/project as base directory. This option is required when using geninfo on projects built with libtool or similar build environments that work with a base directory, i.e. environments, where the current working directory when invoking the compiler is not the same directory in which the source code file is located. Note that this option will not work in environments where multiple base directories are used. In that case use configuration file setting .B geninfo_auto_base=1 (see .BR lcovrc (5)). .RE .B \-\-checksum .br .B \-\-no\-checksum .br .RS Specify whether to generate checksum data when writing tracefiles. Use \-\-checksum to enable checksum generation or \-\-no\-checksum to disable it. Checksum generation is .B disabled by default. When checksum generation is enabled, a checksum will be generated for each source code line and stored along with the coverage data. This checksum will be used to prevent attempts to combine coverage data from different source code versions. If you don't work with different source code versions, disable this option to speed up coverage data processing and to reduce the size of tracefiles. .RE .B \-\-compat .IR mode = value [, mode = value ,...] .br .RS Set compatibility mode. Use \-\-compat to specify that geninfo should enable one or more compatibility modes when capturing coverage data. You can provide a comma-separated list of mode=value pairs to specify the values for multiple modes. Valid .I values are: .B on .RS Enable compatibility mode. .RE .B off .RS Disable compatibility mode. .RE .B auto .RS Apply auto-detection to determine if compatibility mode is required. Note that auto-detection is not available for all compatibility modes. .RE If no value is specified, 'on' is assumed as default value. Valid .I modes are: .B libtool .RS Enable this mode if you are capturing coverage data for a project that was built using the libtool mechanism. See also \-\-compat\-libtool. The default value for this setting is 'on'. .RE .B hammer .RS Enable this mode if you are capturing coverage data for a project that was built using a version of GCC 3.3 that contains a modification (hammer patch) of later GCC versions. You can identify a modified GCC 3.3 by checking the build directory of your project for files ending in the extension '.bbg'. Unmodified versions of GCC 3.3 name these files '.bb'. The default value for this setting is 'auto'. .RE .B split_crc .RS Enable this mode if you are capturing coverage data for a project that was built using a version of GCC 4.6 that contains a modification (split function checksums) of later GCC versions. Typical error messages when running geninfo on coverage data produced by such GCC versions are \'out of memory' and 'reached unexpected end of file'. The default value for this setting is 'auto' .RE .RE .B \-\-compat\-libtool .br .B \-\-no\-compat\-libtool .br .RS Specify whether to enable libtool compatibility mode. Use \-\-compat\-libtool to enable libtool compatibility mode or \-\-no\-compat\-libtool to disable it. The libtool compatibility mode is .B enabled by default. When libtool compatibility mode is enabled, geninfo will assume that the source code relating to a .da file located in a directory named ".libs" can be found in its parent directory. If you have directories named ".libs" in your build environment but don't use libtool, disable this option to prevent problems when capturing coverage data. .RE .B \-\-config\-file .I config\-file .br .RS Specify a configuration file to use. When this option is specified, neither the system\-wide configuration file /etc/lcovrc, nor the per\-user configuration file ~/.lcovrc is read. This option may be useful when there is a need to run several instances of .B geninfo with different configuration file options in parallel. .RE .B \-\-derive\-func\-data .br .RS Calculate function coverage data from line coverage data. Use this option to collect function coverage data, even if the version of the gcov tool installed on the test system does not provide this data. lcov will instead derive function coverage data from line coverage data and information about which lines belong to a function. .RE .B \-\-exclude .I pattern .br .RS Exclude source files matching .IR pattern . Use this switch if you want to exclude coverage data for a particular set of source files matching any of the given patterns. Multiple patterns can be specified by using multiple .B --exclude command line switches. The .I patterns will be interpreted as shell wildcard patterns (note that they may need to be escaped accordingly to prevent the shell from expanding them first). Can be combined with the .B --include command line switch. If a given file matches both the include pattern and the exclude pattern, the exclude pattern will take precedence. .RE .B \-\-external .br .B \-\-no\-external .br .RS Specify whether to capture coverage data for external source files. External source files are files which are not located in one of the directories specified by \-\-directory or \-\-base\-directory. Use \-\-external to include external source files while capturing coverage data or \-\-no\-external to ignore this data. Data for external source files is .B included by default. .RE .B \-f .br .B \-\-follow .RS Follow links when searching .da files. .RE .B \-\-gcov\-tool .I tool .br .RS Specify the location of the gcov tool. .RE .B \-h .br .B \-\-help .RS Print a short help text, then exit. .RE .B \-\-include .I pattern .br .RS Include source files matching .IR pattern . Use this switch if you want to include coverage data for only a particular set of source files matching any of the given patterns. Multiple patterns can be specified by using multiple .B --include command line switches. The .I patterns will be interpreted as shell wildcard patterns (note that they may need to be escaped accordingly to prevent the shell from expanding them first). .RE .B \-\-ignore\-errors .I errors .br .RS Specify a list of errors after which to continue processing. Use this option to specify a list of one or more classes of errors after which geninfo should continue processing instead of aborting. .I errors can be a comma\-separated list of the following keywords: .B gcov: the gcov tool returned with a non\-zero return code. .B source: the source code file for a data set could not be found. .RE .B \-i .br .B \-\-initial .RS Capture initial zero coverage data. Run geninfo with this option on the directories containing .bb, .bbg or .gcno files before running any test case. The result is a "baseline" coverage data file that contains zero coverage for every instrumented line and function. Combine this data file (using lcov \-a) with coverage data files captured after a test run to ensure that the percentage of total lines covered is correct even when not all object code files were loaded during the test. Note: currently, the \-\-initial option does not generate branch coverage information. .RE .B \-\-no\-markers .br .RS Use this option if you want to get coverage data without regard to exclusion markers in the source code file. .RE .B \-\-no\-recursion .br .RS Use this option if you want to get coverage data for the specified directory only without processing subdirectories. .RE .BI "\-o " output\-filename .br .BI "\-\-output\-filename " output\-filename .RS Write all data to .IR output\-filename . If you want to have all data written to a single file (for easier handling), use this option to specify the respective filename. By default, one tracefile will be created for each processed .da file. .RE .B \-q .br .B \-\-quiet .RS Do not print progress messages. Suppresses all informational progress output. When this switch is enabled, only error or warning messages are printed. .RE .B \-\-rc .IR keyword = value .br .RS Override a configuration directive. Use this option to specify a .IR keyword = value statement which overrides the corresponding configuration statement in the lcovrc configuration file. You can specify this option more than once to override multiple configuration statements. See .BR lcovrc (5) for a list of available keywords and their meaning. .RE .BI "\-t " testname .br .BI "\-\-test\-name " testname .RS Use test case name .I testname for resulting data. Valid test case names can consist of letters, decimal digits and the underscore character ('_'). This proves useful when data from several test cases is merged (i.e. by simply concatenating the respective tracefiles) in which case a test name can be used to differentiate between data from each test case. .RE .B \-v .br .B \-\-version .RS Print version number, then exit. .RE .SH FILES .I /etc/lcovrc .RS The system\-wide configuration file. .RE .I ~/.lcovrc .RS The per\-user configuration file. .RE Following is a quick description of the tracefile format as used by .BR genhtml ", " geninfo " and " lcov . A tracefile is made up of several human\-readable lines of text, divided into sections. If available, a tracefile begins with the .I testname which is stored in the following format: TN: For each source file referenced in the .da file, there is a section containing filename and coverage data: SF: Following is a list of line numbers for each function name found in the source file: FN:, Next, there is a list of execution counts for each instrumented function: FNDA:, This list is followed by two lines containing the number of functions found and hit: FNF: FNH: Branch coverage information is stored which one line per branch: BRDA:,,, Block number and branch number are gcc internal IDs for the branch. Taken is either '-' if the basic block containing the branch was never executed or a number indicating how often that branch was taken. Branch coverage summaries are stored in two lines: BRF: BRH: Then there is a list of execution counts for each instrumented line (i.e. a line which resulted in executable code): DA:,[,] Note that there may be an optional checksum present for each instrumented line. The current .B geninfo implementation uses an MD5 hash as checksumming algorithm. At the end of a section, there is a summary about how many lines were found and how many were actually instrumented: LH: LF: Each sections ends with: end_of_record In addition to the main source code file there are sections for all #included files which also contain executable code. Note that the absolute path of a source file is generated by interpreting the contents of the respective .bb file (see .BR "gcov " (1) for more information on this file type). Relative filenames are prefixed with the directory in which the .bb file is found. Note also that symbolic links to the .bb file will be resolved so that the actual file path is used instead of the path to a link. This approach is necessary for the mechanism to work with the /proc/gcov files. .SH AUTHOR Peter Oberparleiter .SH SEE ALSO .BR lcov (1), .BR lcovrc (5), .BR genhtml (1), .BR genpng (1), .BR gendesc (1), .BR gcov (1) lcov-1.14/man/genhtml.10000644000000000000000000003474513436002743013412 0ustar rootroot.TH genhtml 1 "LCOV 1.14" 2019\-02\-28 "User Manuals" .SH NAME genhtml \- Generate HTML view from LCOV coverage data files .SH SYNOPSIS .B genhtml .RB [ \-h | \-\-help ] .RB [ \-v | \-\-version ] .RS 8 .br .RB [ \-q | \-\-quiet ] .RB [ \-s | \-\-show\-details ] .RB [ \-f | \-\-frames ] .br .RB [ \-b | \-\-baseline\-file ] .IR baseline\-file .br .RB [ \-o | \-\-output\-directory .IR output\-directory ] .br .RB [ \-t | \-\-title .IR title ] .br .RB [ \-d | \-\-description\-file .IR description\-file ] .br .RB [ \-k | \-\-keep\-descriptions ] .RB [ \-c | \-\-css\-file .IR css\-file ] .br .RB [ \-p | \-\-prefix .IR prefix ] .RB [ \-\-no\-prefix ] .br .RB [ \-\-no\-source ] .RB [ \-\-num\-spaces .IR num ] .RB [ \-\-highlight ] .br .RB [ \-\-legend ] .RB [ \-\-html\-prolog .IR prolog\-file ] .br .RB [ \-\-html\-epilog .IR epilog\-file ] .RB [ \-\-html\-extension .IR extension ] .br .RB [ \-\-html\-gzip ] .RB [ \-\-sort ] .RB [ \-\-no\-sort ] .br .RB [ \-\-function\-coverage ] .RB [ \-\-no\-function\-coverage ] .br .RB [ \-\-branch\-coverage ] .RB [ \-\-no\-branch\-coverage ] .br .RB [ \-\-demangle\-cpp ] .RB [ \-\-ignore\-errors .IR errors ] .br .RB [ \-\-config\-file .IR config\-file ] .RB [ \-\-rc .IR keyword = value ] .br .RB [ \-\-precision .RB [ \-\-missed ] .br .IR tracefile(s) .RE .SH DESCRIPTION Create an HTML view of coverage data found in .IR tracefile . Note that .I tracefile may also be a list of filenames. HTML output files are created in the current working directory unless the \-\-output\-directory option is used. If .I tracefile ends with ".gz", it is assumed to be GZIP\-compressed and the gunzip tool will be used to decompress it transparently. Note that all source code files have to be present and readable at the exact file system location they were compiled. Use option .I \--css\-file to modify layout and colors of the generated HTML output. Files are marked in different colors depending on the associated coverage rate. By default, the coverage limits for low, medium and high coverage are set to 0\-75%, 75\-90% and 90\-100% percent respectively. To change these values, use configuration file options .IR genhtml_hi_limit " and " genhtml_med_limit . Also note that when displaying percentages, 0% and 100% are only printed when the values are exactly 0% and 100% respectively. Other values which would conventionally be rounded to 0% or 100% are instead printed as nearest non-boundary value. This behavior is in accordance with that of the .BR gcov (1) tool. .SH OPTIONS .B \-h .br .B \-\-help .RS Print a short help text, then exit. .RE .B \-v .br .B \-\-version .RS Print version number, then exit. .RE .B \-q .br .B \-\-quiet .RS Do not print progress messages. Suppresses all informational progress output. When this switch is enabled, only error or warning messages are printed. .RE .B \-f .br .B \-\-frames .RS Use HTML frames for source code view. If enabled, a frameset is created for each source code file, providing an overview of the source code as a "clickable" image. Note that this option will slow down output creation noticeably because each source code character has to be inspected once. Note also that the GD.pm Perl module has to be installed for this option to work (it may be obtained from http://www.cpan.org). .RE .B \-s .br .B \-\-show\-details .RS Generate detailed directory view. When this option is enabled, .B genhtml generates two versions of each file view. One containing the standard information plus a link to a "detailed" version. The latter additionally contains information about which test case covered how many lines of each source file. .RE .BI "\-b " baseline\-file .br .BI "\-\-baseline\-file " baseline\-file .RS Use data in .I baseline\-file as coverage baseline. The tracefile specified by .I baseline\-file is read and all counts found in the original .I tracefile are decremented by the corresponding counts in .I baseline\-file before creating any output. Note that when a count for a particular line in .I baseline\-file is greater than the count in the .IR tracefile , the result is zero. .RE .BI "\-o " output\-directory .br .BI "\-\-output\-directory " output\-directory .RS Create files in .I output\-directory. Use this option to tell .B genhtml to write the resulting files to a directory other than the current one. If .I output\-directory does not exist, it will be created. It is advisable to use this option since depending on the project size, a lot of files and subdirectories may be created. .RE .BI "\-t " title .br .BI "\-\-title " title .RS Display .I title in header of all pages. .I title is written to the header portion of each generated HTML page to identify the context in which a particular output was created. By default this is the name of the tracefile. .RE .BI "\-d " description\-file .br .BI "\-\-description\-file " description\-file .RS Read test case descriptions from .IR description\-file . All test case descriptions found in .I description\-file and referenced in the input data file are read and written to an extra page which is then incorporated into the HTML output. The file format of .IR "description\-file " is: for each test case: .RS TN: .br TD: .RE Valid test case names can consist of letters, numbers and the underscore character ('_'). .RE .B \-k .br .B \-\-keep\-descriptions .RS Do not remove unused test descriptions. Keep descriptions found in the description file even if the coverage data indicates that the associated test case did not cover any lines of code. This option can also be configured permanently using the configuration file option .IR genhtml_keep_descriptions . .RE .BI "\-c " css\-file .br .BI "\-\-css\-file " css\-file .RS Use external style sheet file .IR css\-file . Using this option, an extra .css file may be specified which will replace the default one. This may be helpful if the default colors make your eyes want to jump out of their sockets :) This option can also be configured permanently using the configuration file option .IR genhtml_css_file . .RE .BI "\-p " prefix .br .BI "\-\-prefix " prefix .RS Remove .I prefix from all directory names. Because lists containing long filenames are difficult to read, there is a mechanism implemented that will automatically try to shorten all directory names on the overview page beginning with a common prefix. By default, this is done using an algorithm that tries to find the prefix which, when applied, will minimize the resulting sum of characters of all directory names. Use this option to specify the prefix to be removed by yourself. .RE .B \-\-no\-prefix .RS Do not remove prefix from directory names. This switch will completely disable the prefix mechanism described in the previous section. This option can also be configured permanently using the configuration file option .IR genhtml_no_prefix . .RE .B \-\-no\-source .RS Do not create source code view. Use this switch if you don't want to get a source code view for each file. This option can also be configured permanently using the configuration file option .IR genhtml_no_source . .RE .BI "\-\-num\-spaces " spaces .RS Replace tabs in source view with .I num spaces. Default value is 8. This option can also be configured permanently using the configuration file option .IR genhtml_num_spaces . .RE .B \-\-highlight .RS Highlight lines with converted\-only coverage data. Use this option in conjunction with the \-\-diff option of .B lcov to highlight those lines which were only covered in data sets which were converted from previous source code versions. This option can also be configured permanently using the configuration file option .IR genhtml_highlight . .RE .B \-\-legend .RS Include color legend in HTML output. Use this option to include a legend explaining the meaning of color coding in the resulting HTML output. This option can also be configured permanently using the configuration file option .IR genhtml_legend . .RE .BI "\-\-html\-prolog " prolog\-file .RS Read customized HTML prolog from .IR prolog\-file . Use this option to replace the default HTML prolog (the initial part of the HTML source code leading up to and including the tag) with the contents of .IR prolog\-file . Within the prolog text, the following words will be replaced when a page is generated: .B "@pagetitle@" .br The title of the page. .B "@basedir@" .br A relative path leading to the base directory (e.g. for locating css\-files). This option can also be configured permanently using the configuration file option .IR genhtml_html_prolog . .RE .BI "\-\-html\-epilog " epilog\-file .RS Read customized HTML epilog from .IR epilog\-file . Use this option to replace the default HTML epilog (the final part of the HTML source including ) with the contents of .IR epilog\-file . Within the epilog text, the following words will be replaced when a page is generated: .B "@basedir@" .br A relative path leading to the base directory (e.g. for locating css\-files). This option can also be configured permanently using the configuration file option .IR genhtml_html_epilog . .RE .BI "\-\-html\-extension " extension .RS Use customized filename extension for generated HTML pages. This option is useful in situations where different filename extensions are required to render the resulting pages correctly (e.g. php). Note that a '.' will be inserted between the filename and the extension specified by this option. This option can also be configured permanently using the configuration file option .IR genhtml_html_extension . .RE .B \-\-html\-gzip .RS Compress all generated html files with gzip and add a .htaccess file specifying gzip\-encoding in the root output directory. Use this option if you want to save space on your webserver. Requires a webserver with .htaccess support and a browser with support for gzip compressed html. This option can also be configured permanently using the configuration file option .IR genhtml_html_gzip . .RE .B \-\-sort .br .B \-\-no\-sort .RS Specify whether to include sorted views of file and directory overviews. Use \-\-sort to include sorted views or \-\-no\-sort to not include them. Sorted views are .B enabled by default. When sorted views are enabled, each overview page will contain links to views of that page sorted by coverage rate. This option can also be configured permanently using the configuration file option .IR genhtml_sort . .RE .B \-\-function\-coverage .br .B \-\-no\-function\-coverage .RS Specify whether to display function coverage summaries in HTML output. Use \-\-function\-coverage to enable function coverage summaries or \-\-no\-function\-coverage to disable it. Function coverage summaries are .B enabled by default When function coverage summaries are enabled, each overview page will contain the number of functions found and hit per file or directory, together with the resulting coverage rate. In addition, each source code view will contain a link to a page which lists all functions found in that file plus the respective call count for those functions. This option can also be configured permanently using the configuration file option .IR genhtml_function_coverage . .RE .B \-\-branch\-coverage .br .B \-\-no\-branch\-coverage .RS Specify whether to display branch coverage data in HTML output. Use \-\-branch\-coverage to enable branch coverage display or \-\-no\-branch\-coverage to disable it. Branch coverage data display is .B enabled by default When branch coverage display is enabled, each overview page will contain the number of branches found and hit per file or directory, together with the resulting coverage rate. In addition, each source code view will contain an extra column which lists all branches of a line with indications of whether the branch was taken or not. Branches are shown in the following format: ' + ': Branch was taken at least once .br ' - ': Branch was not taken .br ' # ': The basic block containing the branch was never executed .br Note that it might not always be possible to relate branches to the corresponding source code statements: during compilation, GCC might shuffle branches around or eliminate some of them to generate better code. This option can also be configured permanently using the configuration file option .IR genhtml_branch_coverage . .RE .B \-\-demangle\-cpp .RS Specify whether to demangle C++ function names. Use this option if you want to convert C++ internal function names to human readable format for display on the HTML function overview page. This option requires that the c++filt tool is installed (see .BR c++filt (1)). .RE .B \-\-ignore\-errors .I errors .br .RS Specify a list of errors after which to continue processing. Use this option to specify a list of one or more classes of errors after which geninfo should continue processing instead of aborting. .I errors can be a comma\-separated list of the following keywords: .B source: the source code file for a data set could not be found. .RE .B \-\-config\-file .I config\-file .br .RS Specify a configuration file to use. When this option is specified, neither the system\-wide configuration file /etc/lcovrc, nor the per\-user configuration file ~/.lcovrc is read. This option may be useful when there is a need to run several instances of .B genhtml with different configuration file options in parallel. .RE .B \-\-rc .IR keyword = value .br .RS Override a configuration directive. Use this option to specify a .IR keyword = value statement which overrides the corresponding configuration statement in the lcovrc configuration file. You can specify this option more than once to override multiple configuration statements. See .BR lcovrc (5) for a list of available keywords and their meaning. .RE .BI "\-\-precision " num .RS Show coverage rates with .I num number of digits after the decimal-point. Default value is 1. This option can also be configured permanently using the configuration file option .IR genhtml_precision . .RE .B \-\-missed .RS Show counts of missed lines, functions, or branches Use this option to change overview pages to show the count of lines, functions, or branches that were not hit. These counts are represented by negative numbers. When specified together with \-\-sort, file and directory views will be sorted by missed counts. This option can also be configured permanently using the configuration file option .IR genhtml_missed . .RE .SH FILES .I /etc/lcovrc .RS The system\-wide configuration file. .RE .I ~/.lcovrc .RS The per\-user configuration file. .RE .SH AUTHOR Peter Oberparleiter .SH SEE ALSO .BR lcov (1), .BR lcovrc (5), .BR geninfo (1), .BR genpng (1), .BR gendesc (1), .BR gcov (1) lcov-1.14/man/gendesc.10000644000000000000000000000254713436002743013357 0ustar rootroot.TH gendesc 1 "LCOV 1.14" 2019\-02\-28 "User Manuals" .SH NAME gendesc \- Generate a test case description file .SH SYNOPSIS .B gendesc .RB [ \-h | \-\-help ] .RB [ \-v | \-\-version ] .RS 8 .br .RB [ \-o | \-\-output\-filename .IR filename ] .br .I inputfile .SH DESCRIPTION Convert plain text test case descriptions into a format as understood by .BR genhtml . .I inputfile needs to observe the following format: For each test case: .IP " \-" one line containing the test case name beginning at the start of the line .RE .IP " \-" one or more lines containing the test case description indented with at least one whitespace character (tab or space) .RE .B Example input file: test01 .RS An example test case description. .br Description continued .RE test42 .RS Supposedly the answer to most of your questions .RE Note: valid test names can consist of letters, decimal digits and the underscore character ('_'). .SH OPTIONS .B \-h .br .B \-\-help .RS Print a short help text, then exit. .RE .B \-v .br .B \-\-version .RS Print version number, then exit. .RE .BI "\-o " filename .br .BI "\-\-output\-filename " filename .RS Write description data to .IR filename . By default, output is written to STDOUT. .RE .SH AUTHOR Peter Oberparleiter .SH SEE ALSO .BR lcov (1), .BR genhtml (1), .BR geninfo (1), .BR genpng (1), .BR gcov (1) lcov-1.14/Makefile0000644000000000000000000001237213435775412012560 0ustar rootroot# # Makefile for LCOV # # Make targets: # - install: install LCOV tools and man pages on the system # - uninstall: remove tools and man pages from the system # - dist: create files required for distribution, i.e. the lcov.tar.gz # and the lcov.rpm file. Just make sure to adjust the VERSION # and RELEASE variables below - both version and date strings # will be updated in all necessary files. # - clean: remove all generated files # VERSION := $(shell bin/get_version.sh --version) RELEASE := $(shell bin/get_version.sh --release) FULL := $(shell bin/get_version.sh --full) # Set this variable during 'make install' to specify the Perl interpreter used in # installed scripts, or leave empty to keep the current interpreter. export LCOV_PERL_PATH := /usr/bin/perl PREFIX := /usr/local CFG_DIR := $(PREFIX)/etc BIN_DIR := $(PREFIX)/bin MAN_DIR := $(PREFIX)/share/man TMP_DIR := $(shell mktemp -d) FILES := $(wildcard bin/*) $(wildcard man/*) README Makefile \ $(wildcard rpm/*) lcovrc .PHONY: all info clean install uninstall rpms test all: info info: @echo "Available make targets:" @echo " install : install binaries and man pages in DESTDIR (default /)" @echo " uninstall : delete binaries and man pages from DESTDIR (default /)" @echo " dist : create packages (RPM, tarball) ready for distribution" @echo " test : perform self-tests" clean: rm -f lcov-*.tar.gz rm -f lcov-*.rpm make -C example clean make -C test -s clean install: bin/install.sh bin/lcov $(DESTDIR)$(BIN_DIR)/lcov -m 755 bin/install.sh bin/genhtml $(DESTDIR)$(BIN_DIR)/genhtml -m 755 bin/install.sh bin/geninfo $(DESTDIR)$(BIN_DIR)/geninfo -m 755 bin/install.sh bin/genpng $(DESTDIR)$(BIN_DIR)/genpng -m 755 bin/install.sh bin/gendesc $(DESTDIR)$(BIN_DIR)/gendesc -m 755 bin/install.sh man/lcov.1 $(DESTDIR)$(MAN_DIR)/man1/lcov.1 -m 644 bin/install.sh man/genhtml.1 $(DESTDIR)$(MAN_DIR)/man1/genhtml.1 -m 644 bin/install.sh man/geninfo.1 $(DESTDIR)$(MAN_DIR)/man1/geninfo.1 -m 644 bin/install.sh man/genpng.1 $(DESTDIR)$(MAN_DIR)/man1/genpng.1 -m 644 bin/install.sh man/gendesc.1 $(DESTDIR)$(MAN_DIR)/man1/gendesc.1 -m 644 bin/install.sh man/lcovrc.5 $(DESTDIR)$(MAN_DIR)/man5/lcovrc.5 -m 644 bin/install.sh lcovrc $(DESTDIR)$(CFG_DIR)/lcovrc -m 644 bin/updateversion.pl $(DESTDIR)$(BIN_DIR)/lcov $(VERSION) $(RELEASE) $(FULL) bin/updateversion.pl $(DESTDIR)$(BIN_DIR)/genhtml $(VERSION) $(RELEASE) $(FULL) bin/updateversion.pl $(DESTDIR)$(BIN_DIR)/geninfo $(VERSION) $(RELEASE) $(FULL) bin/updateversion.pl $(DESTDIR)$(BIN_DIR)/genpng $(VERSION) $(RELEASE) $(FULL) bin/updateversion.pl $(DESTDIR)$(BIN_DIR)/gendesc $(VERSION) $(RELEASE) $(FULL) bin/updateversion.pl $(DESTDIR)$(MAN_DIR)/man1/lcov.1 $(VERSION) $(RELEASE) $(FULL) bin/updateversion.pl $(DESTDIR)$(MAN_DIR)/man1/genhtml.1 $(VERSION) $(RELEASE) $(FULL) bin/updateversion.pl $(DESTDIR)$(MAN_DIR)/man1/geninfo.1 $(VERSION) $(RELEASE) $(FULL) bin/updateversion.pl $(DESTDIR)$(MAN_DIR)/man1/genpng.1 $(VERSION) $(RELEASE) $(FULL) bin/updateversion.pl $(DESTDIR)$(MAN_DIR)/man1/gendesc.1 $(VERSION) $(RELEASE) $(FULL) bin/updateversion.pl $(DESTDIR)$(MAN_DIR)/man5/lcovrc.5 $(VERSION) $(RELEASE) $(FULL) uninstall: bin/install.sh --uninstall bin/lcov $(DESTDIR)$(BIN_DIR)/lcov bin/install.sh --uninstall bin/genhtml $(DESTDIR)$(BIN_DIR)/genhtml bin/install.sh --uninstall bin/geninfo $(DESTDIR)$(BIN_DIR)/geninfo bin/install.sh --uninstall bin/genpng $(DESTDIR)$(BIN_DIR)/genpng bin/install.sh --uninstall bin/gendesc $(DESTDIR)$(BIN_DIR)/gendesc bin/install.sh --uninstall man/lcov.1 $(DESTDIR)$(MAN_DIR)/man1/lcov.1 bin/install.sh --uninstall man/genhtml.1 $(DESTDIR)$(MAN_DIR)/man1/genhtml.1 bin/install.sh --uninstall man/geninfo.1 $(DESTDIR)$(MAN_DIR)/man1/geninfo.1 bin/install.sh --uninstall man/genpng.1 $(DESTDIR)$(MAN_DIR)/man1/genpng.1 bin/install.sh --uninstall man/gendesc.1 $(DESTDIR)$(MAN_DIR)/man1/gendesc.1 bin/install.sh --uninstall man/lcovrc.5 $(DESTDIR)$(MAN_DIR)/man5/lcovrc.5 bin/install.sh --uninstall lcovrc $(DESTDIR)$(CFG_DIR)/lcovrc dist: lcov-$(VERSION).tar.gz lcov-$(VERSION)-$(RELEASE).noarch.rpm \ lcov-$(VERSION)-$(RELEASE).src.rpm lcov-$(VERSION).tar.gz: $(FILES) mkdir $(TMP_DIR)/lcov-$(VERSION) cp -r * $(TMP_DIR)/lcov-$(VERSION) bin/copy_dates.sh . $(TMP_DIR)/lcov-$(VERSION) make -C $(TMP_DIR)/lcov-$(VERSION) clean bin/updateversion.pl $(TMP_DIR)/lcov-$(VERSION) $(VERSION) $(RELEASE) $(FULL) bin/get_changes.sh > $(TMP_DIR)/lcov-$(VERSION)/CHANGES cd $(TMP_DIR) ; \ tar cfz $(TMP_DIR)/lcov-$(VERSION).tar.gz lcov-$(VERSION) mv $(TMP_DIR)/lcov-$(VERSION).tar.gz . rm -rf $(TMP_DIR) lcov-$(VERSION)-$(RELEASE).noarch.rpm: rpms lcov-$(VERSION)-$(RELEASE).src.rpm: rpms rpms: lcov-$(VERSION).tar.gz mkdir $(TMP_DIR) mkdir $(TMP_DIR)/BUILD mkdir $(TMP_DIR)/RPMS mkdir $(TMP_DIR)/SOURCES mkdir $(TMP_DIR)/SRPMS cp lcov-$(VERSION).tar.gz $(TMP_DIR)/SOURCES cd $(TMP_DIR)/BUILD ; \ tar xfz $(TMP_DIR)/SOURCES/lcov-$(VERSION).tar.gz \ lcov-$(VERSION)/rpm/lcov.spec rpmbuild --define '_topdir $(TMP_DIR)' \ -ba $(TMP_DIR)/BUILD/lcov-$(VERSION)/rpm/lcov.spec mv $(TMP_DIR)/RPMS/noarch/lcov-$(VERSION)-$(RELEASE).noarch.rpm . mv $(TMP_DIR)/SRPMS/lcov-$(VERSION)-$(RELEASE).src.rpm . rm -rf $(TMP_DIR) test: @make -C test -s all lcov-1.14/lcovrc0000644000000000000000000001246513234076471012332 0ustar rootroot# # /etc/lcovrc - system-wide defaults for LCOV # # To change settings for a single user, place a customized copy of this file # at location ~/.lcovrc # # Specify an external style sheet file (same as --css-file option of genhtml) #genhtml_css_file = gcov.css # Specify coverage rate limits (in %) for classifying file entries # HI: hi_limit <= rate <= 100 graph color: green # MED: med_limit <= rate < hi_limit graph color: orange # LO: 0 <= rate < med_limit graph color: red genhtml_hi_limit = 90 genhtml_med_limit = 75 # Width of line coverage field in source code view genhtml_line_field_width = 12 # Width of branch coverage field in source code view genhtml_branch_field_width = 16 # Width of overview image (used by --frames option of genhtml) genhtml_overview_width = 80 # Resolution of overview navigation: this number specifies the maximum # difference in lines between the position a user selected from the overview # and the position the source code window is scrolled to (used by --frames # option of genhtml) genhtml_nav_resolution = 4 # Clicking a line in the overview image should show the source code view at # a position a bit further up so that the requested line is not the first # line in the window. This number specifies that offset in lines (used by # --frames option of genhtml) genhtml_nav_offset = 10 # Do not remove unused test descriptions if non-zero (same as # --keep-descriptions option of genhtml) genhtml_keep_descriptions = 0 # Do not remove prefix from directory names if non-zero (same as --no-prefix # option of genhtml) genhtml_no_prefix = 0 # Do not create source code view if non-zero (same as --no-source option of # genhtml) genhtml_no_source = 0 # Replace tabs with number of spaces in source view (same as --num-spaces # option of genhtml) genhtml_num_spaces = 8 # Highlight lines with converted-only data if non-zero (same as --highlight # option of genhtml) genhtml_highlight = 0 # Include color legend in HTML output if non-zero (same as --legend option of # genhtml) genhtml_legend = 0 # Use FILE as HTML prolog for generated pages (same as --html-prolog option of # genhtml) #genhtml_html_prolog = FILE # Use FILE as HTML epilog for generated pages (same as --html-epilog option of # genhtml) #genhtml_html_epilog = FILE # Use custom filename extension for pages (same as --html-extension option of # genhtml) #genhtml_html_extension = html # Compress all generated html files with gzip. #genhtml_html_gzip = 1 # Include sorted overview pages (can be disabled by the --no-sort option of # genhtml) genhtml_sort = 1 # Include function coverage data display (can be disabled by the # --no-func-coverage option of genhtml) #genhtml_function_coverage = 1 # Include branch coverage data display (can be disabled by the # --no-branch-coverage option of genhtml) #genhtml_branch_coverage = 1 # Specify the character set of all generated HTML pages genhtml_charset=UTF-8 # Allow HTML markup in test case description text if non-zero genhtml_desc_html=0 # Specify the precision for coverage rates #genhtml_precision=1 # Show missed counts instead of hit counts #genhtml_missed=1 # Demangle C++ symbols #genhtml_demangle_cpp=1 # Location of the gcov tool (same as --gcov-info option of geninfo) #geninfo_gcov_tool = gcov # Adjust test names to include operating system information if non-zero #geninfo_adjust_testname = 0 # Calculate checksum for each source code line if non-zero (same as --checksum # option of geninfo if non-zero, same as --no-checksum if zero) #geninfo_checksum = 1 # Specify whether to capture coverage data for external source files (can # be overridden by the --external and --no-external options of geninfo/lcov) #geninfo_external = 1 # Enable libtool compatibility mode if non-zero (same as --compat-libtool option # of geninfo if non-zero, same as --no-compat-libtool if zero) #geninfo_compat_libtool = 0 # Use gcov's --all-blocks option if non-zero #geninfo_gcov_all_blocks = 1 # Specify compatiblity modes (same as --compat option of geninfo). #geninfo_compat = libtool=on, hammer=auto, split_crc=auto # Adjust path to source files by removing or changing path components that # match the specified pattern (Perl regular expression format) #geninfo_adjust_src_path = /tmp/build => /usr/src # Specify if geninfo should try to automatically determine the base-directory # when collecting coverage data. geninfo_auto_base = 1 # Directory containing gcov kernel files # lcov_gcov_dir = /proc/gcov # Location of the insmod tool lcov_insmod_tool = /sbin/insmod # Location of the modprobe tool lcov_modprobe_tool = /sbin/modprobe # Location of the rmmod tool lcov_rmmod_tool = /sbin/rmmod # Location for temporary directories lcov_tmp_dir = /tmp # Show full paths during list operation if non-zero (same as --list-full-path # option of lcov) lcov_list_full_path = 0 # Specify the maximum width for list output. This value is ignored when # lcov_list_full_path is non-zero. lcov_list_width = 80 # Specify the maximum percentage of file names which may be truncated when # choosing a directory prefix in list output. This value is ignored when # lcov_list_full_path is non-zero. lcov_list_truncate_max = 20 # Specify if function coverage data should be collected and processed. lcov_function_coverage = 1 # Specify if branch coverage data should be collected and processed. lcov_branch_coverage = 0 lcov-1.14/example/0000755000000000000000000000000013436014706012537 5ustar rootrootlcov-1.14/example/descriptions.txt0000644000000000000000000000047412345570444016017 0ustar rootroottest_noargs Example program is called without arguments so that default range [0..9] is used. test_2_to_2000 Example program is called with "2" and "2000" as arguments. test_overflow Example program is called with "0" and "100000" as arguments. The resulting sum is too large to be stored as an int variable. lcov-1.14/example/gauss.h0000644000000000000000000000015412345570444014036 0ustar rootroot#ifndef GAUSS_H #define GAUSS_H GAUSS_h extern int gauss_get_sum (int min, int max); #endif /* GAUSS_H */ lcov-1.14/example/methods/0000755000000000000000000000000013436014706014202 5ustar rootrootlcov-1.14/example/methods/gauss.c0000644000000000000000000000240512345570444015475 0ustar rootroot/* * methods/gauss.c * * Calculate the sum of a given range of integer numbers. * * Somewhat of a more subtle way of calculation - and it even has a story * behind it: * * Supposedly during math classes in elementary school, the teacher of * young mathematician Gauss gave the class an assignment to calculate the * sum of all natural numbers between 1 and 100, hoping that this task would * keep the kids occupied for some time. The story goes that Gauss had the * result ready after only a few minutes. What he had written on his black * board was something like this: * * 1 + 100 = 101 * 2 + 99 = 101 * 3 + 98 = 101 * . * . * 100 + 1 = 101 * * s = (1/2) * 100 * 101 = 5050 * * A more general form of this formula would be * * s = (1/2) * (max + min) * (max - min + 1) * * which is used in the piece of code below to implement the requested * function in constant time, i.e. without dependencies on the size of the * input parameters. * */ #include "gauss.h" int gauss_get_sum (int min, int max) { /* This algorithm doesn't work well with invalid range specifications so we're intercepting them here. */ if (max < min) { return 0; } return (int) ((max + min) * (double) (max - min + 1) / 2); } lcov-1.14/example/methods/iterate.c0000644000000000000000000000174412345570444016015 0ustar rootroot/* * methods/iterate.c * * Calculate the sum of a given range of integer numbers. * * This particular method of implementation works by way of brute force, * i.e. it iterates over the entire range while adding the numbers to finally * get the total sum. As a positive side effect, we're able to easily detect * overflows, i.e. situations in which the sum would exceed the capacity * of an integer variable. * */ #include #include #include "iterate.h" int iterate_get_sum (int min, int max) { int i, total; total = 0; /* This is where we loop over each number in the range, including both the minimum and the maximum number. */ for (i = min; i <= max; i++) { /* We can detect an overflow by checking whether the new sum would become negative. */ if (total + i < total) { printf ("Error: sum too large!\n"); exit (1); } /* Everything seems to fit into an int, so continue adding. */ total += i; } return total; } lcov-1.14/example/example.c0000644000000000000000000000263712345570444014352 0ustar rootroot/* * example.c * * Calculate the sum of a given range of integer numbers. The range is * specified by providing two integer numbers as command line argument. * If no arguments are specified, assume the predefined range [0..9]. * Abort with an error message if the resulting number is too big to be * stored as int variable. * * This program example is similar to the one found in the GCOV documentation. * It is used to demonstrate the HTML output generated by LCOV. * * The program is split into 3 modules to better demonstrate the 'directory * overview' function. There are also a lot of bloated comments inserted to * artificially increase the source code size so that the 'source code * overview' function makes at least a minimum of sense. * */ #include #include #include "iterate.h" #include "gauss.h" static int start = 0; static int end = 9; int main (int argc, char* argv[]) { int total1, total2; /* Accept a pair of numbers as command line arguments. */ if (argc == 3) { start = atoi(argv[1]); end = atoi(argv[2]); } /* Use both methods to calculate the result. */ total1 = iterate_get_sum (start, end); total2 = gauss_get_sum (start, end); /* Make sure both results are the same. */ if (total1 != total2) { printf ("Failure (%d != %d)!\n", total1, total2); } else { printf ("Success, sum[%d..%d] = %d\n", start, end, total1); } return 0; } lcov-1.14/example/iterate.h0000644000000000000000000000016612345570444014354 0ustar rootroot#ifndef ITERATE_H #define ITERATE_H ITERATE_H extern int iterate_get_sum (int min, int max); #endif /* ITERATE_H */ lcov-1.14/example/README0000644000000000000000000000023412345570444013422 0ustar rootroot To get an example of how the LCOV generated HTML output looks like, type 'make output' and point a web browser to the resulting file output/index.html lcov-1.14/example/Makefile0000644000000000000000000000473412345570464014215 0ustar rootroot# # Makefile for the LCOV example program. # # Make targets: # - example: compile the example program # - output: run test cases on example program and create HTML output # - clean: clean up directory # CC := gcc CFLAGS := -Wall -I. -fprofile-arcs -ftest-coverage LCOV := ../bin/lcov GENHTML := ../bin/genhtml GENDESC := ../bin/gendesc GENPNG := ../bin/genpng # Depending on the presence of the GD.pm perl module, we can use the # special option '--frames' for genhtml USE_GENPNG := $(shell $(GENPNG) --help >/dev/null 2>/dev/null; echo $$?) ifeq ($(USE_GENPNG),0) FRAMES := --frames else FRAMES := endif .PHONY: clean output test_noargs test_2_to_2000 test_overflow all: output example: example.o iterate.o gauss.o $(CC) example.o iterate.o gauss.o -o example -lgcov example.o: example.c iterate.h gauss.h $(CC) $(CFLAGS) -c example.c -o example.o iterate.o: methods/iterate.c iterate.h $(CC) $(CFLAGS) -c methods/iterate.c -o iterate.o gauss.o: methods/gauss.c gauss.h $(CC) $(CFLAGS) -c methods/gauss.c -o gauss.o output: example descriptions test_noargs test_2_to_2000 test_overflow @echo @echo '*' @echo '* Generating HTML output' @echo '*' @echo $(GENHTML) trace_noargs.info trace_args.info trace_overflow.info \ --output-directory output --title "Basic example" \ --show-details --description-file descriptions $(FRAMES) \ --legend @echo @echo '*' @echo '* See '`pwd`/output/index.html @echo '*' @echo descriptions: descriptions.txt $(GENDESC) descriptions.txt -o descriptions all_tests: example test_noargs test_2_to_2000 test_overflow test_noargs: @echo @echo '*' @echo '* Test case 1: running ./example without parameters' @echo '*' @echo $(LCOV) --zerocounters --directory . ./example $(LCOV) --capture --directory . --output-file trace_noargs.info --test-name test_noargs --no-external test_2_to_2000: @echo @echo '*' @echo '* Test case 2: running ./example 2 2000' @echo '*' @echo $(LCOV) --zerocounters --directory . ./example 2 2000 $(LCOV) --capture --directory . --output-file trace_args.info --test-name test_2_to_2000 --no-external test_overflow: @echo @echo '*' @echo '* Test case 3: running ./example 0 100000 (causes an overflow)' @echo '*' @echo $(LCOV) --zerocounters --directory . ./example 0 100000 || true $(LCOV) --capture --directory . --output-file trace_overflow.info --test-name "test_overflow" --no-external clean: rm -rf *.o *.bb *.bbg *.da *.gcno *.gcda *.info output example \ descriptions lcov-1.14/COPYING0000644000000000000000000004310312345570461012143 0ustar rootroot GNU GENERAL PUBLIC LICENSE Version 2, June 1991 Copyright (C) 1989, 1991 Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed. Preamble The licenses for most software are designed to take away your freedom to share and change it. By contrast, the GNU General Public License is intended to guarantee your freedom to share and change free software--to make sure the software is free for all its users. This General Public License applies to most of the Free Software Foundation's software and to any other program whose authors commit to using it. (Some other Free Software Foundation software is covered by the GNU Lesser General Public License instead.) You can apply it to your programs, too. When we speak of free software, we are referring to freedom, not price. Our General Public Licenses are designed to make sure that you have the freedom to distribute copies of free software (and charge for this service if you wish), that you receive source code or can get it if you want it, that you can change the software or use pieces of it in new free programs; and that you know you can do these things. To protect your rights, we need to make restrictions that forbid anyone to deny you these rights or to ask you to surrender the rights. These restrictions translate to certain responsibilities for you if you distribute copies of the software, or if you modify it. For example, if you distribute copies of such a program, whether gratis or for a fee, you must give the recipients all the rights that you have. You must make sure that they, too, receive or can get the source code. And you must show them these terms so they know their rights. We protect your rights with two steps: (1) copyright the software, and (2) offer you this license which gives you legal permission to copy, distribute and/or modify the software. Also, for each author's protection and ours, we want to make certain that everyone understands that there is no warranty for this free software. If the software is modified by someone else and passed on, we want its recipients to know that what they have is not the original, so that any problems introduced by others will not reflect on the original authors' reputations. Finally, any free program is threatened constantly by software patents. We wish to avoid the danger that redistributors of a free program will individually obtain patent licenses, in effect making the program proprietary. To prevent this, we have made it clear that any patent must be licensed for everyone's free use or not licensed at all. The precise terms and conditions for copying, distribution and modification follow. GNU GENERAL PUBLIC LICENSE TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION 0. This License applies to any program or other work which contains a notice placed by the copyright holder saying it may be distributed under the terms of this General Public License. The "Program", below, refers to any such program or work, and a "work based on the Program" means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language. (Hereinafter, translation is included without limitation in the term "modification".) Each licensee is addressed as "you". Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted, and the output from the Program is covered only if its contents constitute a work based on the Program (independent of having been made by running the Program). Whether that is true depends on what the Program does. 1. You may copy and distribute verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty; keep intact all the notices that refer to this License and to the absence of any warranty; and give any other recipients of the Program a copy of this License along with the Program. You may charge a fee for the physical act of transferring a copy, and you may at your option offer warranty protection in exchange for a fee. 2. You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program, and copy and distribute such modifications or work under the terms of Section 1 above, provided that you also meet all of these conditions: a) You must cause the modified files to carry prominent notices stating that you changed the files and the date of any change. b) You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License. c) If the modified program normally reads commands interactively when run, you must cause it, when started running for such interactive use in the most ordinary way, to print or display an announcement including an appropriate copyright notice and a notice that there is no warranty (or else, saying that you provide a warranty) and that users may redistribute the program under these conditions, and telling the user how to view a copy of this License. (Exception: if the Program itself is interactive but does not normally print such an announcement, your work based on the Program is not required to print an announcement.) These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it. Thus, it is not the intent of this section to claim rights or contest your rights to work written entirely by you; rather, the intent is to exercise the right to control the distribution of derivative or collective works based on the Program. In addition, mere aggregation of another work not based on the Program with the Program (or with a work based on the Program) on a volume of a storage or distribution medium does not bring the other work under the scope of this License. 3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following: a) Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or, b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or, c) Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above.) The source code for a work means the preferred form of the work for making modifications to it. For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable. However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable. If distribution of executable or object code is made by offering access to copy from a designated place, then offering equivalent access to copy the source code from the same place counts as distribution of the source code, even though third parties are not compelled to copy the source along with the object code. 4. You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License. However, parties who have received copies, or rights, from you under this License will not have their licenses terminated so long as such parties remain in full compliance. 5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it. 6. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License. 7. If, as a consequence of a court judgment or allegation of patent infringement or for any other reason (not limited to patent issues), conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License. If you cannot distribute so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not distribute the Program at all. For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program. If any portion of this section is held invalid or unenforceable under any particular circumstance, the balance of the section is intended to apply and the section as a whole is intended to apply in other circumstances. It is not the purpose of this section to induce you to infringe any patents or other property right claims or to contest validity of any such claims; this section has the sole purpose of protecting the integrity of the free software distribution system, which is implemented by public license practices. Many people have made generous contributions to the wide range of software distributed through that system in reliance on consistent application of that system; it is up to the author/donor to decide if he or she is willing to distribute software through any other system and a licensee cannot impose that choice. This section is intended to make thoroughly clear what is believed to be a consequence of the rest of this License. 8. If the distribution and/or use of the Program is restricted in certain countries either by patents or by copyrighted interfaces, the original copyright holder who places the Program under this License may add an explicit geographical distribution limitation excluding those countries, so that distribution is permitted only in or among countries not thus excluded. In such case, this License incorporates the limitation as if written in the body of this License. 9. The Free Software Foundation may publish revised and/or new versions of the General Public License from time to time. Such new versions will be similar in spirit to the present version, but may differ in detail to address new problems or concerns. Each version is given a distinguishing version number. If the Program specifies a version number of this License which applies to it and "any later version", you have the option of following the terms and conditions either of that version or of any later version published by the Free Software Foundation. If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation. 10. If you wish to incorporate parts of the Program into other free programs whose distribution conditions are different, write to the author to ask for permission. For software which is copyrighted by the Free Software Foundation, write to the Free Software Foundation; we sometimes make exceptions for this. Our decision will be guided by the two goals of preserving the free status of all derivatives of our free software and of promoting the sharing and reuse of software generally. NO WARRANTY 11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION. 12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. END OF TERMS AND CONDITIONS How to Apply These Terms to Your New Programs If you develop a new program, and you want it to be of the greatest possible use to the public, the best way to achieve this is to make it free software which everyone can redistribute and change under these terms. To do so, attach the following notices to the program. It is safest to attach them to the start of each source file to most effectively convey the exclusion of warranty; and each file should have at least the "copyright" line and a pointer to where the full notice is found. Copyright (C) This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. Also add information on how to contact you by electronic and paper mail. If the program is interactive, make it output a short notice like this when it starts in an interactive mode: Gnomovision version 69, Copyright (C) year name of author Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'. This is free software, and you are welcome to redistribute it under certain conditions; type `show c' for details. The hypothetical commands `show w' and `show c' should show the appropriate parts of the General Public License. Of course, the commands you use may be called something other than `show w' and `show c'; they could even be mouse-clicks or menu items--whatever suits your program. You should also get your employer (if you work as a programmer) or your school, if any, to sign a "copyright disclaimer" for the program, if necessary. Here is a sample; alter the names: Yoyodyne, Inc., hereby disclaims all copyright interest in the program `Gnomovision' (which makes passes at compilers) written by James Hacker. , 1 April 1989 Ty Coon, President of Vice This General Public License does not permit incorporating your program into proprietary programs. If your program is a subroutine library, you may consider it more useful to permit linking proprietary applications with the library. If this is what you want to do, use the GNU Lesser General Public License instead of this License. lcov-1.14/CONTRIBUTING0000644000000000000000000000561113403461342012735 0ustar rootrootContributing to LCOV ==================== Please read this document if you would like to help improving the LTP GCOV extension (LCOV). In general, all types of contributions are welcome, for example: * Fixes for code or documentation * Performance and compatibility improvements * Functional enhancements There are some rules that these contributions must follow to be acceptable for inclusion: 1. The contribution must align with the project goals of LCOV. 2. The contribution must follow a particular format. 3. The contribution must be signed. Once you have made sure that your contribution follows these rules, send it via e-mail to the LTP coverage mailing list [1]. Signing your work ================= All contributions to LCOV must be signed by putting the following line at the end of the explanation of a patch: Signed-off-by: Your Name By signing a patch, you certify the following: By making a contribution to the LTP GCOV extension (LCOV) on http://ltp.sourceforge.net, I certify that: a) The contribution was created by me and I have the right to submit it under the terms and conditions of the open source license "GNU General Public License, version 2 or later". (http://www.gnu.org/licenses/old-licenses/gpl-2.0.html). b) The contribution is made free of any other party's intellectual property claims or rights. c) I understand and agree that this project and the contribution are public and that a record of the contribution (including all personal information I submit with it, including my sign-off) is maintained indefinitely and may be redistributed consistent with this project or the open source license(s) involved. Project goals ============= The goal of LCOV is to provide a set of command line tools that can be used to collect, process and visualize code coverage data as produced by the gcov tool that is part of the GNU Compiler Collection (GCC) [2]. If you have an idea for a contribution but are unsure if it aligns with the project goals, feel free to discuss the idea on the LTP coverage mailing list [1]. Contribution format =================== To contribute a change, please create a patch using 'git format-patch'. Alternatively you can use the diff utility with the following command line options: diff -Naurp Please base your changes on the most current version of LCOV. You can use the following command line to obtain this version from the lcov Git repository: git clone https://github.com/linux-test-project/lcov.git Add a meaningful description of the contribution to the top of the patch. The description should follow this format: component: short description detailed description Signed-off-by: Your Name With your Signed-off-by, you certify the rules stated in section "Signing your work". -- [1] ltp-coverage@lists.sourceforge.net [2] http://gcc.gnu.org lcov-1.14/bin/0000755000000000000000000000000013436014710011647 5ustar rootrootlcov-1.14/bin/lcov0000755000000000000000000030430613435751076012562 0ustar rootroot#!/usr/bin/env perl # # Copyright (c) International Business Machines Corp., 2002,2012 # # This program is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 2 of the License, or (at # your option) any later version. # # This program is distributed in the hope that it will be useful, but # WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU # General Public License for more details. # # You should have received a copy of the GNU General Public License # along with this program; if not, write to the Free Software # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA # # # lcov # # This is a wrapper script which provides a single interface for accessing # LCOV coverage data. # # # History: # 2002-08-29 created by Peter Oberparleiter # IBM Lab Boeblingen # 2002-09-05 / Peter Oberparleiter: implemented --kernel-directory + # multiple directories # 2002-10-16 / Peter Oberparleiter: implemented --add-tracefile option # 2002-10-17 / Peter Oberparleiter: implemented --extract option # 2002-11-04 / Peter Oberparleiter: implemented --list option # 2003-03-07 / Paul Larson: Changed to make it work with the latest gcov # kernel patch. This will break it with older gcov-kernel # patches unless you change the value of $gcovmod in this script # 2003-04-07 / Peter Oberparleiter: fixed bug which resulted in an error # when trying to combine .info files containing data without # a test name # 2003-04-10 / Peter Oberparleiter: extended Paul's change so that LCOV # works both with the new and the old gcov-kernel patch # 2003-04-10 / Peter Oberparleiter: added $gcov_dir constant in anticipation # of a possible move of the gcov kernel directory to another # file system in a future version of the gcov-kernel patch # 2003-04-15 / Paul Larson: make info write to STDERR, not STDOUT # 2003-04-15 / Paul Larson: added --remove option # 2003-04-30 / Peter Oberparleiter: renamed --reset to --zerocounters # to remove naming ambiguity with --remove # 2003-04-30 / Peter Oberparleiter: adjusted help text to include --remove # 2003-06-27 / Peter Oberparleiter: implemented --diff # 2003-07-03 / Peter Oberparleiter: added line checksum support, added # --no-checksum # 2003-12-11 / Laurent Deniel: added --follow option # 2004-03-29 / Peter Oberparleiter: modified --diff option to better cope with # ambiguous patch file entries, modified --capture option to use # modprobe before insmod (needed for 2.6) # 2004-03-30 / Peter Oberparleiter: added --path option # 2004-08-09 / Peter Oberparleiter: added configuration file support # 2008-08-13 / Peter Oberparleiter: added function coverage support # use strict; use warnings; use File::Basename; use File::Path; use File::Find; use File::Temp qw /tempdir/; use File::Spec::Functions qw /abs2rel canonpath catdir catfile catpath file_name_is_absolute rootdir splitdir splitpath/; use Getopt::Long; use Cwd qw /abs_path getcwd/; # Global constants our $tool_dir = abs_path(dirname($0)); our $lcov_version = "LCOV version 1.14"; our $lcov_url = "http://ltp.sourceforge.net/coverage/lcov.php"; our $tool_name = basename($0); # Directory containing gcov kernel files our $gcov_dir; # Where to create temporary directories our $tmp_dir; # Internal constants our $GKV_PROC = 0; # gcov-kernel data in /proc via external patch our $GKV_SYS = 1; # gcov-kernel data in /sys via vanilla 2.6.31+ our @GKV_NAME = ( "external", "upstream" ); our $pkg_gkv_file = ".gcov_kernel_version"; our $pkg_build_file = ".build_directory"; # Branch data combination types our $BR_SUB = 0; our $BR_ADD = 1; # Prototypes sub print_usage(*); sub check_options(); sub userspace_reset(); sub userspace_capture(); sub kernel_reset(); sub kernel_capture(); sub kernel_capture_initial(); sub package_capture(); sub add_traces(); sub read_info_file($); sub get_info_entry($); sub set_info_entry($$$$$$$$$;$$$$$$); sub add_counts($$); sub merge_checksums($$$); sub combine_info_entries($$$); sub combine_info_files($$); sub write_info_file(*$); sub extract(); sub remove(); sub list(); sub get_common_filename($$); sub read_diff($); sub diff(); sub system_no_output($@); sub read_config($); sub apply_config($); sub info(@); sub create_temp_dir(); sub transform_pattern($); sub warn_handler($); sub die_handler($); sub abort_handler($); sub temp_cleanup(); sub setup_gkv(); sub get_overall_line($$$$); sub print_overall_rate($$$$$$$$$); sub lcov_geninfo(@); sub create_package($$$;$); sub get_func_found_and_hit($); sub summary(); sub rate($$;$$$); # Global variables & initialization our @directory; # Specifies where to get coverage data from our @kernel_directory; # If set, captures only from specified kernel subdirs our @add_tracefile; # If set, reads in and combines all files in list our $list; # If set, list contents of tracefile our $extract; # If set, extracts parts of tracefile our $remove; # If set, removes parts of tracefile our $diff; # If set, modifies tracefile according to diff our $reset; # If set, reset all coverage data to zero our $capture; # If set, capture data our $output_filename; # Name for file to write coverage data to our $test_name = ""; # Test case name our $quiet = ""; # If set, suppress information messages our $help; # Help option flag our $version; # Version option flag our $convert_filenames; # If set, convert filenames when applying diff our $strip; # If set, strip leading directories when applying diff our $temp_dir_name; # Name of temporary directory our $cwd = `pwd`; # Current working directory our $data_stdout; # If set, indicates that data is written to stdout our $follow; # If set, indicates that find shall follow links our $diff_path = ""; # Path removed from tracefile when applying diff our $base_directory; # Base directory (cwd of gcc during compilation) our $checksum; # If set, calculate a checksum for each line our $no_checksum; # If set, don't calculate a checksum for each line our $compat_libtool; # If set, indicates that libtool mode is to be enabled our $no_compat_libtool; # If set, indicates that libtool mode is to be disabled our $gcov_tool; our @opt_ignore_errors; our $initial; our @include_patterns; # List of source file patterns to include our @exclude_patterns; # List of source file patterns to exclude our $no_recursion = 0; our $to_package; our $from_package; our $maxdepth; our $no_markers; our $config; # Configuration file contents chomp($cwd); our @temp_dirs; our $gcov_gkv; # gcov kernel support version found on machine our $opt_derive_func_data; our $opt_debug; our $opt_list_full_path; our $opt_no_list_full_path; our $opt_list_width = 80; our $opt_list_truncate_max = 20; our $opt_external; our $opt_no_external; our $opt_config_file; our %opt_rc; our @opt_summary; our $opt_compat; our $ln_overall_found; our $ln_overall_hit; our $fn_overall_found; our $fn_overall_hit; our $br_overall_found; our $br_overall_hit; our $func_coverage = 1; our $br_coverage = 0; # # Code entry point # $SIG{__WARN__} = \&warn_handler; $SIG{__DIE__} = \&die_handler; $SIG{'INT'} = \&abort_handler; $SIG{'QUIT'} = \&abort_handler; # Check command line for a configuration file name Getopt::Long::Configure("pass_through", "no_auto_abbrev"); GetOptions("config-file=s" => \$opt_config_file, "rc=s%" => \%opt_rc); Getopt::Long::Configure("default"); { # Remove spaces around rc options my %new_opt_rc; while (my ($key, $value) = each(%opt_rc)) { $key =~ s/^\s+|\s+$//g; $value =~ s/^\s+|\s+$//g; $new_opt_rc{$key} = $value; } %opt_rc = %new_opt_rc; } # Read configuration file if available if (defined($opt_config_file)) { $config = read_config($opt_config_file); } elsif (defined($ENV{"HOME"}) && (-r $ENV{"HOME"}."/.lcovrc")) { $config = read_config($ENV{"HOME"}."/.lcovrc"); } elsif (-r "/etc/lcovrc") { $config = read_config("/etc/lcovrc"); } elsif (-r "/usr/local/etc/lcovrc") { $config = read_config("/usr/local/etc/lcovrc"); } if ($config || %opt_rc) { # Copy configuration file and --rc values to variables apply_config({ "lcov_gcov_dir" => \$gcov_dir, "lcov_tmp_dir" => \$tmp_dir, "lcov_list_full_path" => \$opt_list_full_path, "lcov_list_width" => \$opt_list_width, "lcov_list_truncate_max"=> \$opt_list_truncate_max, "lcov_branch_coverage" => \$br_coverage, "lcov_function_coverage"=> \$func_coverage, }); } # Parse command line options if (!GetOptions("directory|d|di=s" => \@directory, "add-tracefile|a=s" => \@add_tracefile, "list|l=s" => \$list, "kernel-directory|k=s" => \@kernel_directory, "extract|e=s" => \$extract, "remove|r=s" => \$remove, "diff=s" => \$diff, "convert-filenames" => \$convert_filenames, "strip=i" => \$strip, "capture|c" => \$capture, "output-file|o=s" => \$output_filename, "test-name|t=s" => \$test_name, "zerocounters|z" => \$reset, "quiet|q" => \$quiet, "help|h|?" => \$help, "version|v" => \$version, "follow|f" => \$follow, "path=s" => \$diff_path, "base-directory|b=s" => \$base_directory, "checksum" => \$checksum, "no-checksum" => \$no_checksum, "compat-libtool" => \$compat_libtool, "no-compat-libtool" => \$no_compat_libtool, "gcov-tool=s" => \$gcov_tool, "ignore-errors=s" => \@opt_ignore_errors, "initial|i" => \$initial, "include=s" => \@include_patterns, "exclude=s" => \@exclude_patterns, "no-recursion" => \$no_recursion, "to-package=s" => \$to_package, "from-package=s" => \$from_package, "no-markers" => \$no_markers, "derive-func-data" => \$opt_derive_func_data, "debug" => \$opt_debug, "list-full-path" => \$opt_list_full_path, "no-list-full-path" => \$opt_no_list_full_path, "external" => \$opt_external, "no-external" => \$opt_no_external, "summary=s" => \@opt_summary, "compat=s" => \$opt_compat, "config-file=s" => \$opt_config_file, "rc=s%" => \%opt_rc, )) { print(STDERR "Use $tool_name --help to get usage information\n"); exit(1); } else { # Merge options if (defined($no_checksum)) { $checksum = ($no_checksum ? 0 : 1); $no_checksum = undef; } if (defined($no_compat_libtool)) { $compat_libtool = ($no_compat_libtool ? 0 : 1); $no_compat_libtool = undef; } if (defined($opt_no_list_full_path)) { $opt_list_full_path = ($opt_no_list_full_path ? 0 : 1); $opt_no_list_full_path = undef; } if (defined($opt_no_external)) { $opt_external = 0; $opt_no_external = undef; } } # Check for help option if ($help) { print_usage(*STDOUT); exit(0); } # Check for version option if ($version) { print("$tool_name: $lcov_version\n"); exit(0); } # Check list width option if ($opt_list_width <= 40) { die("ERROR: lcov_list_width parameter out of range (needs to be ". "larger than 40)\n"); } # Normalize --path text $diff_path =~ s/\/$//; if ($follow) { $follow = "-follow"; } else { $follow = ""; } if ($no_recursion) { $maxdepth = "-maxdepth 1"; } else { $maxdepth = ""; } # Check for valid options check_options(); # Only --extract, --remove and --diff allow unnamed parameters if (@ARGV && !($extract || $remove || $diff || @opt_summary)) { die("Extra parameter found: '".join(" ", @ARGV)."'\n". "Use $tool_name --help to get usage information\n"); } # Check for output filename $data_stdout = !($output_filename && ($output_filename ne "-")); if ($capture) { if ($data_stdout) { # Option that tells geninfo to write to stdout $output_filename = "-"; } } # Determine kernel directory for gcov data if (!$from_package && !@directory && ($capture || $reset)) { ($gcov_gkv, $gcov_dir) = setup_gkv(); } # Check for requested functionality if ($reset) { $data_stdout = 0; # Differentiate between user space and kernel reset if (@directory) { userspace_reset(); } else { kernel_reset(); } } elsif ($capture) { # Capture source can be user space, kernel or package if ($from_package) { package_capture(); } elsif (@directory) { userspace_capture(); } else { if ($initial) { if (defined($to_package)) { die("ERROR: --initial cannot be used together ". "with --to-package\n"); } kernel_capture_initial(); } else { kernel_capture(); } } } elsif (@add_tracefile) { ($ln_overall_found, $ln_overall_hit, $fn_overall_found, $fn_overall_hit, $br_overall_found, $br_overall_hit) = add_traces(); } elsif ($remove) { ($ln_overall_found, $ln_overall_hit, $fn_overall_found, $fn_overall_hit, $br_overall_found, $br_overall_hit) = remove(); } elsif ($extract) { ($ln_overall_found, $ln_overall_hit, $fn_overall_found, $fn_overall_hit, $br_overall_found, $br_overall_hit) = extract(); } elsif ($list) { $data_stdout = 0; list(); } elsif ($diff) { if (scalar(@ARGV) != 1) { die("ERROR: option --diff requires one additional argument!\n". "Use $tool_name --help to get usage information\n"); } ($ln_overall_found, $ln_overall_hit, $fn_overall_found, $fn_overall_hit, $br_overall_found, $br_overall_hit) = diff(); } elsif (@opt_summary) { $data_stdout = 0; ($ln_overall_found, $ln_overall_hit, $fn_overall_found, $fn_overall_hit, $br_overall_found, $br_overall_hit) = summary(); } temp_cleanup(); if (defined($ln_overall_found)) { print_overall_rate(1, $ln_overall_found, $ln_overall_hit, 1, $fn_overall_found, $fn_overall_hit, 1, $br_overall_found, $br_overall_hit); } else { info("Done.\n") if (!$list && !$capture); } exit(0); # # print_usage(handle) # # Print usage information. # sub print_usage(*) { local *HANDLE = $_[0]; print(HANDLE < 1) { die("ERROR: only one of -z, -c, -a, -e, -r, -l, ". "--diff or --summary allowed!\n". "Use $tool_name --help to get usage information\n"); } } # # userspace_reset() # # Reset coverage data found in DIRECTORY by deleting all contained .da files. # # Die on error. # sub userspace_reset() { my $current_dir; my @file_list; foreach $current_dir (@directory) { info("Deleting all .da files in $current_dir". ($no_recursion?"\n":" and subdirectories\n")); @file_list = `find "$current_dir" $maxdepth $follow -name \\*\\.da -type f -o -name \\*\\.gcda -type f 2>/dev/null`; chomp(@file_list); foreach (@file_list) { unlink($_) or die("ERROR: cannot remove file $_!\n"); } } } # # userspace_capture() # # Capture coverage data found in DIRECTORY and write it to a package (if # TO_PACKAGE specified) or to OUTPUT_FILENAME or STDOUT. # # Die on error. # sub userspace_capture() { my $dir; my $build; if (!defined($to_package)) { lcov_geninfo(@directory); return; } if (scalar(@directory) != 1) { die("ERROR: -d may be specified only once with --to-package\n"); } $dir = $directory[0]; if (defined($base_directory)) { $build = $base_directory; } else { $build = $dir; } create_package($to_package, $dir, $build); } # # kernel_reset() # # Reset kernel coverage. # # Die on error. # sub kernel_reset() { local *HANDLE; my $reset_file; info("Resetting kernel execution counters\n"); if (-e "$gcov_dir/vmlinux") { $reset_file = "$gcov_dir/vmlinux"; } elsif (-e "$gcov_dir/reset") { $reset_file = "$gcov_dir/reset"; } else { die("ERROR: no reset control found in $gcov_dir\n"); } open(HANDLE, ">", $reset_file) or die("ERROR: cannot write to $reset_file!\n"); print(HANDLE "0"); close(HANDLE); } # # lcov_copy_single(from, to) # # Copy single regular file FROM to TO without checking its size. This is # required to work with special files generated by the kernel # seq_file-interface. # # sub lcov_copy_single($$) { my ($from, $to) = @_; my $content; local $/; local *HANDLE; open(HANDLE, "<", $from) or die("ERROR: cannot read $from: $!\n"); $content = ; close(HANDLE); open(HANDLE, ">", $to) or die("ERROR: cannot write $from: $!\n"); if (defined($content)) { print(HANDLE $content); } close(HANDLE); } # # lcov_find(dir, function, data[, extension, ...)]) # # Search DIR for files and directories whose name matches PATTERN and run # FUNCTION for each match. If not pattern is specified, match all names. # # FUNCTION has the following prototype: # function(dir, relative_name, data) # # Where: # dir: the base directory for this search # relative_name: the name relative to the base directory of this entry # data: the DATA variable passed to lcov_find # sub lcov_find($$$;@) { my ($dir, $fn, $data, @pattern) = @_; my $result; my $_fn = sub { my $filename = $File::Find::name; if (defined($result)) { return; } $filename = abs2rel($filename, $dir); foreach (@pattern) { if ($filename =~ /$_/) { goto ok; } } return; ok: $result = &$fn($dir, $filename, $data); }; if (scalar(@pattern) == 0) { @pattern = ".*"; } find( { wanted => $_fn, no_chdir => 1 }, $dir); return $result; } # # lcov_copy_fn(from, rel, to) # # Copy directories, files and links from/rel to to/rel. # sub lcov_copy_fn($$$) { my ($from, $rel, $to) = @_; my $absfrom = canonpath(catfile($from, $rel)); my $absto = canonpath(catfile($to, $rel)); if (-d) { if (! -d $absto) { mkpath($absto) or die("ERROR: cannot create directory $absto\n"); chmod(0700, $absto); } } elsif (-l) { # Copy symbolic link my $link = readlink($absfrom); if (!defined($link)) { die("ERROR: cannot read link $absfrom: $!\n"); } symlink($link, $absto) or die("ERROR: cannot create link $absto: $!\n"); } else { lcov_copy_single($absfrom, $absto); chmod(0600, $absto); } return undef; } # # lcov_copy(from, to, subdirs) # # Copy all specified SUBDIRS and files from directory FROM to directory TO. For # regular files, copy file contents without checking its size. This is required # to work with seq_file-generated files. # sub lcov_copy($$;@) { my ($from, $to, @subdirs) = @_; my @pattern; foreach (@subdirs) { push(@pattern, "^$_"); } lcov_find($from, \&lcov_copy_fn, $to, @pattern); } # # lcov_geninfo(directory) # # Call geninfo for the specified directory and with the parameters specified # at the command line. # sub lcov_geninfo(@) { my (@dir) = @_; my @param; # Capture data info("Capturing coverage data from ".join(" ", @dir)."\n"); @param = ("$tool_dir/geninfo", @dir); if ($output_filename) { @param = (@param, "--output-filename", $output_filename); } if ($test_name) { @param = (@param, "--test-name", $test_name); } if ($follow) { @param = (@param, "--follow"); } if ($quiet) { @param = (@param, "--quiet"); } if (defined($checksum)) { if ($checksum) { @param = (@param, "--checksum"); } else { @param = (@param, "--no-checksum"); } } if ($base_directory) { @param = (@param, "--base-directory", $base_directory); } if ($no_compat_libtool) { @param = (@param, "--no-compat-libtool"); } elsif ($compat_libtool) { @param = (@param, "--compat-libtool"); } if ($gcov_tool) { @param = (@param, "--gcov-tool", $gcov_tool); } foreach (@opt_ignore_errors) { @param = (@param, "--ignore-errors", $_); } if ($no_recursion) { @param = (@param, "--no-recursion"); } if ($initial) { @param = (@param, "--initial"); } if ($no_markers) { @param = (@param, "--no-markers"); } if ($opt_derive_func_data) { @param = (@param, "--derive-func-data"); } if ($opt_debug) { @param = (@param, "--debug"); } if (defined($opt_external) && $opt_external) { @param = (@param, "--external"); } if (defined($opt_external) && !$opt_external) { @param = (@param, "--no-external"); } if (defined($opt_compat)) { @param = (@param, "--compat", $opt_compat); } if (%opt_rc) { foreach my $key (keys(%opt_rc)) { @param = (@param, "--rc", "$key=".$opt_rc{$key}); } } if (defined($opt_config_file)) { @param = (@param, "--config-file", $opt_config_file); } foreach (@include_patterns) { @param = (@param, "--include", $_); } foreach (@exclude_patterns) { @param = (@param, "--exclude", $_); } system(@param) and exit($? >> 8); } # # read_file(filename) # # Return the contents of the file defined by filename. # sub read_file($) { my ($filename) = @_; my $content; local $\; local *HANDLE; open(HANDLE, "<", $filename) || return undef; $content = ; close(HANDLE); return $content; } # # get_package(package_file) # # Unpack unprocessed coverage data files from package_file to a temporary # directory and return directory name, build directory and gcov kernel version # as found in package. # sub get_package($) { my ($file) = @_; my $dir = create_temp_dir(); my $gkv; my $build; my $cwd = getcwd(); my $count; local *HANDLE; info("Reading package $file:\n"); $file = abs_path($file); chdir($dir); open(HANDLE, "-|", "tar xvfz '$file' 2>/dev/null") or die("ERROR: could not process package $file\n"); $count = 0; while () { if (/\.da$/ || /\.gcda$/) { $count++; } } close(HANDLE); if ($count == 0) { die("ERROR: no data file found in package $file\n"); } info(" data directory .......: $dir\n"); $build = read_file("$dir/$pkg_build_file"); if (defined($build)) { info(" build directory ......: $build\n"); } $gkv = read_file("$dir/$pkg_gkv_file"); if (defined($gkv)) { $gkv = int($gkv); if ($gkv != $GKV_PROC && $gkv != $GKV_SYS) { die("ERROR: unsupported gcov kernel version found ". "($gkv)\n"); } info(" content type .........: kernel data\n"); info(" gcov kernel version ..: %s\n", $GKV_NAME[$gkv]); } else { info(" content type .........: application data\n"); } info(" data files ...........: $count\n"); chdir($cwd); return ($dir, $build, $gkv); } # # write_file(filename, $content) # # Create a file named filename and write the specified content to it. # sub write_file($$) { my ($filename, $content) = @_; local *HANDLE; open(HANDLE, ">", $filename) || return 0; print(HANDLE $content); close(HANDLE) || return 0; return 1; } # count_package_data(filename) # # Count the number of coverage data files in the specified package file. # sub count_package_data($) { my ($filename) = @_; local *HANDLE; my $count = 0; open(HANDLE, "-|", "tar tfz '$filename'") or return undef; while () { if (/\.da$/ || /\.gcda$/) { $count++; } } close(HANDLE); return $count; } # # create_package(package_file, source_directory, build_directory[, # kernel_gcov_version]) # # Store unprocessed coverage data files from source_directory to package_file. # sub create_package($$$;$) { my ($file, $dir, $build, $gkv) = @_; my $cwd = getcwd(); # Check for availability of tar tool first system("tar --help > /dev/null") and die("ERROR: tar command not available\n"); # Print information about the package info("Creating package $file:\n"); info(" data directory .......: $dir\n"); # Handle build directory if (defined($build)) { info(" build directory ......: $build\n"); write_file("$dir/$pkg_build_file", $build) or die("ERROR: could not write to ". "$dir/$pkg_build_file\n"); } # Handle gcov kernel version data if (defined($gkv)) { info(" content type .........: kernel data\n"); info(" gcov kernel version ..: %s\n", $GKV_NAME[$gkv]); write_file("$dir/$pkg_gkv_file", $gkv) or die("ERROR: could not write to ". "$dir/$pkg_gkv_file\n"); } else { info(" content type .........: application data\n"); } # Create package $file = abs_path($file); chdir($dir); system("tar cfz $file .") and die("ERROR: could not create package $file\n"); chdir($cwd); # Remove temporary files unlink("$dir/$pkg_build_file"); unlink("$dir/$pkg_gkv_file"); # Show number of data files if (!$quiet) { my $count = count_package_data($file); if (defined($count)) { info(" data files ...........: $count\n"); } } } sub find_link_fn($$$) { my ($from, $rel, $filename) = @_; my $absfile = catfile($from, $rel, $filename); if (-l $absfile) { return $absfile; } return undef; } # # get_base(dir) # # Return (BASE, OBJ), where # - BASE: is the path to the kernel base directory relative to dir # - OBJ: is the absolute path to the kernel build directory # sub get_base($) { my ($dir) = @_; my $marker = "kernel/gcov/base.gcno"; my $markerfile; my $sys; my $obj; my $link; $markerfile = lcov_find($dir, \&find_link_fn, $marker); if (!defined($markerfile)) { return (undef, undef); } # sys base is parent of parent of markerfile. $sys = abs2rel(dirname(dirname(dirname($markerfile))), $dir); # obj base is parent of parent of markerfile link target. $link = readlink($markerfile); if (!defined($link)) { die("ERROR: could not read $markerfile\n"); } $obj = dirname(dirname(dirname($link))); return ($sys, $obj); } # # apply_base_dir(data_dir, base_dir, build_dir, @directories) # # Make entries in @directories relative to data_dir. # sub apply_base_dir($$$@) { my ($data, $base, $build, @dirs) = @_; my $dir; my @result; foreach $dir (@dirs) { # Is directory path relative to data directory? if (-d catdir($data, $dir)) { push(@result, $dir); next; } # Relative to the auto-detected base-directory? if (defined($base)) { if (-d catdir($data, $base, $dir)) { push(@result, catdir($base, $dir)); next; } } # Relative to the specified base-directory? if (defined($base_directory)) { if (file_name_is_absolute($base_directory)) { $base = abs2rel($base_directory, rootdir()); } else { $base = $base_directory; } if (-d catdir($data, $base, $dir)) { push(@result, catdir($base, $dir)); next; } } # Relative to the build directory? if (defined($build)) { if (file_name_is_absolute($build)) { $base = abs2rel($build, rootdir()); } else { $base = $build; } if (-d catdir($data, $base, $dir)) { push(@result, catdir($base, $dir)); next; } } die("ERROR: subdirectory $dir not found\n". "Please use -b to specify the correct directory\n"); } return @result; } # # copy_gcov_dir(dir, [@subdirectories]) # # Create a temporary directory and copy all or, if specified, only some # subdirectories from dir to that directory. Return the name of the temporary # directory. # sub copy_gcov_dir($;@) { my ($data, @dirs) = @_; my $tempdir = create_temp_dir(); info("Copying data to temporary directory $tempdir\n"); lcov_copy($data, $tempdir, @dirs); return $tempdir; } # # kernel_capture_initial # # Capture initial kernel coverage data, i.e. create a coverage data file from # static graph files which contains zero coverage data for all instrumented # lines. # sub kernel_capture_initial() { my $build; my $source; my @params; if (defined($base_directory)) { $build = $base_directory; $source = "specified"; } else { (undef, $build) = get_base($gcov_dir); if (!defined($build)) { die("ERROR: could not auto-detect build directory.\n". "Please use -b to specify the build directory\n"); } $source = "auto-detected"; } info("Using $build as kernel build directory ($source)\n"); # Build directory needs to be passed to geninfo $base_directory = $build; if (@kernel_directory) { foreach my $dir (@kernel_directory) { push(@params, "$build/$dir"); } } else { push(@params, $build); } lcov_geninfo(@params); } # # kernel_capture_from_dir(directory, gcov_kernel_version, build) # # Perform the actual kernel coverage capturing from the specified directory # assuming that the data was copied from the specified gcov kernel version. # sub kernel_capture_from_dir($$$) { my ($dir, $gkv, $build) = @_; # Create package or coverage file if (defined($to_package)) { create_package($to_package, $dir, $build, $gkv); } else { # Build directory needs to be passed to geninfo $base_directory = $build; lcov_geninfo($dir); } } # # adjust_kernel_dir(dir, build) # # Adjust directories specified with -k so that they point to the directory # relative to DIR. Return the build directory if specified or the auto- # detected build-directory. # sub adjust_kernel_dir($$) { my ($dir, $build) = @_; my ($sys_base, $build_auto) = get_base($dir); if (!defined($build)) { $build = $build_auto; } if (!defined($build)) { die("ERROR: could not auto-detect build directory.\n". "Please use -b to specify the build directory\n"); } # Make @kernel_directory relative to sysfs base if (@kernel_directory) { @kernel_directory = apply_base_dir($dir, $sys_base, $build, @kernel_directory); } return $build; } sub kernel_capture() { my $data_dir; my $build = $base_directory; if ($gcov_gkv == $GKV_SYS) { $build = adjust_kernel_dir($gcov_dir, $build); } $data_dir = copy_gcov_dir($gcov_dir, @kernel_directory); kernel_capture_from_dir($data_dir, $gcov_gkv, $build); } # # link_data_cb(datadir, rel, graphdir) # # Create symbolic link in GRAPDIR/REL pointing to DATADIR/REL. # sub link_data_cb($$$) { my ($datadir, $rel, $graphdir) = @_; my $absfrom = catfile($datadir, $rel); my $absto = catfile($graphdir, $rel); my $base; my $dir; if (-e $absto) { die("ERROR: could not create symlink at $absto: ". "File already exists!\n"); } if (-l $absto) { # Broken link - possibly from an interrupted earlier run unlink($absto); } # Check for graph file $base = $absto; $base =~ s/\.(gcda|da)$//; if (! -e $base.".gcno" && ! -e $base.".bbg" && ! -e $base.".bb") { die("ERROR: No graph file found for $absfrom in ". dirname($base)."!\n"); } symlink($absfrom, $absto) or die("ERROR: could not create symlink at $absto: $!\n"); } # # unlink_data_cb(datadir, rel, graphdir) # # Remove symbolic link from GRAPHDIR/REL to DATADIR/REL. # sub unlink_data_cb($$$) { my ($datadir, $rel, $graphdir) = @_; my $absfrom = catfile($datadir, $rel); my $absto = catfile($graphdir, $rel); my $target; return if (!-l $absto); $target = readlink($absto); return if (!defined($target) || $target ne $absfrom); unlink($absto) or warn("WARNING: could not remove symlink $absto: $!\n"); } # # link_data(datadir, graphdir, create) # # If CREATE is non-zero, create symbolic links in GRAPHDIR for data files # found in DATADIR. Otherwise remove link in GRAPHDIR. # sub link_data($$$) { my ($datadir, $graphdir, $create) = @_; $datadir = abs_path($datadir); $graphdir = abs_path($graphdir); if ($create) { lcov_find($datadir, \&link_data_cb, $graphdir, '\.gcda$', '\.da$'); } else { lcov_find($datadir, \&unlink_data_cb, $graphdir, '\.gcda$', '\.da$'); } } # # find_graph_cb(datadir, rel, count_ref) # # Count number of files found. # sub find_graph_cb($$$) { my ($dir, $rel, $count_ref) = @_; ($$count_ref)++; } # # find_graph(dir) # # Search DIR for a graph file. Return non-zero if one was found, zero otherwise. # sub find_graph($) { my ($dir) = @_; my $count = 0; lcov_find($dir, \&find_graph_cb, \$count, '\.gcno$', '\.bb$', '\.bbg$'); return $count > 0 ? 1 : 0; } # # package_capture() # # Capture coverage data from a package of unprocessed coverage data files # as generated by lcov --to-package. # sub package_capture() { my $dir; my $build; my $gkv; ($dir, $build, $gkv) = get_package($from_package); # Check for build directory if (defined($base_directory)) { if (defined($build)) { info("Using build directory specified by -b.\n"); } $build = $base_directory; } # Do the actual capture if (defined($gkv)) { if ($gkv == $GKV_SYS) { $build = adjust_kernel_dir($dir, $build); } if (@kernel_directory) { $dir = copy_gcov_dir($dir, @kernel_directory); } kernel_capture_from_dir($dir, $gkv, $build); } else { # Build directory needs to be passed to geninfo $base_directory = $build; if (find_graph($dir)) { # Package contains graph files - collect from there lcov_geninfo($dir); } else { # No graph files found, link data files next to # graph files link_data($dir, $base_directory, 1); lcov_geninfo($base_directory); link_data($dir, $base_directory, 0); } } } # # info(printf_parameter) # # Use printf to write PRINTF_PARAMETER to stdout only when the $quiet flag # is not set. # sub info(@) { if (!$quiet) { # Print info string if (!$data_stdout) { printf(@_) } else { # Don't interfere with the .info output to STDOUT printf(STDERR @_); } } } # # create_temp_dir() # # Create a temporary directory and return its path. # # Die on error. # sub create_temp_dir() { my $dir; if (defined($tmp_dir)) { $dir = tempdir(DIR => $tmp_dir, CLEANUP => 1); } else { $dir = tempdir(CLEANUP => 1); } if (!defined($dir)) { die("ERROR: cannot create temporary directory\n"); } push(@temp_dirs, $dir); return $dir; } sub compress_brcount($) { my ($brcount) = @_; my $db; $db = brcount_to_db($brcount); return db_to_brcount($db, $brcount); } sub get_br_found_and_hit($) { my ($brcount) = @_; my $db; $db = brcount_to_db($brcount); return brcount_db_get_found_and_hit($db); } # # read_info_file(info_filename) # # Read in the contents of the .info file specified by INFO_FILENAME. Data will # be returned as a reference to a hash containing the following mappings: # # %result: for each filename found in file -> \%data # # %data: "test" -> \%testdata # "sum" -> \%sumcount # "func" -> \%funcdata # "found" -> $lines_found (number of instrumented lines found in file) # "hit" -> $lines_hit (number of executed lines in file) # "f_found" -> $fn_found (number of instrumented functions found in file) # "f_hit" -> $fn_hit (number of executed functions in file) # "b_found" -> $br_found (number of instrumented branches found in file) # "b_hit" -> $br_hit (number of executed branches in file) # "check" -> \%checkdata # "testfnc" -> \%testfncdata # "sumfnc" -> \%sumfnccount # "testbr" -> \%testbrdata # "sumbr" -> \%sumbrcount # # %testdata : name of test affecting this file -> \%testcount # %testfncdata: name of test affecting this file -> \%testfnccount # %testbrdata: name of test affecting this file -> \%testbrcount # # %testcount : line number -> execution count for a single test # %testfnccount: function name -> execution count for a single test # %testbrcount : line number -> branch coverage data for a single test # %sumcount : line number -> execution count for all tests # %sumfnccount : function name -> execution count for all tests # %sumbrcount : line number -> branch coverage data for all tests # %funcdata : function name -> line number # %checkdata : line number -> checksum of source code line # $brdata : text "block,branch,taken:..." # # Note that .info file sections referring to the same file and test name # will automatically be combined by adding all execution counts. # # Note that if INFO_FILENAME ends with ".gz", it is assumed that the file # is compressed using GZIP. If available, GUNZIP will be used to decompress # this file. # # Die on error. # sub read_info_file($) { my $tracefile = $_[0]; # Name of tracefile my %result; # Resulting hash: file -> data my $data; # Data handle for current entry my $testdata; # " " my $testcount; # " " my $sumcount; # " " my $funcdata; # " " my $checkdata; # " " my $testfncdata; my $testfnccount; my $sumfnccount; my $testbrdata; my $testbrcount; my $sumbrcount; my $line; # Current line read from .info file my $testname; # Current test name my $filename; # Current filename my $hitcount; # Count for lines hit my $count; # Execution count of current line my $negative; # If set, warn about negative counts my $changed_testname; # If set, warn about changed testname my $line_checksum; # Checksum of current line local *INFO_HANDLE; # Filehandle for .info file info("Reading tracefile $tracefile\n"); # Check if file exists and is readable stat($_[0]); if (!(-r _)) { die("ERROR: cannot read file $_[0]!\n"); } # Check if this is really a plain file if (!(-f _)) { die("ERROR: not a plain file: $_[0]!\n"); } # Check for .gz extension if ($_[0] =~ /\.gz$/) { # Check for availability of GZIP tool system_no_output(1, "gunzip" ,"-h") and die("ERROR: gunzip command not available!\n"); # Check integrity of compressed file system_no_output(1, "gunzip", "-t", $_[0]) and die("ERROR: integrity check failed for ". "compressed file $_[0]!\n"); # Open compressed file open(INFO_HANDLE, "-|", "gunzip -c '$_[0]'") or die("ERROR: cannot start gunzip to decompress ". "file $_[0]!\n"); } else { # Open decompressed file open(INFO_HANDLE, "<", $_[0]) or die("ERROR: cannot read file $_[0]!\n"); } $testname = ""; while () { chomp($_); $line = $_; # Switch statement foreach ($line) { /^TN:([^,]*)(,diff)?/ && do { # Test name information found $testname = defined($1) ? $1 : ""; if ($testname =~ s/\W/_/g) { $changed_testname = 1; } $testname .= $2 if (defined($2)); last; }; /^[SK]F:(.*)/ && do { # Filename information found # Retrieve data for new entry $filename = $1; $data = $result{$filename}; ($testdata, $sumcount, $funcdata, $checkdata, $testfncdata, $sumfnccount, $testbrdata, $sumbrcount) = get_info_entry($data); if (defined($testname)) { $testcount = $testdata->{$testname}; $testfnccount = $testfncdata->{$testname}; $testbrcount = $testbrdata->{$testname}; } else { $testcount = {}; $testfnccount = {}; $testbrcount = {}; } last; }; /^DA:(\d+),(-?\d+)(,[^,\s]+)?/ && do { # Fix negative counts $count = $2 < 0 ? 0 : $2; if ($2 < 0) { $negative = 1; } # Execution count found, add to structure # Add summary counts $sumcount->{$1} += $count; # Add test-specific counts if (defined($testname)) { $testcount->{$1} += $count; } # Store line checksum if available if (defined($3)) { $line_checksum = substr($3, 1); # Does it match a previous definition if (defined($checkdata->{$1}) && ($checkdata->{$1} ne $line_checksum)) { die("ERROR: checksum mismatch ". "at $filename:$1\n"); } $checkdata->{$1} = $line_checksum; } last; }; /^FN:(\d+),([^,]+)/ && do { last if (!$func_coverage); # Function data found, add to structure $funcdata->{$2} = $1; # Also initialize function call data if (!defined($sumfnccount->{$2})) { $sumfnccount->{$2} = 0; } if (defined($testname)) { if (!defined($testfnccount->{$2})) { $testfnccount->{$2} = 0; } } last; }; /^FNDA:(\d+),([^,]+)/ && do { last if (!$func_coverage); # Function call count found, add to structure # Add summary counts $sumfnccount->{$2} += $1; # Add test-specific counts if (defined($testname)) { $testfnccount->{$2} += $1; } last; }; /^BRDA:(\d+),(\d+),(\d+),(\d+|-)/ && do { # Branch coverage data found my ($line, $block, $branch, $taken) = ($1, $2, $3, $4); last if (!$br_coverage); $sumbrcount->{$line} .= "$block,$branch,$taken:"; # Add test-specific counts if (defined($testname)) { $testbrcount->{$line} .= "$block,$branch,$taken:"; } last; }; /^end_of_record/ && do { # Found end of section marker if ($filename) { # Store current section data if (defined($testname)) { $testdata->{$testname} = $testcount; $testfncdata->{$testname} = $testfnccount; $testbrdata->{$testname} = $testbrcount; } set_info_entry($data, $testdata, $sumcount, $funcdata, $checkdata, $testfncdata, $sumfnccount, $testbrdata, $sumbrcount); $result{$filename} = $data; last; } }; # default last; } } close(INFO_HANDLE); # Calculate hit and found values for lines and functions of each file foreach $filename (keys(%result)) { $data = $result{$filename}; ($testdata, $sumcount, undef, undef, $testfncdata, $sumfnccount, $testbrdata, $sumbrcount) = get_info_entry($data); # Filter out empty files if (scalar(keys(%{$sumcount})) == 0) { delete($result{$filename}); next; } # Filter out empty test cases foreach $testname (keys(%{$testdata})) { if (!defined($testdata->{$testname}) || scalar(keys(%{$testdata->{$testname}})) == 0) { delete($testdata->{$testname}); delete($testfncdata->{$testname}); } } $data->{"found"} = scalar(keys(%{$sumcount})); $hitcount = 0; foreach (keys(%{$sumcount})) { if ($sumcount->{$_} > 0) { $hitcount++; } } $data->{"hit"} = $hitcount; # Get found/hit values for function call data $data->{"f_found"} = scalar(keys(%{$sumfnccount})); $hitcount = 0; foreach (keys(%{$sumfnccount})) { if ($sumfnccount->{$_} > 0) { $hitcount++; } } $data->{"f_hit"} = $hitcount; # Combine branch data for the same branches (undef, $data->{"b_found"}, $data->{"b_hit"}) = compress_brcount($sumbrcount); foreach $testname (keys(%{$testbrdata})) { compress_brcount($testbrdata->{$testname}); } } if (scalar(keys(%result)) == 0) { die("ERROR: no valid records found in tracefile $tracefile\n"); } if ($negative) { warn("WARNING: negative counts found in tracefile ". "$tracefile\n"); } if ($changed_testname) { warn("WARNING: invalid characters removed from testname in ". "tracefile $tracefile\n"); } return(\%result); } # # get_info_entry(hash_ref) # # Retrieve data from an entry of the structure generated by read_info_file(). # Return a list of references to hashes: # (test data hash ref, sum count hash ref, funcdata hash ref, checkdata hash # ref, testfncdata hash ref, sumfnccount hash ref, testbrdata hash ref, # sumbrcount hash ref, lines found, lines hit, functions found, # functions hit, branches found, branches hit) # sub get_info_entry($) { my $testdata_ref = $_[0]->{"test"}; my $sumcount_ref = $_[0]->{"sum"}; my $funcdata_ref = $_[0]->{"func"}; my $checkdata_ref = $_[0]->{"check"}; my $testfncdata = $_[0]->{"testfnc"}; my $sumfnccount = $_[0]->{"sumfnc"}; my $testbrdata = $_[0]->{"testbr"}; my $sumbrcount = $_[0]->{"sumbr"}; my $lines_found = $_[0]->{"found"}; my $lines_hit = $_[0]->{"hit"}; my $f_found = $_[0]->{"f_found"}; my $f_hit = $_[0]->{"f_hit"}; my $br_found = $_[0]->{"b_found"}; my $br_hit = $_[0]->{"b_hit"}; return ($testdata_ref, $sumcount_ref, $funcdata_ref, $checkdata_ref, $testfncdata, $sumfnccount, $testbrdata, $sumbrcount, $lines_found, $lines_hit, $f_found, $f_hit, $br_found, $br_hit); } # # set_info_entry(hash_ref, testdata_ref, sumcount_ref, funcdata_ref, # checkdata_ref, testfncdata_ref, sumfcncount_ref, # testbrdata_ref, sumbrcount_ref[,lines_found, # lines_hit, f_found, f_hit, $b_found, $b_hit]) # # Update the hash referenced by HASH_REF with the provided data references. # sub set_info_entry($$$$$$$$$;$$$$$$) { my $data_ref = $_[0]; $data_ref->{"test"} = $_[1]; $data_ref->{"sum"} = $_[2]; $data_ref->{"func"} = $_[3]; $data_ref->{"check"} = $_[4]; $data_ref->{"testfnc"} = $_[5]; $data_ref->{"sumfnc"} = $_[6]; $data_ref->{"testbr"} = $_[7]; $data_ref->{"sumbr"} = $_[8]; if (defined($_[9])) { $data_ref->{"found"} = $_[9]; } if (defined($_[10])) { $data_ref->{"hit"} = $_[10]; } if (defined($_[11])) { $data_ref->{"f_found"} = $_[11]; } if (defined($_[12])) { $data_ref->{"f_hit"} = $_[12]; } if (defined($_[13])) { $data_ref->{"b_found"} = $_[13]; } if (defined($_[14])) { $data_ref->{"b_hit"} = $_[14]; } } # # add_counts(data1_ref, data2_ref) # # DATA1_REF and DATA2_REF are references to hashes containing a mapping # # line number -> execution count # # Return a list (RESULT_REF, LINES_FOUND, LINES_HIT) where RESULT_REF # is a reference to a hash containing the combined mapping in which # execution counts are added. # sub add_counts($$) { my $data1_ref = $_[0]; # Hash 1 my $data2_ref = $_[1]; # Hash 2 my %result; # Resulting hash my $line; # Current line iteration scalar my $data1_count; # Count of line in hash1 my $data2_count; # Count of line in hash2 my $found = 0; # Total number of lines found my $hit = 0; # Number of lines with a count > 0 foreach $line (keys(%$data1_ref)) { $data1_count = $data1_ref->{$line}; $data2_count = $data2_ref->{$line}; # Add counts if present in both hashes if (defined($data2_count)) { $data1_count += $data2_count; } # Store sum in %result $result{$line} = $data1_count; $found++; if ($data1_count > 0) { $hit++; } } # Add lines unique to data2_ref foreach $line (keys(%$data2_ref)) { # Skip lines already in data1_ref if (defined($data1_ref->{$line})) { next; } # Copy count from data2_ref $result{$line} = $data2_ref->{$line}; $found++; if ($result{$line} > 0) { $hit++; } } return (\%result, $found, $hit); } # # merge_checksums(ref1, ref2, filename) # # REF1 and REF2 are references to hashes containing a mapping # # line number -> checksum # # Merge checksum lists defined in REF1 and REF2 and return reference to # resulting hash. Die if a checksum for a line is defined in both hashes # but does not match. # sub merge_checksums($$$) { my $ref1 = $_[0]; my $ref2 = $_[1]; my $filename = $_[2]; my %result; my $line; foreach $line (keys(%{$ref1})) { if (defined($ref2->{$line}) && ($ref1->{$line} ne $ref2->{$line})) { die("ERROR: checksum mismatch at $filename:$line\n"); } $result{$line} = $ref1->{$line}; } foreach $line (keys(%{$ref2})) { $result{$line} = $ref2->{$line}; } return \%result; } # # merge_func_data(funcdata1, funcdata2, filename) # sub merge_func_data($$$) { my ($funcdata1, $funcdata2, $filename) = @_; my %result; my $func; if (defined($funcdata1)) { %result = %{$funcdata1}; } foreach $func (keys(%{$funcdata2})) { my $line1 = $result{$func}; my $line2 = $funcdata2->{$func}; if (defined($line1) && ($line1 != $line2)) { warn("WARNING: function data mismatch at ". "$filename:$line2\n"); next; } $result{$func} = $line2; } return \%result; } # # add_fnccount(fnccount1, fnccount2) # # Add function call count data. Return list (fnccount_added, f_found, f_hit) # sub add_fnccount($$) { my ($fnccount1, $fnccount2) = @_; my %result; my $f_found; my $f_hit; my $function; if (defined($fnccount1)) { %result = %{$fnccount1}; } foreach $function (keys(%{$fnccount2})) { $result{$function} += $fnccount2->{$function}; } $f_found = scalar(keys(%result)); $f_hit = 0; foreach $function (keys(%result)) { if ($result{$function} > 0) { $f_hit++; } } return (\%result, $f_found, $f_hit); } # # add_testfncdata(testfncdata1, testfncdata2) # # Add function call count data for several tests. Return reference to # added_testfncdata. # sub add_testfncdata($$) { my ($testfncdata1, $testfncdata2) = @_; my %result; my $testname; foreach $testname (keys(%{$testfncdata1})) { if (defined($testfncdata2->{$testname})) { my $fnccount; # Function call count data for this testname exists # in both data sets: merge ($fnccount) = add_fnccount( $testfncdata1->{$testname}, $testfncdata2->{$testname}); $result{$testname} = $fnccount; next; } # Function call count data for this testname is unique to # data set 1: copy $result{$testname} = $testfncdata1->{$testname}; } # Add count data for testnames unique to data set 2 foreach $testname (keys(%{$testfncdata2})) { if (!defined($result{$testname})) { $result{$testname} = $testfncdata2->{$testname}; } } return \%result; } # # brcount_to_db(brcount) # # Convert brcount data to the following format: # # db: line number -> block hash # block hash: block number -> branch hash # branch hash: branch number -> taken value # sub brcount_to_db($) { my ($brcount) = @_; my $line; my $db = {}; # Add branches to database foreach $line (keys(%{$brcount})) { my $brdata = $brcount->{$line}; foreach my $entry (split(/:/, $brdata)) { my ($block, $branch, $taken) = split(/,/, $entry); my $old = $db->{$line}->{$block}->{$branch}; if (!defined($old) || $old eq "-") { $old = $taken; } elsif ($taken ne "-") { $old += $taken; } $db->{$line}->{$block}->{$branch} = $old; } } return $db; } # # db_to_brcount(db[, brcount]) # # Convert branch coverage data back to brcount format. If brcount is specified, # the converted data is directly inserted in brcount. # sub db_to_brcount($;$) { my ($db, $brcount) = @_; my $line; my $br_found = 0; my $br_hit = 0; # Convert database back to brcount format foreach $line (sort({$a <=> $b} keys(%{$db}))) { my $ldata = $db->{$line}; my $brdata; my $block; foreach $block (sort({$a <=> $b} keys(%{$ldata}))) { my $bdata = $ldata->{$block}; my $branch; foreach $branch (sort({$a <=> $b} keys(%{$bdata}))) { my $taken = $bdata->{$branch}; $br_found++; $br_hit++ if ($taken ne "-" && $taken > 0); $brdata .= "$block,$branch,$taken:"; } } $brcount->{$line} = $brdata; } return ($brcount, $br_found, $br_hit); } # # brcount_db_combine(db1, db2, op) # # db1 := db1 op db2, where # db1, db2: brcount data as returned by brcount_to_db # op: one of $BR_ADD and BR_SUB # sub brcount_db_combine($$$) { my ($db1, $db2, $op) = @_; foreach my $line (keys(%{$db2})) { my $ldata = $db2->{$line}; foreach my $block (keys(%{$ldata})) { my $bdata = $ldata->{$block}; foreach my $branch (keys(%{$bdata})) { my $taken = $bdata->{$branch}; my $new = $db1->{$line}->{$block}->{$branch}; if (!defined($new) || $new eq "-") { $new = $taken; } elsif ($taken ne "-") { if ($op == $BR_ADD) { $new += $taken; } elsif ($op == $BR_SUB) { $new -= $taken; $new = 0 if ($new < 0); } } $db1->{$line}->{$block}->{$branch} = $new; } } } } # # brcount_db_get_found_and_hit(db) # # Return (br_found, br_hit) for db. # sub brcount_db_get_found_and_hit($) { my ($db) = @_; my ($br_found , $br_hit) = (0, 0); foreach my $line (keys(%{$db})) { my $ldata = $db->{$line}; foreach my $block (keys(%{$ldata})) { my $bdata = $ldata->{$block}; foreach my $branch (keys(%{$bdata})) { my $taken = $bdata->{$branch}; $br_found++; $br_hit++ if ($taken ne "-" && $taken > 0); } } } return ($br_found, $br_hit); } # combine_brcount(brcount1, brcount2, type, inplace) # # If add is BR_ADD, add branch coverage data and return list brcount_added. # If add is BR_SUB, subtract the taken values of brcount2 from brcount1 and # return brcount_sub. If inplace is set, the result is inserted into brcount1. # sub combine_brcount($$$;$) { my ($brcount1, $brcount2, $type, $inplace) = @_; my ($db1, $db2); $db1 = brcount_to_db($brcount1); $db2 = brcount_to_db($brcount2); brcount_db_combine($db1, $db2, $type); return db_to_brcount($db1, $inplace ? $brcount1 : undef); } # # add_testbrdata(testbrdata1, testbrdata2) # # Add branch coverage data for several tests. Return reference to # added_testbrdata. # sub add_testbrdata($$) { my ($testbrdata1, $testbrdata2) = @_; my %result; my $testname; foreach $testname (keys(%{$testbrdata1})) { if (defined($testbrdata2->{$testname})) { my $brcount; # Branch coverage data for this testname exists # in both data sets: add ($brcount) = combine_brcount( $testbrdata1->{$testname}, $testbrdata2->{$testname}, $BR_ADD); $result{$testname} = $brcount; next; } # Branch coverage data for this testname is unique to # data set 1: copy $result{$testname} = $testbrdata1->{$testname}; } # Add count data for testnames unique to data set 2 foreach $testname (keys(%{$testbrdata2})) { if (!defined($result{$testname})) { $result{$testname} = $testbrdata2->{$testname}; } } return \%result; } # # combine_info_entries(entry_ref1, entry_ref2, filename) # # Combine .info data entry hashes referenced by ENTRY_REF1 and ENTRY_REF2. # Return reference to resulting hash. # sub combine_info_entries($$$) { my $entry1 = $_[0]; # Reference to hash containing first entry my $testdata1; my $sumcount1; my $funcdata1; my $checkdata1; my $testfncdata1; my $sumfnccount1; my $testbrdata1; my $sumbrcount1; my $entry2 = $_[1]; # Reference to hash containing second entry my $testdata2; my $sumcount2; my $funcdata2; my $checkdata2; my $testfncdata2; my $sumfnccount2; my $testbrdata2; my $sumbrcount2; my %result; # Hash containing combined entry my %result_testdata; my $result_sumcount = {}; my $result_funcdata; my $result_testfncdata; my $result_sumfnccount; my $result_testbrdata; my $result_sumbrcount; my $lines_found; my $lines_hit; my $f_found; my $f_hit; my $br_found; my $br_hit; my $testname; my $filename = $_[2]; # Retrieve data ($testdata1, $sumcount1, $funcdata1, $checkdata1, $testfncdata1, $sumfnccount1, $testbrdata1, $sumbrcount1) = get_info_entry($entry1); ($testdata2, $sumcount2, $funcdata2, $checkdata2, $testfncdata2, $sumfnccount2, $testbrdata2, $sumbrcount2) = get_info_entry($entry2); # Merge checksums $checkdata1 = merge_checksums($checkdata1, $checkdata2, $filename); # Combine funcdata $result_funcdata = merge_func_data($funcdata1, $funcdata2, $filename); # Combine function call count data $result_testfncdata = add_testfncdata($testfncdata1, $testfncdata2); ($result_sumfnccount, $f_found, $f_hit) = add_fnccount($sumfnccount1, $sumfnccount2); # Combine branch coverage data $result_testbrdata = add_testbrdata($testbrdata1, $testbrdata2); ($result_sumbrcount, $br_found, $br_hit) = combine_brcount($sumbrcount1, $sumbrcount2, $BR_ADD); # Combine testdata foreach $testname (keys(%{$testdata1})) { if (defined($testdata2->{$testname})) { # testname is present in both entries, requires # combination ($result_testdata{$testname}) = add_counts($testdata1->{$testname}, $testdata2->{$testname}); } else { # testname only present in entry1, add to result $result_testdata{$testname} = $testdata1->{$testname}; } # update sum count hash ($result_sumcount, $lines_found, $lines_hit) = add_counts($result_sumcount, $result_testdata{$testname}); } foreach $testname (keys(%{$testdata2})) { # Skip testnames already covered by previous iteration if (defined($testdata1->{$testname})) { next; } # testname only present in entry2, add to result hash $result_testdata{$testname} = $testdata2->{$testname}; # update sum count hash ($result_sumcount, $lines_found, $lines_hit) = add_counts($result_sumcount, $result_testdata{$testname}); } # Calculate resulting sumcount # Store result set_info_entry(\%result, \%result_testdata, $result_sumcount, $result_funcdata, $checkdata1, $result_testfncdata, $result_sumfnccount, $result_testbrdata, $result_sumbrcount, $lines_found, $lines_hit, $f_found, $f_hit, $br_found, $br_hit); return(\%result); } # # combine_info_files(info_ref1, info_ref2) # # Combine .info data in hashes referenced by INFO_REF1 and INFO_REF2. Return # reference to resulting hash. # sub combine_info_files($$) { my %hash1 = %{$_[0]}; my %hash2 = %{$_[1]}; my $filename; foreach $filename (keys(%hash2)) { if ($hash1{$filename}) { # Entry already exists in hash1, combine them $hash1{$filename} = combine_info_entries($hash1{$filename}, $hash2{$filename}, $filename); } else { # Entry is unique in both hashes, simply add to # resulting hash $hash1{$filename} = $hash2{$filename}; } } return(\%hash1); } # # add_traces() # sub add_traces() { my $total_trace; my $current_trace; my $tracefile; my @result; local *INFO_HANDLE; info("Combining tracefiles.\n"); foreach $tracefile (@add_tracefile) { $current_trace = read_info_file($tracefile); if ($total_trace) { $total_trace = combine_info_files($total_trace, $current_trace); } else { $total_trace = $current_trace; } } # Write combined data if (!$data_stdout) { info("Writing data to $output_filename\n"); open(INFO_HANDLE, ">", $output_filename) or die("ERROR: cannot write to $output_filename!\n"); @result = write_info_file(*INFO_HANDLE, $total_trace); close(*INFO_HANDLE); } else { @result = write_info_file(*STDOUT, $total_trace); } return @result; } # # write_info_file(filehandle, data) # sub write_info_file(*$) { local *INFO_HANDLE = $_[0]; my %data = %{$_[1]}; my $source_file; my $entry; my $testdata; my $sumcount; my $funcdata; my $checkdata; my $testfncdata; my $sumfnccount; my $testbrdata; my $sumbrcount; my $testname; my $line; my $func; my $testcount; my $testfnccount; my $testbrcount; my $found; my $hit; my $f_found; my $f_hit; my $br_found; my $br_hit; my $ln_total_found = 0; my $ln_total_hit = 0; my $fn_total_found = 0; my $fn_total_hit = 0; my $br_total_found = 0; my $br_total_hit = 0; foreach $source_file (sort(keys(%data))) { $entry = $data{$source_file}; ($testdata, $sumcount, $funcdata, $checkdata, $testfncdata, $sumfnccount, $testbrdata, $sumbrcount, $found, $hit, $f_found, $f_hit, $br_found, $br_hit) = get_info_entry($entry); # Add to totals $ln_total_found += $found; $ln_total_hit += $hit; $fn_total_found += $f_found; $fn_total_hit += $f_hit; $br_total_found += $br_found; $br_total_hit += $br_hit; foreach $testname (sort(keys(%{$testdata}))) { $testcount = $testdata->{$testname}; $testfnccount = $testfncdata->{$testname}; $testbrcount = $testbrdata->{$testname}; $found = 0; $hit = 0; print(INFO_HANDLE "TN:$testname\n"); print(INFO_HANDLE "SF:$source_file\n"); # Write function related data foreach $func ( sort({$funcdata->{$a} <=> $funcdata->{$b}} keys(%{$funcdata}))) { print(INFO_HANDLE "FN:".$funcdata->{$func}. ",$func\n"); } foreach $func (keys(%{$testfnccount})) { print(INFO_HANDLE "FNDA:". $testfnccount->{$func}. ",$func\n"); } ($f_found, $f_hit) = get_func_found_and_hit($testfnccount); print(INFO_HANDLE "FNF:$f_found\n"); print(INFO_HANDLE "FNH:$f_hit\n"); # Write branch related data $br_found = 0; $br_hit = 0; foreach $line (sort({$a <=> $b} keys(%{$testbrcount}))) { my $brdata = $testbrcount->{$line}; foreach my $brentry (split(/:/, $brdata)) { my ($block, $branch, $taken) = split(/,/, $brentry); print(INFO_HANDLE "BRDA:$line,$block,". "$branch,$taken\n"); $br_found++; $br_hit++ if ($taken ne '-' && $taken > 0); } } if ($br_found > 0) { print(INFO_HANDLE "BRF:$br_found\n"); print(INFO_HANDLE "BRH:$br_hit\n"); } # Write line related data foreach $line (sort({$a <=> $b} keys(%{$testcount}))) { print(INFO_HANDLE "DA:$line,". $testcount->{$line}. (defined($checkdata->{$line}) && $checksum ? ",".$checkdata->{$line} : "")."\n"); $found++; if ($testcount->{$line} > 0) { $hit++; } } print(INFO_HANDLE "LF:$found\n"); print(INFO_HANDLE "LH:$hit\n"); print(INFO_HANDLE "end_of_record\n"); } } return ($ln_total_found, $ln_total_hit, $fn_total_found, $fn_total_hit, $br_total_found, $br_total_hit); } # # transform_pattern(pattern) # # Transform shell wildcard expression to equivalent Perl regular expression. # Return transformed pattern. # sub transform_pattern($) { my $pattern = $_[0]; # Escape special chars $pattern =~ s/\\/\\\\/g; $pattern =~ s/\//\\\//g; $pattern =~ s/\^/\\\^/g; $pattern =~ s/\$/\\\$/g; $pattern =~ s/\(/\\\(/g; $pattern =~ s/\)/\\\)/g; $pattern =~ s/\[/\\\[/g; $pattern =~ s/\]/\\\]/g; $pattern =~ s/\{/\\\{/g; $pattern =~ s/\}/\\\}/g; $pattern =~ s/\./\\\./g; $pattern =~ s/\,/\\\,/g; $pattern =~ s/\|/\\\|/g; $pattern =~ s/\+/\\\+/g; $pattern =~ s/\!/\\\!/g; # Transform ? => (.) and * => (.*) $pattern =~ s/\*/\(\.\*\)/g; $pattern =~ s/\?/\(\.\)/g; return $pattern; } # # extract() # sub extract() { my $data = read_info_file($extract); my $filename; my $keep; my $pattern; my @pattern_list; my $extracted = 0; my @result; local *INFO_HANDLE; # Need perlreg expressions instead of shell pattern @pattern_list = map({ transform_pattern($_); } @ARGV); # Filter out files which do not match any pattern foreach $filename (sort(keys(%{$data}))) { $keep = 0; foreach $pattern (@pattern_list) { $keep ||= ($filename =~ (/^$pattern$/)); } if (!$keep) { delete($data->{$filename}); } else { info("Extracting $filename\n"), $extracted++; } } # Write extracted data if (!$data_stdout) { info("Extracted $extracted files\n"); info("Writing data to $output_filename\n"); open(INFO_HANDLE, ">", $output_filename) or die("ERROR: cannot write to $output_filename!\n"); @result = write_info_file(*INFO_HANDLE, $data); close(*INFO_HANDLE); } else { @result = write_info_file(*STDOUT, $data); } return @result; } # # remove() # sub remove() { my $data = read_info_file($remove); my $filename; my $match_found; my $pattern; my @pattern_list; my $removed = 0; my @result; local *INFO_HANDLE; # Need perlreg expressions instead of shell pattern @pattern_list = map({ transform_pattern($_); } @ARGV); # Filter out files that match the pattern foreach $filename (sort(keys(%{$data}))) { $match_found = 0; foreach $pattern (@pattern_list) { $match_found ||= ($filename =~ (/^$pattern$/)); } if ($match_found) { delete($data->{$filename}); info("Removing $filename\n"), $removed++; } } # Write data if (!$data_stdout) { info("Deleted $removed files\n"); info("Writing data to $output_filename\n"); open(INFO_HANDLE, ">", $output_filename) or die("ERROR: cannot write to $output_filename!\n"); @result = write_info_file(*INFO_HANDLE, $data); close(*INFO_HANDLE); } else { @result = write_info_file(*STDOUT, $data); } return @result; } # get_prefix(max_width, max_percentage_too_long, path_list) # # Return a path prefix that satisfies the following requirements: # - is shared by more paths in path_list than any other prefix # - the percentage of paths which would exceed the given max_width length # after applying the prefix does not exceed max_percentage_too_long # # If multiple prefixes satisfy all requirements, the longest prefix is # returned. Return an empty string if no prefix could be found. sub get_prefix($$@) { my ($max_width, $max_long, @path_list) = @_; my $path; my $ENTRY_NUM = 0; my $ENTRY_LONG = 1; my %prefix; # Build prefix hash foreach $path (@path_list) { my ($v, $d, $f) = splitpath($path); my @dirs = splitdir($d); my $p_len = length($path); my $i; # Remove trailing '/' pop(@dirs) if ($dirs[scalar(@dirs) - 1] eq ''); for ($i = 0; $i < scalar(@dirs); $i++) { my $subpath = catpath($v, catdir(@dirs[0..$i]), ''); my $entry = $prefix{$subpath}; $entry = [ 0, 0 ] if (!defined($entry)); $entry->[$ENTRY_NUM]++; if (($p_len - length($subpath) - 1) > $max_width) { $entry->[$ENTRY_LONG]++; } $prefix{$subpath} = $entry; } } # Find suitable prefix (sort descending by two keys: 1. number of # entries covered by a prefix, 2. length of prefix) foreach $path (sort {($prefix{$a}->[$ENTRY_NUM] == $prefix{$b}->[$ENTRY_NUM]) ? length($b) <=> length($a) : $prefix{$b}->[$ENTRY_NUM] <=> $prefix{$a}->[$ENTRY_NUM]} keys(%prefix)) { my ($num, $long) = @{$prefix{$path}}; # Check for additional requirement: number of filenames # that would be too long may not exceed a certain percentage if ($long <= $num * $max_long / 100) { return $path; } } return ""; } # # shorten_filename(filename, width) # # Truncate filename if it is longer than width characters. # sub shorten_filename($$) { my ($filename, $width) = @_; my $l = length($filename); my $s; my $e; return $filename if ($l <= $width); $e = int(($width - 3) / 2); $s = $width - 3 - $e; return substr($filename, 0, $s).'...'.substr($filename, $l - $e); } sub shorten_number($$) { my ($number, $width) = @_; my $result = sprintf("%*d", $width, $number); return $result if (length($result) <= $width); $number = $number / 1000; return $result if (length($result) <= $width); $result = sprintf("%*dk", $width - 1, $number); return $result if (length($result) <= $width); $number = $number / 1000; $result = sprintf("%*dM", $width - 1, $number); return $result if (length($result) <= $width); return '#'; } sub shorten_rate($$$) { my ($hit, $found, $width) = @_; my $result = rate($hit, $found, "%", 1, $width); return $result if (length($result) <= $width); $result = rate($hit, $found, "%", 0, $width); return $result if (length($result) <= $width); return "#"; } # # list() # sub list() { my $data = read_info_file($list); my $filename; my $found; my $hit; my $entry; my $fn_found; my $fn_hit; my $br_found; my $br_hit; my $total_found = 0; my $total_hit = 0; my $fn_total_found = 0; my $fn_total_hit = 0; my $br_total_found = 0; my $br_total_hit = 0; my $prefix; my $strlen = length("Filename"); my $format; my $heading1; my $heading2; my @footer; my $barlen; my $rate; my $fnrate; my $brrate; my $lastpath; my $F_LN_NUM = 0; my $F_LN_RATE = 1; my $F_FN_NUM = 2; my $F_FN_RATE = 3; my $F_BR_NUM = 4; my $F_BR_RATE = 5; my @fwidth_narrow = (5, 5, 3, 5, 4, 5); my @fwidth_wide = (6, 5, 5, 5, 6, 5); my @fwidth = @fwidth_wide; my $w; my $max_width = $opt_list_width; my $max_long = $opt_list_truncate_max; my $fwidth_narrow_length; my $fwidth_wide_length; my $got_prefix = 0; my $root_prefix = 0; # Calculate total width of narrow fields $fwidth_narrow_length = 0; foreach $w (@fwidth_narrow) { $fwidth_narrow_length += $w + 1; } # Calculate total width of wide fields $fwidth_wide_length = 0; foreach $w (@fwidth_wide) { $fwidth_wide_length += $w + 1; } # Get common file path prefix $prefix = get_prefix($max_width - $fwidth_narrow_length, $max_long, keys(%{$data})); $root_prefix = 1 if ($prefix eq rootdir()); $got_prefix = 1 if (length($prefix) > 0); $prefix =~ s/\/$//; # Get longest filename length foreach $filename (keys(%{$data})) { if (!$opt_list_full_path) { if (!$got_prefix || !$root_prefix && !($filename =~ s/^\Q$prefix\/\E//)) { my ($v, $d, $f) = splitpath($filename); $filename = $f; } } # Determine maximum length of entries if (length($filename) > $strlen) { $strlen = length($filename) } } if (!$opt_list_full_path) { my $blanks; $w = $fwidth_wide_length; # Check if all columns fit into max_width characters if ($strlen + $fwidth_wide_length > $max_width) { # Use narrow fields @fwidth = @fwidth_narrow; $w = $fwidth_narrow_length; if (($strlen + $fwidth_narrow_length) > $max_width) { # Truncate filenames at max width $strlen = $max_width - $fwidth_narrow_length; } } # Add some blanks between filename and fields if possible $blanks = int($strlen * 0.5); $blanks = 4 if ($blanks < 4); $blanks = 8 if ($blanks > 8); if (($strlen + $w + $blanks) < $max_width) { $strlen += $blanks; } else { $strlen = $max_width - $w; } } # Filename $w = $strlen; $format = "%-${w}s|"; $heading1 = sprintf("%*s|", $w, ""); $heading2 = sprintf("%-*s|", $w, "Filename"); $barlen = $w + 1; # Line coverage rate $w = $fwidth[$F_LN_RATE]; $format .= "%${w}s "; $heading1 .= sprintf("%-*s |", $w + $fwidth[$F_LN_NUM], "Lines"); $heading2 .= sprintf("%-*s ", $w, "Rate"); $barlen += $w + 1; # Number of lines $w = $fwidth[$F_LN_NUM]; $format .= "%${w}s|"; $heading2 .= sprintf("%*s|", $w, "Num"); $barlen += $w + 1; # Function coverage rate $w = $fwidth[$F_FN_RATE]; $format .= "%${w}s "; $heading1 .= sprintf("%-*s|", $w + $fwidth[$F_FN_NUM] + 1, "Functions"); $heading2 .= sprintf("%-*s ", $w, "Rate"); $barlen += $w + 1; # Number of functions $w = $fwidth[$F_FN_NUM]; $format .= "%${w}s|"; $heading2 .= sprintf("%*s|", $w, "Num"); $barlen += $w + 1; # Branch coverage rate $w = $fwidth[$F_BR_RATE]; $format .= "%${w}s "; $heading1 .= sprintf("%-*s", $w + $fwidth[$F_BR_NUM] + 1, "Branches"); $heading2 .= sprintf("%-*s ", $w, "Rate"); $barlen += $w + 1; # Number of branches $w = $fwidth[$F_BR_NUM]; $format .= "%${w}s"; $heading2 .= sprintf("%*s", $w, "Num"); $barlen += $w; # Line end $format .= "\n"; $heading1 .= "\n"; $heading2 .= "\n"; # Print heading print($heading1); print($heading2); print(("="x$barlen)."\n"); # Print per file information foreach $filename (sort(keys(%{$data}))) { my @file_data; my $print_filename = $filename; $entry = $data->{$filename}; if (!$opt_list_full_path) { my $p; $print_filename = $filename; if (!$got_prefix || !$root_prefix && !($print_filename =~ s/^\Q$prefix\/\E//)) { my ($v, $d, $f) = splitpath($filename); $p = catpath($v, $d, ""); $p =~ s/\/$//; $print_filename = $f; } else { $p = $prefix; } if (!defined($lastpath) || $lastpath ne $p) { print("\n") if (defined($lastpath)); $lastpath = $p; print("[$lastpath/]\n") if (!$root_prefix); } $print_filename = shorten_filename($print_filename, $strlen); } (undef, undef, undef, undef, undef, undef, undef, undef, $found, $hit, $fn_found, $fn_hit, $br_found, $br_hit) = get_info_entry($entry); # Assume zero count if there is no function data for this file if (!defined($fn_found) || !defined($fn_hit)) { $fn_found = 0; $fn_hit = 0; } # Assume zero count if there is no branch data for this file if (!defined($br_found) || !defined($br_hit)) { $br_found = 0; $br_hit = 0; } # Add line coverage totals $total_found += $found; $total_hit += $hit; # Add function coverage totals $fn_total_found += $fn_found; $fn_total_hit += $fn_hit; # Add branch coverage totals $br_total_found += $br_found; $br_total_hit += $br_hit; # Determine line coverage rate for this file $rate = shorten_rate($hit, $found, $fwidth[$F_LN_RATE]); # Determine function coverage rate for this file $fnrate = shorten_rate($fn_hit, $fn_found, $fwidth[$F_FN_RATE]); # Determine branch coverage rate for this file $brrate = shorten_rate($br_hit, $br_found, $fwidth[$F_BR_RATE]); # Assemble line parameters push(@file_data, $print_filename); push(@file_data, $rate); push(@file_data, shorten_number($found, $fwidth[$F_LN_NUM])); push(@file_data, $fnrate); push(@file_data, shorten_number($fn_found, $fwidth[$F_FN_NUM])); push(@file_data, $brrate); push(@file_data, shorten_number($br_found, $fwidth[$F_BR_NUM])); # Print assembled line printf($format, @file_data); } # Determine total line coverage rate $rate = shorten_rate($total_hit, $total_found, $fwidth[$F_LN_RATE]); # Determine total function coverage rate $fnrate = shorten_rate($fn_total_hit, $fn_total_found, $fwidth[$F_FN_RATE]); # Determine total branch coverage rate $brrate = shorten_rate($br_total_hit, $br_total_found, $fwidth[$F_BR_RATE]); # Print separator print(("="x$barlen)."\n"); # Assemble line parameters push(@footer, sprintf("%*s", $strlen, "Total:")); push(@footer, $rate); push(@footer, shorten_number($total_found, $fwidth[$F_LN_NUM])); push(@footer, $fnrate); push(@footer, shorten_number($fn_total_found, $fwidth[$F_FN_NUM])); push(@footer, $brrate); push(@footer, shorten_number($br_total_found, $fwidth[$F_BR_NUM])); # Print assembled line printf($format, @footer); } # # get_common_filename(filename1, filename2) # # Check for filename components which are common to FILENAME1 and FILENAME2. # Upon success, return # # (common, path1, path2) # # or 'undef' in case there are no such parts. # sub get_common_filename($$) { my @list1 = split("/", $_[0]); my @list2 = split("/", $_[1]); my @result; # Work in reverse order, i.e. beginning with the filename itself while (@list1 && @list2 && ($list1[$#list1] eq $list2[$#list2])) { unshift(@result, pop(@list1)); pop(@list2); } # Did we find any similarities? if (scalar(@result) > 0) { return (join("/", @result), join("/", @list1), join("/", @list2)); } else { return undef; } } # # strip_directories($path, $depth) # # Remove DEPTH leading directory levels from PATH. # sub strip_directories($$) { my $filename = $_[0]; my $depth = $_[1]; my $i; if (!defined($depth) || ($depth < 1)) { return $filename; } for ($i = 0; $i < $depth; $i++) { $filename =~ s/^[^\/]*\/+(.*)$/$1/; } return $filename; } # # read_diff(filename) # # Read diff output from FILENAME to memory. The diff file has to follow the # format generated by 'diff -u'. Returns a list of hash references: # # (mapping, path mapping) # # mapping: filename -> reference to line hash # line hash: line number in new file -> corresponding line number in old file # # path mapping: filename -> old filename # # Die in case of error. # sub read_diff($) { my $diff_file = $_[0]; # Name of diff file my %diff; # Resulting mapping filename -> line hash my %paths; # Resulting mapping old path -> new path my $mapping; # Reference to current line hash my $line; # Contents of current line my $num_old; # Current line number in old file my $num_new; # Current line number in new file my $file_old; # Name of old file in diff section my $file_new; # Name of new file in diff section my $filename; # Name of common filename of diff section my $in_block = 0; # Non-zero while we are inside a diff block local *HANDLE; # File handle for reading the diff file info("Reading diff $diff_file\n"); # Check if file exists and is readable stat($diff_file); if (!(-r _)) { die("ERROR: cannot read file $diff_file!\n"); } # Check if this is really a plain file if (!(-f _)) { die("ERROR: not a plain file: $diff_file!\n"); } # Check for .gz extension if ($diff_file =~ /\.gz$/) { # Check for availability of GZIP tool system_no_output(1, "gunzip", "-h") and die("ERROR: gunzip command not available!\n"); # Check integrity of compressed file system_no_output(1, "gunzip", "-t", $diff_file) and die("ERROR: integrity check failed for ". "compressed file $diff_file!\n"); # Open compressed file open(HANDLE, "-|", "gunzip -c '$diff_file'") or die("ERROR: cannot start gunzip to decompress ". "file $_[0]!\n"); } else { # Open decompressed file open(HANDLE, "<", $diff_file) or die("ERROR: cannot read file $_[0]!\n"); } # Parse diff file line by line while () { chomp($_); $line = $_; foreach ($line) { # Filename of old file: # --- /^--- (\S+)/ && do { $file_old = strip_directories($1, $strip); last; }; # Filename of new file: # +++ /^\+\+\+ (\S+)/ && do { # Add last file to resulting hash if ($filename) { my %new_hash; $diff{$filename} = $mapping; $mapping = \%new_hash; } $file_new = strip_directories($1, $strip); $filename = $file_old; $paths{$filename} = $file_new; $num_old = 1; $num_new = 1; last; }; # Start of diff block: # @@ -old_start,old_num, +new_start,new_num @@ /^\@\@\s+-(\d+),(\d+)\s+\+(\d+),(\d+)\s+\@\@$/ && do { $in_block = 1; while ($num_old < $1) { $mapping->{$num_new} = $num_old; $num_old++; $num_new++; } last; }; # Unchanged line # /^ / && do { if ($in_block == 0) { last; } $mapping->{$num_new} = $num_old; $num_old++; $num_new++; last; }; # Line as seen in old file # /^-/ && do { if ($in_block == 0) { last; } $num_old++; last; }; # Line as seen in new file # /^\+/ && do { if ($in_block == 0) { last; } $num_new++; last; }; # Empty line /^$/ && do { if ($in_block == 0) { last; } $mapping->{$num_new} = $num_old; $num_old++; $num_new++; last; }; } } close(HANDLE); # Add final diff file section to resulting hash if ($filename) { $diff{$filename} = $mapping; } if (!%diff) { die("ERROR: no valid diff data found in $diff_file!\n". "Make sure to use 'diff -u' when generating the diff ". "file.\n"); } return (\%diff, \%paths); } # # apply_diff($count_data, $line_hash) # # Transform count data using a mapping of lines: # # $count_data: reference to hash: line number -> data # $line_hash: reference to hash: line number new -> line number old # # Return a reference to transformed count data. # sub apply_diff($$) { my $count_data = $_[0]; # Reference to data hash: line -> hash my $line_hash = $_[1]; # Reference to line hash: new line -> old line my %result; # Resulting hash my $last_new = 0; # Last new line number found in line hash my $last_old = 0; # Last old line number found in line hash # Iterate all new line numbers found in the diff foreach (sort({$a <=> $b} keys(%{$line_hash}))) { $last_new = $_; $last_old = $line_hash->{$last_new}; # Is there data associated with the corresponding old line? if (defined($count_data->{$line_hash->{$_}})) { # Copy data to new hash with a new line number $result{$_} = $count_data->{$line_hash->{$_}}; } } # Transform all other lines which come after the last diff entry foreach (sort({$a <=> $b} keys(%{$count_data}))) { if ($_ <= $last_old) { # Skip lines which were covered by line hash next; } # Copy data to new hash with an offset $result{$_ + ($last_new - $last_old)} = $count_data->{$_}; } return \%result; } # # apply_diff_to_brcount(brcount, linedata) # # Adjust line numbers of branch coverage data according to linedata. # sub apply_diff_to_brcount($$) { my ($brcount, $linedata) = @_; my $db; # Convert brcount to db format $db = brcount_to_db($brcount); # Apply diff to db format $db = apply_diff($db, $linedata); # Convert db format back to brcount format ($brcount) = db_to_brcount($db); return $brcount; } # # get_hash_max(hash_ref) # # Return the highest integer key from hash. # sub get_hash_max($) { my ($hash) = @_; my $max; foreach (keys(%{$hash})) { if (!defined($max)) { $max = $_; } elsif ($hash->{$_} > $max) { $max = $_; } } return $max; } sub get_hash_reverse($) { my ($hash) = @_; my %result; foreach (keys(%{$hash})) { $result{$hash->{$_}} = $_; } return \%result; } # # apply_diff_to_funcdata(funcdata, line_hash) # sub apply_diff_to_funcdata($$) { my ($funcdata, $linedata) = @_; my $last_new = get_hash_max($linedata); my $last_old = $linedata->{$last_new}; my $func; my %result; my $line_diff = get_hash_reverse($linedata); foreach $func (keys(%{$funcdata})) { my $line = $funcdata->{$func}; if (defined($line_diff->{$line})) { $result{$func} = $line_diff->{$line}; } elsif ($line > $last_old) { $result{$func} = $line + $last_new - $last_old; } } return \%result; } # # get_line_hash($filename, $diff_data, $path_data) # # Find line hash in DIFF_DATA which matches FILENAME. On success, return list # line hash. or undef in case of no match. Die if more than one line hashes in # DIFF_DATA match. # sub get_line_hash($$$) { my $filename = $_[0]; my $diff_data = $_[1]; my $path_data = $_[2]; my $conversion; my $old_path; my $new_path; my $diff_name; my $common; my $old_depth; my $new_depth; # Remove trailing slash from diff path $diff_path =~ s/\/$//; foreach (keys(%{$diff_data})) { my $sep = ""; $sep = '/' if (!/^\//); # Try to match diff filename with filename if ($filename =~ /^\Q$diff_path$sep$_\E$/) { if ($diff_name) { # Two files match, choose the more specific one # (the one with more path components) $old_depth = ($diff_name =~ tr/\///); $new_depth = (tr/\///); if ($old_depth == $new_depth) { die("ERROR: diff file contains ". "ambiguous entries for ". "$filename\n"); } elsif ($new_depth > $old_depth) { $diff_name = $_; } } else { $diff_name = $_; } }; } if ($diff_name) { # Get converted path if ($filename =~ /^(.*)$diff_name$/) { ($common, $old_path, $new_path) = get_common_filename($filename, $1.$path_data->{$diff_name}); } return ($diff_data->{$diff_name}, $old_path, $new_path); } else { return undef; } } # # convert_paths(trace_data, path_conversion_data) # # Rename all paths in TRACE_DATA which show up in PATH_CONVERSION_DATA. # sub convert_paths($$) { my $trace_data = $_[0]; my $path_conversion_data = $_[1]; my $filename; my $new_path; if (scalar(keys(%{$path_conversion_data})) == 0) { info("No path conversion data available.\n"); return; } # Expand path conversion list foreach $filename (keys(%{$path_conversion_data})) { $new_path = $path_conversion_data->{$filename}; while (($filename =~ s/^(.*)\/[^\/]+$/$1/) && ($new_path =~ s/^(.*)\/[^\/]+$/$1/) && ($filename ne $new_path)) { $path_conversion_data->{$filename} = $new_path; } } # Adjust paths FILENAME: foreach $filename (keys(%{$trace_data})) { # Find a path in our conversion table that matches, starting # with the longest path foreach (sort({length($b) <=> length($a)} keys(%{$path_conversion_data}))) { # Is this path a prefix of our filename? if (!($filename =~ /^$_(.*)$/)) { next; } $new_path = $path_conversion_data->{$_}.$1; # Make sure not to overwrite an existing entry under # that path name if ($trace_data->{$new_path}) { # Need to combine entries $trace_data->{$new_path} = combine_info_entries( $trace_data->{$filename}, $trace_data->{$new_path}, $filename); } else { # Simply rename entry $trace_data->{$new_path} = $trace_data->{$filename}; } delete($trace_data->{$filename}); next FILENAME; } info("No conversion available for filename $filename\n"); } } # # sub adjust_fncdata(funcdata, testfncdata, sumfnccount) # # Remove function call count data from testfncdata and sumfnccount which # is no longer present in funcdata. # sub adjust_fncdata($$$) { my ($funcdata, $testfncdata, $sumfnccount) = @_; my $testname; my $func; my $f_found; my $f_hit; # Remove count data in testfncdata for functions which are no longer # in funcdata foreach $testname (keys(%{$testfncdata})) { my $fnccount = $testfncdata->{$testname}; foreach $func (keys(%{$fnccount})) { if (!defined($funcdata->{$func})) { delete($fnccount->{$func}); } } } # Remove count data in sumfnccount for functions which are no longer # in funcdata foreach $func (keys(%{$sumfnccount})) { if (!defined($funcdata->{$func})) { delete($sumfnccount->{$func}); } } } # # get_func_found_and_hit(sumfnccount) # # Return (f_found, f_hit) for sumfnccount # sub get_func_found_and_hit($) { my ($sumfnccount) = @_; my $function; my $f_found; my $f_hit; $f_found = scalar(keys(%{$sumfnccount})); $f_hit = 0; foreach $function (keys(%{$sumfnccount})) { if ($sumfnccount->{$function} > 0) { $f_hit++; } } return ($f_found, $f_hit); } # # diff() # sub diff() { my $trace_data = read_info_file($diff); my $diff_data; my $path_data; my $old_path; my $new_path; my %path_conversion_data; my $filename; my $line_hash; my $new_name; my $entry; my $testdata; my $testname; my $sumcount; my $funcdata; my $checkdata; my $testfncdata; my $sumfnccount; my $testbrdata; my $sumbrcount; my $found; my $hit; my $f_found; my $f_hit; my $br_found; my $br_hit; my $converted = 0; my $unchanged = 0; my @result; local *INFO_HANDLE; ($diff_data, $path_data) = read_diff($ARGV[0]); foreach $filename (sort(keys(%{$trace_data}))) { # Find a diff section corresponding to this file ($line_hash, $old_path, $new_path) = get_line_hash($filename, $diff_data, $path_data); if (!$line_hash) { # There's no diff section for this file $unchanged++; next; } $converted++; if ($old_path && $new_path && ($old_path ne $new_path)) { $path_conversion_data{$old_path} = $new_path; } # Check for deleted files if (scalar(keys(%{$line_hash})) == 0) { info("Removing $filename\n"); delete($trace_data->{$filename}); next; } info("Converting $filename\n"); $entry = $trace_data->{$filename}; ($testdata, $sumcount, $funcdata, $checkdata, $testfncdata, $sumfnccount, $testbrdata, $sumbrcount) = get_info_entry($entry); # Convert test data foreach $testname (keys(%{$testdata})) { # Adjust line numbers of line coverage data $testdata->{$testname} = apply_diff($testdata->{$testname}, $line_hash); # Adjust line numbers of branch coverage data $testbrdata->{$testname} = apply_diff_to_brcount($testbrdata->{$testname}, $line_hash); # Remove empty sets of test data if (scalar(keys(%{$testdata->{$testname}})) == 0) { delete($testdata->{$testname}); delete($testfncdata->{$testname}); delete($testbrdata->{$testname}); } } # Rename test data to indicate conversion foreach $testname (keys(%{$testdata})) { # Skip testnames which already contain an extension if ($testname =~ /,[^,]+$/) { next; } # Check for name conflict if (defined($testdata->{$testname.",diff"})) { # Add counts ($testdata->{$testname}) = add_counts( $testdata->{$testname}, $testdata->{$testname.",diff"}); delete($testdata->{$testname.",diff"}); # Add function call counts ($testfncdata->{$testname}) = add_fnccount( $testfncdata->{$testname}, $testfncdata->{$testname.",diff"}); delete($testfncdata->{$testname.",diff"}); # Add branch counts combine_brcount( $testbrdata->{$testname}, $testbrdata->{$testname.",diff"}, $BR_ADD, 1); delete($testbrdata->{$testname.",diff"}); } # Move test data to new testname $testdata->{$testname.",diff"} = $testdata->{$testname}; delete($testdata->{$testname}); # Move function call count data to new testname $testfncdata->{$testname.",diff"} = $testfncdata->{$testname}; delete($testfncdata->{$testname}); # Move branch count data to new testname $testbrdata->{$testname.",diff"} = $testbrdata->{$testname}; delete($testbrdata->{$testname}); } # Convert summary of test data $sumcount = apply_diff($sumcount, $line_hash); # Convert function data $funcdata = apply_diff_to_funcdata($funcdata, $line_hash); # Convert branch coverage data $sumbrcount = apply_diff_to_brcount($sumbrcount, $line_hash); # Update found/hit numbers # Convert checksum data $checkdata = apply_diff($checkdata, $line_hash); # Convert function call count data adjust_fncdata($funcdata, $testfncdata, $sumfnccount); ($f_found, $f_hit) = get_func_found_and_hit($sumfnccount); ($br_found, $br_hit) = get_br_found_and_hit($sumbrcount); # Update found/hit numbers $found = 0; $hit = 0; foreach (keys(%{$sumcount})) { $found++; if ($sumcount->{$_} > 0) { $hit++; } } if ($found > 0) { # Store converted entry set_info_entry($entry, $testdata, $sumcount, $funcdata, $checkdata, $testfncdata, $sumfnccount, $testbrdata, $sumbrcount, $found, $hit, $f_found, $f_hit, $br_found, $br_hit); } else { # Remove empty data set delete($trace_data->{$filename}); } } # Convert filenames as well if requested if ($convert_filenames) { convert_paths($trace_data, \%path_conversion_data); } info("$converted entr".($converted != 1 ? "ies" : "y")." converted, ". "$unchanged entr".($unchanged != 1 ? "ies" : "y")." left ". "unchanged.\n"); # Write data if (!$data_stdout) { info("Writing data to $output_filename\n"); open(INFO_HANDLE, ">", $output_filename) or die("ERROR: cannot write to $output_filename!\n"); @result = write_info_file(*INFO_HANDLE, $trace_data); close(*INFO_HANDLE); } else { @result = write_info_file(*STDOUT, $trace_data); } return @result; } # # summary() # sub summary() { my $filename; my $current; my $total; my $ln_total_found; my $ln_total_hit; my $fn_total_found; my $fn_total_hit; my $br_total_found; my $br_total_hit; # Read and combine trace files foreach $filename (@opt_summary) { $current = read_info_file($filename); if (!defined($total)) { $total = $current; } else { $total = combine_info_files($total, $current); } } # Calculate coverage data foreach $filename (keys(%{$total})) { my $entry = $total->{$filename}; my $ln_found; my $ln_hit; my $fn_found; my $fn_hit; my $br_found; my $br_hit; (undef, undef, undef, undef, undef, undef, undef, undef, $ln_found, $ln_hit, $fn_found, $fn_hit, $br_found, $br_hit) = get_info_entry($entry); # Add to totals $ln_total_found += $ln_found; $ln_total_hit += $ln_hit; $fn_total_found += $fn_found; $fn_total_hit += $fn_hit; $br_total_found += $br_found; $br_total_hit += $br_hit; } return ($ln_total_found, $ln_total_hit, $fn_total_found, $fn_total_hit, $br_total_found, $br_total_hit); } # # system_no_output(mode, parameters) # # Call an external program using PARAMETERS while suppressing depending on # the value of MODE: # # MODE & 1: suppress STDOUT # MODE & 2: suppress STDERR # # Return 0 on success, non-zero otherwise. # sub system_no_output($@) { my $mode = shift; my $result; local *OLD_STDERR; local *OLD_STDOUT; # Save old stdout and stderr handles ($mode & 1) && open(OLD_STDOUT, ">>&", "STDOUT"); ($mode & 2) && open(OLD_STDERR, ">>&", "STDERR"); # Redirect to /dev/null ($mode & 1) && open(STDOUT, ">", "/dev/null"); ($mode & 2) && open(STDERR, ">", "/dev/null"); system(@_); $result = $?; # Close redirected handles ($mode & 1) && close(STDOUT); ($mode & 2) && close(STDERR); # Restore old handles ($mode & 1) && open(STDOUT, ">>&", "OLD_STDOUT"); ($mode & 2) && open(STDERR, ">>&", "OLD_STDERR"); return $result; } # # read_config(filename) # # Read configuration file FILENAME and return a reference to a hash containing # all valid key=value pairs found. # sub read_config($) { my $filename = $_[0]; my %result; my $key; my $value; local *HANDLE; if (!open(HANDLE, "<", $filename)) { warn("WARNING: cannot read configuration file $filename\n"); return undef; } while () { chomp; # Skip comments s/#.*//; # Remove leading blanks s/^\s+//; # Remove trailing blanks s/\s+$//; next unless length; ($key, $value) = split(/\s*=\s*/, $_, 2); if (defined($key) && defined($value)) { $result{$key} = $value; } else { warn("WARNING: malformed statement in line $. ". "of configuration file $filename\n"); } } close(HANDLE); return \%result; } # # apply_config(REF) # # REF is a reference to a hash containing the following mapping: # # key_string => var_ref # # where KEY_STRING is a keyword and VAR_REF is a reference to an associated # variable. If the global configuration hashes CONFIG or OPT_RC contain a value # for keyword KEY_STRING, VAR_REF will be assigned the value for that keyword. # sub apply_config($) { my $ref = $_[0]; foreach (keys(%{$ref})) { if (defined($opt_rc{$_})) { ${$ref->{$_}} = $opt_rc{$_}; } elsif (defined($config->{$_})) { ${$ref->{$_}} = $config->{$_}; } } } sub warn_handler($) { my ($msg) = @_; warn("$tool_name: $msg"); } sub die_handler($) { my ($msg) = @_; temp_cleanup(); die("$tool_name: $msg"); } sub abort_handler($) { temp_cleanup(); exit(1); } sub temp_cleanup() { # Ensure temp directory is not in use by current process chdir("/"); if (@temp_dirs) { info("Removing temporary directories.\n"); foreach (@temp_dirs) { rmtree($_); } @temp_dirs = (); } } sub setup_gkv_sys() { system_no_output(3, "mount", "-t", "debugfs", "nodev", "/sys/kernel/debug"); } sub setup_gkv_proc() { if (system_no_output(3, "modprobe", "gcov_proc")) { system_no_output(3, "modprobe", "gcov_prof"); } } sub check_gkv_sys($) { my ($dir) = @_; if (-e "$dir/reset") { return 1; } return 0; } sub check_gkv_proc($) { my ($dir) = @_; if (-e "$dir/vmlinux") { return 1; } return 0; } sub setup_gkv() { my $dir; my $sys_dir = "/sys/kernel/debug/gcov"; my $proc_dir = "/proc/gcov"; my @todo; if (!defined($gcov_dir)) { info("Auto-detecting gcov kernel support.\n"); @todo = ( "cs", "cp", "ss", "cs", "sp", "cp" ); } elsif ($gcov_dir =~ /proc/) { info("Checking gcov kernel support at $gcov_dir ". "(user-specified).\n"); @todo = ( "cp", "sp", "cp", "cs", "ss", "cs"); } else { info("Checking gcov kernel support at $gcov_dir ". "(user-specified).\n"); @todo = ( "cs", "ss", "cs", "cp", "sp", "cp", ); } foreach (@todo) { if ($_ eq "cs") { # Check /sys $dir = defined($gcov_dir) ? $gcov_dir : $sys_dir; if (check_gkv_sys($dir)) { info("Found ".$GKV_NAME[$GKV_SYS]." gcov ". "kernel support at $dir\n"); return ($GKV_SYS, $dir); } } elsif ($_ eq "cp") { # Check /proc $dir = defined($gcov_dir) ? $gcov_dir : $proc_dir; if (check_gkv_proc($dir)) { info("Found ".$GKV_NAME[$GKV_PROC]." gcov ". "kernel support at $dir\n"); return ($GKV_PROC, $dir); } } elsif ($_ eq "ss") { # Setup /sys setup_gkv_sys(); } elsif ($_ eq "sp") { # Setup /proc setup_gkv_proc(); } } if (defined($gcov_dir)) { die("ERROR: could not find gcov kernel data at $gcov_dir\n"); } else { die("ERROR: no gcov kernel data found\n"); } } # # get_overall_line(found, hit, name_singular, name_plural) # # Return a string containing overall information for the specified # found/hit data. # sub get_overall_line($$$$) { my ($found, $hit, $name_sn, $name_pl) = @_; my $name; return "no data found" if (!defined($found) || $found == 0); $name = ($found == 1) ? $name_sn : $name_pl; return rate($hit, $found, "% ($hit of $found $name)"); } # # print_overall_rate(ln_do, ln_found, ln_hit, fn_do, fn_found, fn_hit, br_do # br_found, br_hit) # # Print overall coverage rates for the specified coverage types. # sub print_overall_rate($$$$$$$$$) { my ($ln_do, $ln_found, $ln_hit, $fn_do, $fn_found, $fn_hit, $br_do, $br_found, $br_hit) = @_; info("Summary coverage rate:\n"); info(" lines......: %s\n", get_overall_line($ln_found, $ln_hit, "line", "lines")) if ($ln_do); info(" functions..: %s\n", get_overall_line($fn_found, $fn_hit, "function", "functions")) if ($fn_do); info(" branches...: %s\n", get_overall_line($br_found, $br_hit, "branch", "branches")) if ($br_do); } # # rate(hit, found[, suffix, precision, width]) # # Return the coverage rate [0..100] for HIT and FOUND values. 0 is only # returned when HIT is 0. 100 is only returned when HIT equals FOUND. # PRECISION specifies the precision of the result. SUFFIX defines a # string that is appended to the result if FOUND is non-zero. Spaces # are added to the start of the resulting string until it is at least WIDTH # characters wide. # sub rate($$;$$$) { my ($hit, $found, $suffix, $precision, $width) = @_; my $rate; # Assign defaults if necessary $precision = 1 if (!defined($precision)); $suffix = "" if (!defined($suffix)); $width = 0 if (!defined($width)); return sprintf("%*s", $width, "-") if (!defined($found) || $found == 0); $rate = sprintf("%.*f", $precision, $hit * 100 / $found); # Adjust rates if necessary if ($rate == 0 && $hit > 0) { $rate = sprintf("%.*f", $precision, 1 / 10 ** $precision); } elsif ($rate == 100 && $hit != $found) { $rate = sprintf("%.*f", $precision, 100 - 1 / 10 ** $precision); } return sprintf("%*s", $width, $rate.$suffix); } lcov-1.14/bin/genpng0000755000000000000000000002272513435751076013077 0ustar rootroot#!/usr/bin/env perl # # Copyright (c) International Business Machines Corp., 2002 # # This program is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 2 of the License, or (at # your option) any later version. # # This program is distributed in the hope that it will be useful, but # WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU # General Public License for more details. # # You should have received a copy of the GNU General Public License # along with this program; if not, write to the Free Software # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA # # # genpng # # This script creates an overview PNG image of a source code file by # representing each source code character by a single pixel. # # Note that the Perl module GD.pm is required for this script to work. # It may be obtained from http://www.cpan.org # # History: # 2002-08-26: created by Peter Oberparleiter # use strict; use warnings; use File::Basename; use Getopt::Long; use Cwd qw/abs_path/; # Constants our $tool_dir = abs_path(dirname($0)); our $lcov_version = "LCOV version 1.14"; our $lcov_url = "http://ltp.sourceforge.net/coverage/lcov.php"; our $tool_name = basename($0); # Prototypes sub gen_png($$$@); sub check_and_load_module($); sub genpng_print_usage(*); sub genpng_process_file($$$$); sub genpng_warn_handler($); sub genpng_die_handler($); # # Code entry point # # Check whether required module GD.pm is installed if (check_and_load_module("GD")) { # Note: cannot use die() to print this message because inserting this # code into another script via do() would not fail as required! print(STDERR < \$tab_size, "width=i" => \$width, "output-filename=s" => \$out_filename, "help" => \$help, "version" => \$version)) { print(STDERR "Use $tool_name --help to get usage ". "information\n"); exit(1); } $filename = $ARGV[0]; # Check for help flag if ($help) { genpng_print_usage(*STDOUT); exit(0); } # Check for version flag if ($version) { print("$tool_name: $lcov_version\n"); exit(0); } # Check options if (!$filename) { die("No filename specified\n"); } # Check for output filename if (!$out_filename) { $out_filename = "$filename.png"; } genpng_process_file($filename, $out_filename, $width, $tab_size); exit(0); } # # genpng_print_usage(handle) # # Write out command line usage information to given filehandle. # sub genpng_print_usage(*) { local *HANDLE = $_[0]; print(HANDLE <) { if (/^\t\t(.*)$/) { # Uninstrumented line push(@source, ":$1"); } elsif (/^ ###### (.*)$/) { # Line with zero execution count push(@source, "0:$1"); } elsif (/^( *)(\d*) (.*)$/) { # Line with positive execution count push(@source, "$2:$3"); } } } else { # Plain text file while () { push(@source, ":$_"); } } close(HANDLE); gen_png($out_filename, $width, $tab_size, @source); } # # gen_png(filename, width, tab_size, source) # # Write an overview PNG file to FILENAME. Source code is defined by SOURCE # which is a list of lines : per source code line. # The output image will be made up of one pixel per character of source, # coloring will be done according to execution counts. WIDTH defines the # image width. TAB_SIZE specifies the number of spaces to use as replacement # string for tabulator signs in source code text. # # Die on error. # sub gen_png($$$@) { my $filename = shift(@_); # Filename for PNG file my $overview_width = shift(@_); # Imagewidth for image my $tab_size = shift(@_); # Replacement string for tab signs my @source = @_; # Source code as passed via argument 2 my $height; # Height as define by source size my $overview; # Source code overview image data my $col_plain_back; # Color for overview background my $col_plain_text; # Color for uninstrumented text my $col_cov_back; # Color for background of covered lines my $col_cov_text; # Color for text of covered lines my $col_nocov_back; # Color for background of lines which # were not covered (count == 0) my $col_nocov_text; # Color for test of lines which were not # covered (count == 0) my $col_hi_back; # Color for background of highlighted lines my $col_hi_text; # Color for text of highlighted lines my $line; # Current line during iteration my $row = 0; # Current row number during iteration my $column; # Current column number during iteration my $color_text; # Current text color during iteration my $color_back; # Current background color during iteration my $last_count; # Count of last processed line my $count; # Count of current line my $source; # Source code of current line my $replacement; # Replacement string for tabulator chars local *PNG_HANDLE; # Handle for output PNG file # Handle empty source files if (!@source) { @source = ( "" ); } $height = scalar(@source); # Create image $overview = new GD::Image($overview_width, $height) or die("ERROR: cannot allocate overview image!\n"); # Define colors $col_plain_back = $overview->colorAllocate(0xff, 0xff, 0xff); $col_plain_text = $overview->colorAllocate(0xaa, 0xaa, 0xaa); $col_cov_back = $overview->colorAllocate(0xaa, 0xa7, 0xef); $col_cov_text = $overview->colorAllocate(0x5d, 0x5d, 0xea); $col_nocov_back = $overview->colorAllocate(0xff, 0x00, 0x00); $col_nocov_text = $overview->colorAllocate(0xaa, 0x00, 0x00); $col_hi_back = $overview->colorAllocate(0x00, 0xff, 0x00); $col_hi_text = $overview->colorAllocate(0x00, 0xaa, 0x00); # Visualize each line foreach $line (@source) { # Replace tabs with spaces to keep consistent with source # code view while ($line =~ /^([^\t]*)(\t)/) { $replacement = " "x($tab_size - ((length($1) - 1) % $tab_size)); $line =~ s/^([^\t]*)(\t)/$1$replacement/; } # Skip lines which do not follow the : # specification, otherwise $1 = count, $2 = source code if (!($line =~ /(\*?)(\d*):(.*)$/)) { next; } $count = $2; $source = $3; # Decide which color pair to use # If this line was not instrumented but the one before was, # take the color of that line to widen color areas in # resulting image if (($count eq "") && defined($last_count) && ($last_count ne "")) { $count = $last_count; } if ($count eq "") { # Line was not instrumented $color_text = $col_plain_text; $color_back = $col_plain_back; } elsif ($count == 0) { # Line was instrumented but not executed $color_text = $col_nocov_text; $color_back = $col_nocov_back; } elsif ($1 eq "*") { # Line was highlighted $color_text = $col_hi_text; $color_back = $col_hi_back; } else { # Line was instrumented and executed $color_text = $col_cov_text; $color_back = $col_cov_back; } # Write one pixel for each source character $column = 0; foreach (split("", $source)) { # Check for width if ($column >= $overview_width) { last; } if ($_ eq " ") { # Space $overview->setPixel($column++, $row, $color_back); } else { # Text $overview->setPixel($column++, $row, $color_text); } } # Fill rest of line while ($column < $overview_width) { $overview->setPixel($column++, $row, $color_back); } $last_count = $2; $row++; } # Write PNG file open (PNG_HANDLE, ">", $filename) or die("ERROR: cannot write png file $filename!\n"); binmode(*PNG_HANDLE); print(PNG_HANDLE $overview->png()); close(PNG_HANDLE); } sub genpng_warn_handler($) { my ($msg) = @_; warn("$tool_name: $msg"); } sub genpng_die_handler($) { my ($msg) = @_; die("$tool_name: $msg"); } lcov-1.14/bin/geninfo0000755000000000000000000026463313435777305013257 0ustar rootroot#!/usr/bin/env perl # # Copyright (c) International Business Machines Corp., 2002,2012 # # This program is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 2 of the License, or (at # your option) any later version. # # This program is distributed in the hope that it will be useful, but # WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU # General Public License for more details. # # You should have received a copy of the GNU General Public License # along with this program; if not, write to the Free Software # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA # # # geninfo # # This script generates .info files from data files as created by code # instrumented with gcc's built-in profiling mechanism. Call it with # --help and refer to the geninfo man page to get information on usage # and available options. # # # Authors: # 2002-08-23 created by Peter Oberparleiter # IBM Lab Boeblingen # based on code by Manoj Iyer and # Megan Bock # IBM Austin # 2002-09-05 / Peter Oberparleiter: implemented option that allows file list # 2003-04-16 / Peter Oberparleiter: modified read_gcov so that it can also # parse the new gcov format which is to be introduced in gcc 3.3 # 2003-04-30 / Peter Oberparleiter: made info write to STDERR, not STDOUT # 2003-07-03 / Peter Oberparleiter: added line checksum support, added # --no-checksum # 2003-09-18 / Nigel Hinds: capture branch coverage data from GCOV # 2003-12-11 / Laurent Deniel: added --follow option # workaround gcov (<= 3.2.x) bug with empty .da files # 2004-01-03 / Laurent Deniel: Ignore empty .bb files # 2004-02-16 / Andreas Krebbel: Added support for .gcno/.gcda files and # gcov versioning # 2004-08-09 / Peter Oberparleiter: added configuration file support # 2008-07-14 / Tom Zoerner: added --function-coverage command line option # 2008-08-13 / Peter Oberparleiter: modified function coverage # implementation (now enabled per default) # use strict; use warnings; use File::Basename; use File::Spec::Functions qw /abs2rel catdir file_name_is_absolute splitdir splitpath catpath/; use Getopt::Long; use Digest::MD5 qw(md5_base64); use Cwd qw/abs_path/; if( $^O eq "msys" ) { require File::Spec::Win32; } # Constants our $tool_dir = abs_path(dirname($0)); our $lcov_version = "LCOV version 1.14"; our $lcov_url = "http://ltp.sourceforge.net/coverage/lcov.php"; our $gcov_tool = "gcov"; our $tool_name = basename($0); our $GCOV_VERSION_8_0_0 = 0x80000; our $GCOV_VERSION_4_7_0 = 0x40700; our $GCOV_VERSION_3_4_0 = 0x30400; our $GCOV_VERSION_3_3_0 = 0x30300; our $GCNO_FUNCTION_TAG = 0x01000000; our $GCNO_LINES_TAG = 0x01450000; our $GCNO_FILE_MAGIC = 0x67636e6f; our $BBG_FILE_MAGIC = 0x67626267; # Error classes which users may specify to ignore during processing our $ERROR_GCOV = 0; our $ERROR_SOURCE = 1; our $ERROR_GRAPH = 2; our %ERROR_ID = ( "gcov" => $ERROR_GCOV, "source" => $ERROR_SOURCE, "graph" => $ERROR_GRAPH, ); our $EXCL_START = "LCOV_EXCL_START"; our $EXCL_STOP = "LCOV_EXCL_STOP"; # Marker to exclude branch coverage but keep function and line coveage our $EXCL_BR_START = "LCOV_EXCL_BR_START"; our $EXCL_BR_STOP = "LCOV_EXCL_BR_STOP"; # Compatibility mode values our $COMPAT_VALUE_OFF = 0; our $COMPAT_VALUE_ON = 1; our $COMPAT_VALUE_AUTO = 2; # Compatibility mode value names our %COMPAT_NAME_TO_VALUE = ( "off" => $COMPAT_VALUE_OFF, "on" => $COMPAT_VALUE_ON, "auto" => $COMPAT_VALUE_AUTO, ); # Compatiblity modes our $COMPAT_MODE_LIBTOOL = 1 << 0; our $COMPAT_MODE_HAMMER = 1 << 1; our $COMPAT_MODE_SPLIT_CRC = 1 << 2; # Compatibility mode names our %COMPAT_NAME_TO_MODE = ( "libtool" => $COMPAT_MODE_LIBTOOL, "hammer" => $COMPAT_MODE_HAMMER, "split_crc" => $COMPAT_MODE_SPLIT_CRC, "android_4_4_0" => $COMPAT_MODE_SPLIT_CRC, ); # Map modes to names our %COMPAT_MODE_TO_NAME = ( $COMPAT_MODE_LIBTOOL => "libtool", $COMPAT_MODE_HAMMER => "hammer", $COMPAT_MODE_SPLIT_CRC => "split_crc", ); # Compatibility mode default values our %COMPAT_MODE_DEFAULTS = ( $COMPAT_MODE_LIBTOOL => $COMPAT_VALUE_ON, $COMPAT_MODE_HAMMER => $COMPAT_VALUE_AUTO, $COMPAT_MODE_SPLIT_CRC => $COMPAT_VALUE_AUTO, ); # Compatibility mode auto-detection routines sub compat_hammer_autodetect(); our %COMPAT_MODE_AUTO = ( $COMPAT_MODE_HAMMER => \&compat_hammer_autodetect, $COMPAT_MODE_SPLIT_CRC => 1, # will be done later ); our $BR_LINE = 0; our $BR_BLOCK = 1; our $BR_BRANCH = 2; our $BR_TAKEN = 3; our $BR_VEC_ENTRIES = 4; our $BR_VEC_WIDTH = 32; our $BR_VEC_MAX = vec(pack('b*', 1 x $BR_VEC_WIDTH), 0, $BR_VEC_WIDTH); our $UNNAMED_BLOCK = -1; # Prototypes sub print_usage(*); sub transform_pattern($); sub gen_info($); sub process_dafile($$); sub match_filename($@); sub solve_ambiguous_match($$$); sub split_filename($); sub solve_relative_path($$); sub read_gcov_header($); sub read_gcov_file($); sub info(@); sub map_llvm_version($); sub version_to_str($); sub get_gcov_version(); sub system_no_output($@); sub read_config($); sub apply_config($); sub get_exclusion_data($); sub apply_exclusion_data($$); sub process_graphfile($$); sub filter_fn_name($); sub warn_handler($); sub die_handler($); sub graph_error($$); sub graph_expect($); sub graph_read(*$;$$); sub graph_skip(*$;$); sub uniq(@); sub sort_uniq(@); sub sort_uniq_lex(@); sub graph_cleanup($); sub graph_find_base($); sub graph_from_bb($$$$); sub graph_add_order($$$); sub read_bb_word(*;$); sub read_bb_value(*;$); sub read_bb_string(*$); sub read_bb($); sub read_bbg_word(*;$); sub read_bbg_value(*;$); sub read_bbg_string(*); sub read_bbg_lines_record(*$$$$$); sub read_bbg($); sub read_gcno_word(*;$$); sub read_gcno_value(*$;$$); sub read_gcno_string(*$); sub read_gcno_lines_record(*$$$$$$); sub determine_gcno_split_crc($$$$); sub read_gcno_function_record(*$$$$$); sub read_gcno($); sub get_gcov_capabilities(); sub get_overall_line($$$$); sub print_overall_rate($$$$$$$$$); sub br_gvec_len($); sub br_gvec_get($$); sub debug($); sub int_handler(); sub parse_ignore_errors(@); sub is_external($); sub compat_name($); sub parse_compat_modes($); sub is_compat($); sub is_compat_auto($); # Global variables our $gcov_version; our $gcov_version_string; our $graph_file_extension; our $data_file_extension; our @data_directory; our $test_name = ""; our $quiet; our $help; our $output_filename; our $base_directory; our $version; our $follow; our $checksum; our $no_checksum; our $opt_compat_libtool; our $opt_no_compat_libtool; our $rc_adjust_src_path;# Regexp specifying parts to remove from source path our $adjust_src_pattern; our $adjust_src_replace; our $adjust_testname; our $config; # Configuration file contents our @ignore_errors; # List of errors to ignore (parameter) our @ignore; # List of errors to ignore (array) our $initial; our @include_patterns; # List of source file patterns to include our @exclude_patterns; # List of source file patterns to exclude our %excluded_files; # Files excluded due to include/exclude options our $no_recursion = 0; our $maxdepth; our $no_markers = 0; our $opt_derive_func_data = 0; our $opt_external = 1; our $opt_no_external; our $debug = 0; our $gcov_caps; our @gcov_options; our @internal_dirs; our $opt_config_file; our $opt_gcov_all_blocks = 1; our $opt_compat; our %opt_rc; our %compat_value; our $gcno_split_crc; our $func_coverage = 1; our $br_coverage = 0; our $rc_auto_base = 1; our $excl_line = "LCOV_EXCL_LINE"; our $excl_br_line = "LCOV_EXCL_BR_LINE"; our $cwd = `pwd`; chomp($cwd); # # Code entry point # # Register handler routine to be called when interrupted $SIG{"INT"} = \&int_handler; $SIG{__WARN__} = \&warn_handler; $SIG{__DIE__} = \&die_handler; # Set LC_ALL so that gcov output will be in a unified format $ENV{"LC_ALL"} = "C"; # Check command line for a configuration file name Getopt::Long::Configure("pass_through", "no_auto_abbrev"); GetOptions("config-file=s" => \$opt_config_file, "rc=s%" => \%opt_rc); Getopt::Long::Configure("default"); { # Remove spaces around rc options my %new_opt_rc; while (my ($key, $value) = each(%opt_rc)) { $key =~ s/^\s+|\s+$//g; $value =~ s/^\s+|\s+$//g; $new_opt_rc{$key} = $value; } %opt_rc = %new_opt_rc; } # Read configuration file if available if (defined($opt_config_file)) { $config = read_config($opt_config_file); } elsif (defined($ENV{"HOME"}) && (-r $ENV{"HOME"}."/.lcovrc")) { $config = read_config($ENV{"HOME"}."/.lcovrc"); } elsif (-r "/etc/lcovrc") { $config = read_config("/etc/lcovrc"); } elsif (-r "/usr/local/etc/lcovrc") { $config = read_config("/usr/local/etc/lcovrc"); } if ($config || %opt_rc) { # Copy configuration file and --rc values to variables apply_config({ "geninfo_gcov_tool" => \$gcov_tool, "geninfo_adjust_testname" => \$adjust_testname, "geninfo_checksum" => \$checksum, "geninfo_no_checksum" => \$no_checksum, # deprecated "geninfo_compat_libtool" => \$opt_compat_libtool, "geninfo_external" => \$opt_external, "geninfo_gcov_all_blocks" => \$opt_gcov_all_blocks, "geninfo_compat" => \$opt_compat, "geninfo_adjust_src_path" => \$rc_adjust_src_path, "geninfo_auto_base" => \$rc_auto_base, "lcov_function_coverage" => \$func_coverage, "lcov_branch_coverage" => \$br_coverage, "lcov_excl_line" => \$excl_line, "lcov_excl_br_line" => \$excl_br_line, }); # Merge options if (defined($no_checksum)) { $checksum = ($no_checksum ? 0 : 1); $no_checksum = undef; } # Check regexp if (defined($rc_adjust_src_path)) { my ($pattern, $replace) = split(/\s*=>\s*/, $rc_adjust_src_path); local $SIG{__DIE__}; eval '$adjust_src_pattern = qr>'.$pattern.'>;'; if (!defined($adjust_src_pattern)) { my $msg = $@; chomp($msg); $msg =~ s/at \(eval.*$//; warn("WARNING: invalid pattern in ". "geninfo_adjust_src_path: $msg\n"); } elsif (!defined($replace)) { # If no replacement is specified, simply remove pattern $adjust_src_replace = ""; } else { $adjust_src_replace = $replace; } } for my $regexp (($excl_line, $excl_br_line)) { eval 'qr/'.$regexp.'/'; my $error = $@; chomp($error); $error =~ s/at \(eval.*$//; die("ERROR: invalid exclude pattern: $error") if $error; } } # Parse command line options if (!GetOptions("test-name|t=s" => \$test_name, "output-filename|o=s" => \$output_filename, "checksum" => \$checksum, "no-checksum" => \$no_checksum, "base-directory|b=s" => \$base_directory, "version|v" =>\$version, "quiet|q" => \$quiet, "help|h|?" => \$help, "follow|f" => \$follow, "compat-libtool" => \$opt_compat_libtool, "no-compat-libtool" => \$opt_no_compat_libtool, "gcov-tool=s" => \$gcov_tool, "ignore-errors=s" => \@ignore_errors, "initial|i" => \$initial, "include=s" => \@include_patterns, "exclude=s" => \@exclude_patterns, "no-recursion" => \$no_recursion, "no-markers" => \$no_markers, "derive-func-data" => \$opt_derive_func_data, "debug" => \$debug, "external|e" => \$opt_external, "no-external" => \$opt_no_external, "compat=s" => \$opt_compat, "config-file=s" => \$opt_config_file, "rc=s%" => \%opt_rc, )) { print(STDERR "Use $tool_name --help to get usage information\n"); exit(1); } else { # Merge options if (defined($no_checksum)) { $checksum = ($no_checksum ? 0 : 1); $no_checksum = undef; } if (defined($opt_no_compat_libtool)) { $opt_compat_libtool = ($opt_no_compat_libtool ? 0 : 1); $opt_no_compat_libtool = undef; } if (defined($opt_no_external)) { $opt_external = 0; $opt_no_external = undef; } if(@include_patterns) { # Need perlreg expressions instead of shell pattern @include_patterns = map({ transform_pattern($_); } @include_patterns); } if(@exclude_patterns) { # Need perlreg expressions instead of shell pattern @exclude_patterns = map({ transform_pattern($_); } @exclude_patterns); } } @data_directory = @ARGV; debug("$lcov_version\n"); # Check for help option if ($help) { print_usage(*STDOUT); exit(0); } # Check for version option if ($version) { print("$tool_name: $lcov_version\n"); exit(0); } # Check gcov tool if (system_no_output(3, $gcov_tool, "--help") == -1) { die("ERROR: need tool $gcov_tool!\n"); } ($gcov_version, $gcov_version_string) = get_gcov_version(); # Determine gcov options $gcov_caps = get_gcov_capabilities(); push(@gcov_options, "-b") if ($gcov_caps->{'branch-probabilities'} && ($br_coverage || $func_coverage)); push(@gcov_options, "-c") if ($gcov_caps->{'branch-counts'} && $br_coverage); push(@gcov_options, "-a") if ($gcov_caps->{'all-blocks'} && $opt_gcov_all_blocks && $br_coverage); if ($gcov_caps->{'hash-filenames'}) { push(@gcov_options, "-x"); } else { push(@gcov_options, "-p") if ($gcov_caps->{'preserve-paths'}); } # Determine compatibility modes parse_compat_modes($opt_compat); # Determine which errors the user wants us to ignore parse_ignore_errors(@ignore_errors); # Make sure test names only contain valid characters if ($test_name =~ s/\W/_/g) { warn("WARNING: invalid characters removed from testname!\n"); } # Adjust test name to include uname output if requested if ($adjust_testname) { $test_name .= "__".`uname -a`; $test_name =~ s/\W/_/g; } # Make sure base_directory contains an absolute path specification if ($base_directory) { $base_directory = solve_relative_path($cwd, $base_directory); } # Check for follow option if ($follow) { $follow = "-follow" } else { $follow = ""; } # Determine checksum mode if (defined($checksum)) { # Normalize to boolean $checksum = ($checksum ? 1 : 0); } else { # Default is off $checksum = 0; } # Determine max depth for recursion if ($no_recursion) { $maxdepth = "-maxdepth 1"; } else { $maxdepth = ""; } # Check for directory name if (!@data_directory) { die("No directory specified\n". "Use $tool_name --help to get usage information\n"); } else { foreach (@data_directory) { stat($_); if (!-r _) { die("ERROR: cannot read $_!\n"); } } } if ($gcov_version < $GCOV_VERSION_3_4_0) { if (is_compat($COMPAT_MODE_HAMMER)) { $data_file_extension = ".da"; $graph_file_extension = ".bbg"; } else { $data_file_extension = ".da"; $graph_file_extension = ".bb"; } } else { $data_file_extension = ".gcda"; $graph_file_extension = ".gcno"; } # Check output filename if (defined($output_filename) && ($output_filename ne "-")) { # Initially create output filename, data is appended # for each data file processed local *DUMMY_HANDLE; open(DUMMY_HANDLE, ">", $output_filename) or die("ERROR: cannot create $output_filename!\n"); close(DUMMY_HANDLE); # Make $output_filename an absolute path because we're going # to change directories while processing files if (!($output_filename =~ /^\/(.*)$/)) { $output_filename = $cwd."/".$output_filename; } } # Build list of directories to identify external files foreach my $entry(@data_directory, $base_directory) { next if (!defined($entry)); push(@internal_dirs, solve_relative_path($cwd, $entry)); } # Do something foreach my $entry (@data_directory) { gen_info($entry); } if ($initial && $br_coverage) { warn("Note: --initial does not generate branch coverage ". "data\n"); } info("Finished .info-file creation\n"); exit(0); # # print_usage(handle) # # Print usage information. # sub print_usage(*) { local *HANDLE = $_[0]; print(HANDLE < (.) and * => (.*) $pattern =~ s/\*/\(\.\*\)/g; $pattern =~ s/\?/\(\.\)/g; return $pattern; } # # get_common_prefix(min_dir, filenames) # # Return the longest path prefix shared by all filenames. MIN_DIR specifies # the minimum number of directories that a filename may have after removing # the prefix. # sub get_common_prefix($@) { my ($min_dir, @files) = @_; my $file; my @prefix; my $i; foreach $file (@files) { my ($v, $d, $f) = splitpath($file); my @comp = splitdir($d); if (!@prefix) { @prefix = @comp; next; } for ($i = 0; $i < scalar(@comp) && $i < scalar(@prefix); $i++) { if ($comp[$i] ne $prefix[$i] || ((scalar(@comp) - ($i + 1)) <= $min_dir)) { delete(@prefix[$i..scalar(@prefix)]); last; } } } return catdir(@prefix); } # # gen_info(directory) # # Traverse DIRECTORY and create a .info file for each data file found. # The .info file contains TEST_NAME in the following format: # # TN: # # For each source file name referenced in the data file, there is a section # containing source code and coverage data: # # SF: # FN:, for each function # DA:, for each instrumented line # LH: greater than 0 # LF: # # Sections are separated by: # # end_of_record # # In addition to the main source code file there are sections for each # #included file containing executable code. Note that the absolute path # of a source file is generated by interpreting the contents of the respective # graph file. Relative filenames are prefixed with the directory in which the # graph file is found. Note also that symbolic links to the graph file will be # resolved so that the actual file path is used instead of the path to a link. # This approach is necessary for the mechanism to work with the /proc/gcov # files. # # Die on error. # sub gen_info($) { my $directory = $_[0]; my @file_list; my $file; my $prefix; my $type; my $ext; if ($initial) { $type = "graph"; $ext = $graph_file_extension; } else { $type = "data"; $ext = $data_file_extension; } if (-d $directory) { info("Scanning $directory for $ext files ...\n"); @file_list = `find "$directory" $maxdepth $follow -name \\*$ext -type f -o -name \\*$ext -type l 2>/dev/null`; chomp(@file_list); if (!@file_list) { warn("WARNING: no $ext files found in $directory - ". "skipping!\n"); return; } $prefix = get_common_prefix(1, @file_list); info("Found %d %s files in %s\n", $#file_list+1, $type, $directory); } else { @file_list = ($directory); $prefix = ""; } # Process all files in list foreach $file (@file_list) { # Process file if ($initial) { process_graphfile($file, $prefix); } else { process_dafile($file, $prefix); } } # Report whether files were excluded. if (%excluded_files) { info("Excluded data for %d files due to include/exclude options\n", scalar keys %excluded_files); } } # # derive_data(contentdata, funcdata, bbdata) # # Calculate function coverage data by combining line coverage data and the # list of lines belonging to a function. # # contentdata: [ instr1, count1, source1, instr2, count2, source2, ... ] # instr: Instrumentation flag for line n # count: Execution count for line n # source: Source code for line n # # funcdata: [ count1, func1, count2, func2, ... ] # count: Execution count for function number n # func: Function name for function number n # # bbdata: function_name -> [ line1, line2, ... ] # line: Line number belonging to the corresponding function # sub derive_data($$$) { my ($contentdata, $funcdata, $bbdata) = @_; my @gcov_content = @{$contentdata}; my @gcov_functions = @{$funcdata}; my %fn_count; my %ln_fn; my $line; my $maxline; my %fn_name; my $fn; my $count; if (!defined($bbdata)) { return @gcov_functions; } # First add existing function data while (@gcov_functions) { $count = shift(@gcov_functions); $fn = shift(@gcov_functions); $fn_count{$fn} = $count; } # Convert line coverage data to function data foreach $fn (keys(%{$bbdata})) { my $line_data = $bbdata->{$fn}; my $line; my $fninstr = 0; if ($fn eq "") { next; } # Find the lowest line count for this function $count = 0; foreach $line (@$line_data) { my $linstr = $gcov_content[ ( $line - 1 ) * 3 + 0 ]; my $lcount = $gcov_content[ ( $line - 1 ) * 3 + 1 ]; next if (!$linstr); $fninstr = 1; if (($lcount > 0) && (($count == 0) || ($lcount < $count))) { $count = $lcount; } } next if (!$fninstr); $fn_count{$fn} = $count; } # Check if we got data for all functions foreach $fn (keys(%fn_name)) { if ($fn eq "") { next; } if (defined($fn_count{$fn})) { next; } warn("WARNING: no derived data found for function $fn\n"); } # Convert hash to list in @gcov_functions format foreach $fn (sort(keys(%fn_count))) { push(@gcov_functions, $fn_count{$fn}, $fn); } return @gcov_functions; } # # get_filenames(directory, pattern) # # Return a list of filenames found in directory which match the specified # pattern. # # Die on error. # sub get_filenames($$) { my ($dirname, $pattern) = @_; my @result; my $directory; local *DIR; opendir(DIR, $dirname) or die("ERROR: cannot read directory $dirname\n"); while ($directory = readdir(DIR)) { push(@result, $directory) if ($directory =~ /$pattern/); } closedir(DIR); return @result; } # # process_dafile(da_filename, dir) # # Create a .info file for a single data file. # # Die on error. # sub process_dafile($$) { my ($file, $dir) = @_; my $da_filename; # Name of data file to process my $da_dir; # Directory of data file my $source_dir; # Directory of source file my $da_basename; # data filename without ".da/.gcda" extension my $bb_filename; # Name of respective graph file my $bb_basename; # Basename of the original graph file my $graph; # Contents of graph file my $instr; # Contents of graph file part 2 my $gcov_error; # Error code of gcov tool my $object_dir; # Directory containing all object files my $source_filename; # Name of a source code file my $gcov_file; # Name of a .gcov file my @gcov_content; # Content of a .gcov file my $gcov_branches; # Branch content of a .gcov file my @gcov_functions; # Function calls of a .gcov file my @gcov_list; # List of generated .gcov files my $line_number; # Line number count my $lines_hit; # Number of instrumented lines hit my $lines_found; # Number of instrumented lines found my $funcs_hit; # Number of instrumented functions hit my $funcs_found; # Number of instrumented functions found my $br_hit; my $br_found; my $source; # gcov source header information my $object; # gcov object header information my @matches; # List of absolute paths matching filename my $base_dir; # Base directory for current file my @tmp_links; # Temporary links to be cleaned up my @result; my $index; my $da_renamed; # If data file is to be renamed local *INFO_HANDLE; info("Processing %s\n", abs2rel($file, $dir)); # Get path to data file in absolute and normalized form (begins with /, # contains no more ../ or ./) $da_filename = solve_relative_path($cwd, $file); # Get directory and basename of data file ($da_dir, $da_basename) = split_filename($da_filename); $source_dir = $da_dir; if (is_compat($COMPAT_MODE_LIBTOOL)) { # Avoid files from .libs dirs $source_dir =~ s/\.libs$//; } if (-z $da_filename) { $da_renamed = 1; } else { $da_renamed = 0; } # Construct base_dir for current file if ($base_directory) { $base_dir = $base_directory; } else { $base_dir = $source_dir; } # Check for writable $base_dir (gcov will try to write files there) stat($base_dir); if (!-w _) { die("ERROR: cannot write to directory $base_dir!\n"); } # Construct name of graph file $bb_basename = $da_basename.$graph_file_extension; $bb_filename = "$da_dir/$bb_basename"; # Find out the real location of graph file in case we're just looking at # a link while (readlink($bb_filename)) { my $last_dir = dirname($bb_filename); $bb_filename = readlink($bb_filename); $bb_filename = solve_relative_path($last_dir, $bb_filename); } # Ignore empty graph file (e.g. source file with no statement) if (-z $bb_filename) { warn("WARNING: empty $bb_filename (skipped)\n"); return; } # Read contents of graph file into hash. We need it later to find out # the absolute path to each .gcov file created as well as for # information about functions and their source code positions. if ($gcov_version < $GCOV_VERSION_3_4_0) { if (is_compat($COMPAT_MODE_HAMMER)) { ($instr, $graph) = read_bbg($bb_filename); } else { ($instr, $graph) = read_bb($bb_filename); } } else { ($instr, $graph) = read_gcno($bb_filename); } # Try to find base directory automatically if requested by user if ($rc_auto_base) { $base_dir = find_base_from_graph($base_dir, $instr, $graph); } ($instr, $graph) = adjust_graph_filenames($base_dir, $instr, $graph); # Set $object_dir to real location of object files. This may differ # from $da_dir if the graph file is just a link to the "real" object # file location. $object_dir = dirname($bb_filename); # Is the data file in a different directory? (this happens e.g. with # the gcov-kernel patch) if ($object_dir ne $da_dir) { # Need to create link to data file in $object_dir system("ln", "-s", $da_filename, "$object_dir/$da_basename$data_file_extension") and die ("ERROR: cannot create link $object_dir/". "$da_basename$data_file_extension!\n"); push(@tmp_links, "$object_dir/$da_basename$data_file_extension"); # Need to create link to graph file if basename of link # and file are different (CONFIG_MODVERSION compat) if ((basename($bb_filename) ne $bb_basename) && (! -e "$object_dir/$bb_basename")) { symlink($bb_filename, "$object_dir/$bb_basename") or warn("WARNING: cannot create link ". "$object_dir/$bb_basename\n"); push(@tmp_links, "$object_dir/$bb_basename"); } } # Change to directory containing data files and apply GCOV debug("chdir($base_dir)\n"); chdir($base_dir); if ($da_renamed) { # Need to rename empty data file to workaround # gcov <= 3.2.x bug (Abort) system_no_output(3, "mv", "$da_filename", "$da_filename.ori") and die ("ERROR: cannot rename $da_filename\n"); } # Execute gcov command and suppress standard output $gcov_error = system_no_output(1, $gcov_tool, $da_filename, "-o", $object_dir, @gcov_options); if ($da_renamed) { system_no_output(3, "mv", "$da_filename.ori", "$da_filename") and die ("ERROR: cannot rename $da_filename.ori"); } # Clean up temporary links foreach (@tmp_links) { unlink($_); } if ($gcov_error) { if ($ignore[$ERROR_GCOV]) { warn("WARNING: GCOV failed for $da_filename!\n"); return; } die("ERROR: GCOV failed for $da_filename!\n"); } # Collect data from resulting .gcov files and create .info file @gcov_list = get_filenames('.', '\.gcov$'); # Check for files if (!@gcov_list) { warn("WARNING: gcov did not create any files for ". "$da_filename!\n"); } # Check whether we're writing to a single file if ($output_filename) { if ($output_filename eq "-") { *INFO_HANDLE = *STDOUT; } else { # Append to output file open(INFO_HANDLE, ">>", $output_filename) or die("ERROR: cannot write to ". "$output_filename!\n"); } } else { # Open .info file for output open(INFO_HANDLE, ">", "$da_filename.info") or die("ERROR: cannot create $da_filename.info!\n"); } # Write test name printf(INFO_HANDLE "TN:%s\n", $test_name); # Traverse the list of generated .gcov files and combine them into a # single .info file foreach $gcov_file (sort(@gcov_list)) { my $i; my $num; # Skip gcov file for gcc built-in code next if ($gcov_file eq ".gcov"); ($source, $object) = read_gcov_header($gcov_file); if (!defined($source)) { # Derive source file name from gcov file name if # header format could not be parsed $source = $gcov_file; $source =~ s/\.gcov$//; } $source = solve_relative_path($base_dir, $source); if (defined($adjust_src_pattern)) { # Apply transformation as specified by user $source =~ s/$adjust_src_pattern/$adjust_src_replace/g; } # gcov will happily create output even if there's no source code # available - this interferes with checksum creation so we need # to pull the emergency brake here. if (! -r $source && $checksum) { if ($ignore[$ERROR_SOURCE]) { warn("WARNING: could not read source file ". "$source\n"); next; } die("ERROR: could not read source file $source\n"); } @matches = match_filename($source, keys(%{$instr})); # Skip files that are not mentioned in the graph file if (!@matches) { warn("WARNING: cannot find an entry for ".$gcov_file. " in $graph_file_extension file, skipping ". "file!\n"); unlink($gcov_file); next; } # Read in contents of gcov file @result = read_gcov_file($gcov_file); if (!defined($result[0])) { warn("WARNING: skipping unreadable file ". $gcov_file."\n"); unlink($gcov_file); next; } @gcov_content = @{$result[0]}; $gcov_branches = $result[1]; @gcov_functions = @{$result[2]}; # Skip empty files if (!@gcov_content) { warn("WARNING: skipping empty file ".$gcov_file."\n"); unlink($gcov_file); next; } if (scalar(@matches) == 1) { # Just one match $source_filename = $matches[0]; } else { # Try to solve the ambiguity $source_filename = solve_ambiguous_match($gcov_file, \@matches, \@gcov_content); } if (@include_patterns) { my $keep = 0; foreach my $pattern (@include_patterns) { $keep ||= ($source_filename =~ (/^$pattern$/)); } if (!$keep) { $excluded_files{$source_filename} = (); unlink($gcov_file); next; } } if (@exclude_patterns) { my $exclude = 0; foreach my $pattern (@exclude_patterns) { $exclude ||= ($source_filename =~ (/^$pattern$/)); } if ($exclude) { $excluded_files{$source_filename} = (); unlink($gcov_file); next; } } # Skip external files if requested if (!$opt_external) { if (is_external($source_filename)) { info(" ignoring data for external file ". "$source_filename\n"); unlink($gcov_file); next; } } # Write absolute path of source file printf(INFO_HANDLE "SF:%s\n", $source_filename); # If requested, derive function coverage data from # line coverage data of the first line of a function if ($opt_derive_func_data) { @gcov_functions = derive_data(\@gcov_content, \@gcov_functions, $graph->{$source_filename}); } # Write function-related information if (defined($graph->{$source_filename})) { my $fn_data = $graph->{$source_filename}; my $fn; foreach $fn (sort {$fn_data->{$a}->[0] <=> $fn_data->{$b}->[0]} keys(%{$fn_data})) { my $ln_data = $fn_data->{$fn}; my $line = $ln_data->[0]; # Skip empty function if ($fn eq "") { next; } # Remove excluded functions if (!$no_markers) { my $gfn; my $found = 0; foreach $gfn (@gcov_functions) { if ($gfn eq $fn) { $found = 1; last; } } if (!$found) { next; } } # Normalize function name $fn = filter_fn_name($fn); print(INFO_HANDLE "FN:$line,$fn\n"); } } #-- #-- FNDA: , #-- FNF: overall count of functions #-- FNH: overall count of functions with non-zero call count #-- $funcs_found = 0; $funcs_hit = 0; while (@gcov_functions) { my $count = shift(@gcov_functions); my $fn = shift(@gcov_functions); $fn = filter_fn_name($fn); printf(INFO_HANDLE "FNDA:$count,$fn\n"); $funcs_found++; $funcs_hit++ if ($count > 0); } if ($funcs_found > 0) { printf(INFO_HANDLE "FNF:%s\n", $funcs_found); printf(INFO_HANDLE "FNH:%s\n", $funcs_hit); } # Write coverage information for each instrumented branch: # # BRDA:,,, # # where 'taken' is the number of times the branch was taken # or '-' if the block to which the branch belongs was never # executed $br_found = 0; $br_hit = 0; $num = br_gvec_len($gcov_branches); for ($i = 0; $i < $num; $i++) { my ($line, $block, $branch, $taken) = br_gvec_get($gcov_branches, $i); $block = $BR_VEC_MAX if ($block < 0); print(INFO_HANDLE "BRDA:$line,$block,$branch,$taken\n"); $br_found++; $br_hit++ if ($taken ne '-' && $taken > 0); } if ($br_found > 0) { printf(INFO_HANDLE "BRF:%s\n", $br_found); printf(INFO_HANDLE "BRH:%s\n", $br_hit); } # Reset line counters $line_number = 0; $lines_found = 0; $lines_hit = 0; # Write coverage information for each instrumented line # Note: @gcov_content contains a list of (flag, count, source) # tuple for each source code line while (@gcov_content) { $line_number++; # Check for instrumented line if ($gcov_content[0]) { $lines_found++; printf(INFO_HANDLE "DA:".$line_number.",". $gcov_content[1].($checksum ? ",". md5_base64($gcov_content[2]) : ""). "\n"); # Increase $lines_hit in case of an execution # count>0 if ($gcov_content[1] > 0) { $lines_hit++; } } # Remove already processed data from array splice(@gcov_content,0,3); } # Write line statistics and section separator printf(INFO_HANDLE "LF:%s\n", $lines_found); printf(INFO_HANDLE "LH:%s\n", $lines_hit); print(INFO_HANDLE "end_of_record\n"); # Remove .gcov file after processing unlink($gcov_file); } if (!($output_filename && ($output_filename eq "-"))) { close(INFO_HANDLE); } # Change back to initial directory chdir($cwd); } # # solve_relative_path(path, dir) # # Solve relative path components of DIR which, if not absolute, resides in PATH. # sub solve_relative_path($$) { my $path = $_[0]; my $dir = $_[1]; my $volume; my $directories; my $filename; my @dirs; # holds path elements my $result; # Convert from Windows path to msys path if( $^O eq "msys" ) { # search for a windows drive letter at the beginning ($volume, $directories, $filename) = File::Spec::Win32->splitpath( $dir ); if( $volume ne '' ) { my $uppercase_volume; # transform c/d\../e/f\g to Windows style c\d\..\e\f\g $dir = File::Spec::Win32->canonpath( $dir ); # use Win32 module to retrieve path components # $uppercase_volume is not used any further ( $uppercase_volume, $directories, $filename ) = File::Spec::Win32->splitpath( $dir ); @dirs = File::Spec::Win32->splitdir( $directories ); # prepend volume, since in msys C: is always mounted to /c $volume =~ s|^([a-zA-Z]+):|/\L$1\E|; unshift( @dirs, $volume ); # transform to Unix style '/' path $directories = File::Spec->catdir( @dirs ); $dir = File::Spec->catpath( '', $directories, $filename ); } else { # eliminate '\' path separators $dir = File::Spec->canonpath( $dir ); } } $result = $dir; # Prepend path if not absolute if ($dir =~ /^[^\/]/) { $result = "$path/$result"; } # Remove // $result =~ s/\/\//\//g; # Remove . while ($result =~ s/\/\.\//\//g) { } $result =~ s/\/\.$/\//g; # Remove trailing / $result =~ s/\/$//g; # Solve .. while ($result =~ s/\/[^\/]+\/\.\.\//\//) { } # Remove preceding .. $result =~ s/^\/\.\.\//\//g; return $result; } # # match_filename(gcov_filename, list) # # Return a list of those entries of LIST which match the relative filename # GCOV_FILENAME. # sub match_filename($@) { my ($filename, @list) = @_; my ($vol, $dir, $file) = splitpath($filename); my @comp = splitdir($dir); my $comps = scalar(@comp); my $entry; my @result; entry: foreach $entry (@list) { my ($evol, $edir, $efile) = splitpath($entry); my @ecomp; my $ecomps; my $i; # Filename component must match if ($efile ne $file) { next; } # Check directory components last to first for match @ecomp = splitdir($edir); $ecomps = scalar(@ecomp); if ($ecomps < $comps) { next; } for ($i = 0; $i < $comps; $i++) { if ($comp[$comps - $i - 1] ne $ecomp[$ecomps - $i - 1]) { next entry; } } push(@result, $entry), } return @result; } # # solve_ambiguous_match(rel_filename, matches_ref, gcov_content_ref) # # Try to solve ambiguous matches of mapping (gcov file) -> (source code) file # by comparing source code provided in the GCOV file with that of the files # in MATCHES. REL_FILENAME identifies the relative filename of the gcov # file. # # Return the one real match or die if there is none. # sub solve_ambiguous_match($$$) { my $rel_name = $_[0]; my $matches = $_[1]; my $content = $_[2]; my $filename; my $index; my $no_match; local *SOURCE; # Check the list of matches foreach $filename (@$matches) { # Compare file contents open(SOURCE, "<", $filename) or die("ERROR: cannot read $filename!\n"); $no_match = 0; for ($index = 2; ; $index += 3) { chomp; # Also remove CR from line-end s/\015$//; if ($_ ne @$content[$index]) { $no_match = 1; last; } } close(SOURCE); if (!$no_match) { info("Solved source file ambiguity for $rel_name\n"); return $filename; } } die("ERROR: could not match gcov data for $rel_name!\n"); } # # split_filename(filename) # # Return (path, filename, extension) for a given FILENAME. # sub split_filename($) { my @path_components = split('/', $_[0]); my @file_components = split('\.', pop(@path_components)); my $extension = pop(@file_components); return (join("/",@path_components), join(".",@file_components), $extension); } # # read_gcov_header(gcov_filename) # # Parse file GCOV_FILENAME and return a list containing the following # information: # # (source, object) # # where: # # source: complete relative path of the source code file (gcc >= 3.3 only) # object: name of associated graph file # # Die on error. # sub read_gcov_header($) { my $source; my $object; local *INPUT; if (!open(INPUT, "<", $_[0])) { if ($ignore_errors[$ERROR_GCOV]) { warn("WARNING: cannot read $_[0]!\n"); return (undef,undef); } die("ERROR: cannot read $_[0]!\n"); } while () { chomp($_); # Also remove CR from line-end s/\015$//; if (/^\s+-:\s+0:Source:(.*)$/) { # Source: header entry $source = $1; } elsif (/^\s+-:\s+0:Object:(.*)$/) { # Object: header entry $object = $1; } else { last; } } close(INPUT); return ($source, $object); } # # br_gvec_len(vector) # # Return the number of entries in the branch coverage vector. # sub br_gvec_len($) { my ($vec) = @_; return 0 if (!defined($vec)); return (length($vec) * 8 / $BR_VEC_WIDTH) / $BR_VEC_ENTRIES; } # # br_gvec_get(vector, number) # # Return an entry from the branch coverage vector. # sub br_gvec_get($$) { my ($vec, $num) = @_; my $line; my $block; my $branch; my $taken; my $offset = $num * $BR_VEC_ENTRIES; # Retrieve data from vector $line = vec($vec, $offset + $BR_LINE, $BR_VEC_WIDTH); $block = vec($vec, $offset + $BR_BLOCK, $BR_VEC_WIDTH); $block = -1 if ($block == $BR_VEC_MAX); $branch = vec($vec, $offset + $BR_BRANCH, $BR_VEC_WIDTH); $taken = vec($vec, $offset + $BR_TAKEN, $BR_VEC_WIDTH); # Decode taken value from an integer if ($taken == 0) { $taken = "-"; } else { $taken--; } return ($line, $block, $branch, $taken); } # # br_gvec_push(vector, line, block, branch, taken) # # Add an entry to the branch coverage vector. # sub br_gvec_push($$$$$) { my ($vec, $line, $block, $branch, $taken) = @_; my $offset; $vec = "" if (!defined($vec)); $offset = br_gvec_len($vec) * $BR_VEC_ENTRIES; $block = $BR_VEC_MAX if $block < 0; # Encode taken value into an integer if ($taken eq "-") { $taken = 0; } else { $taken++; } # Add to vector vec($vec, $offset + $BR_LINE, $BR_VEC_WIDTH) = $line; vec($vec, $offset + $BR_BLOCK, $BR_VEC_WIDTH) = $block; vec($vec, $offset + $BR_BRANCH, $BR_VEC_WIDTH) = $branch; vec($vec, $offset + $BR_TAKEN, $BR_VEC_WIDTH) = $taken; return $vec; } # # read_gcov_file(gcov_filename) # # Parse file GCOV_FILENAME (.gcov file format) and return the list: # (reference to gcov_content, reference to gcov_branch, reference to gcov_func) # # gcov_content is a list of 3 elements # (flag, count, source) for each source code line: # # $result[($line_number-1)*3+0] = instrumentation flag for line $line_number # $result[($line_number-1)*3+1] = execution count for line $line_number # $result[($line_number-1)*3+2] = source code text for line $line_number # # gcov_branch is a vector of 4 4-byte long elements for each branch: # line number, block number, branch number, count + 1 or 0 # # gcov_func is a list of 2 elements # (number of calls, function name) for each function # # Die on error. # sub read_gcov_file($) { my $filename = $_[0]; my @result = (); my $branches = ""; my @functions = (); my $number; my $exclude_flag = 0; my $exclude_line = 0; my $exclude_br_flag = 0; my $exclude_branch = 0; my $last_block = $UNNAMED_BLOCK; my $last_line = 0; local *INPUT; if (!open(INPUT, "<", $filename)) { if ($ignore_errors[$ERROR_GCOV]) { warn("WARNING: cannot read $filename!\n"); return (undef, undef, undef); } die("ERROR: cannot read $filename!\n"); } if ($gcov_version < $GCOV_VERSION_3_3_0) { # Expect gcov format as used in gcc < 3.3 while () { chomp($_); # Also remove CR from line-end s/\015$//; if (/^branch\s+(\d+)\s+taken\s+=\s+(\d+)/) { next if (!$br_coverage); next if ($exclude_line); next if ($exclude_branch); $branches = br_gvec_push($branches, $last_line, $last_block, $1, $2); } elsif (/^branch\s+(\d+)\s+never\s+executed/) { next if (!$br_coverage); next if ($exclude_line); next if ($exclude_branch); $branches = br_gvec_push($branches, $last_line, $last_block, $1, '-'); } elsif (/^call/ || /^function/) { # Function call return data } else { $last_line++; # Check for exclusion markers if (!$no_markers) { if (/$EXCL_STOP/) { $exclude_flag = 0; } elsif (/$EXCL_START/) { $exclude_flag = 1; } if (/$excl_line/ || $exclude_flag) { $exclude_line = 1; } else { $exclude_line = 0; } } # Check for exclusion markers (branch exclude) if (!$no_markers) { if (/$EXCL_BR_STOP/) { $exclude_br_flag = 0; } elsif (/$EXCL_BR_START/) { $exclude_br_flag = 1; } if (/$excl_br_line/ || $exclude_br_flag) { $exclude_branch = 1; } else { $exclude_branch = 0; } } # Source code execution data if (/^\t\t(.*)$/) { # Uninstrumented line push(@result, 0); push(@result, 0); push(@result, $1); next; } $number = (split(" ",substr($_, 0, 16)))[0]; # Check for zero count which is indicated # by ###### if ($number eq "######") { $number = 0; } if ($exclude_line) { # Register uninstrumented line instead push(@result, 0); push(@result, 0); } else { push(@result, 1); push(@result, $number); } push(@result, substr($_, 16)); } } } else { # Expect gcov format as used in gcc >= 3.3 while () { chomp($_); # Also remove CR from line-end s/\015$//; if (/^\s*(\d+|\$+|\%+):\s*(\d+)-block\s+(\d+)\s*$/) { # Block information - used to group related # branches $last_line = $2; $last_block = $3; } elsif (/^branch\s+(\d+)\s+taken\s+(\d+)/) { next if (!$br_coverage); next if ($exclude_line); next if ($exclude_branch); $branches = br_gvec_push($branches, $last_line, $last_block, $1, $2); } elsif (/^branch\s+(\d+)\s+never\s+executed/) { next if (!$br_coverage); next if ($exclude_line); next if ($exclude_branch); $branches = br_gvec_push($branches, $last_line, $last_block, $1, '-'); } elsif (/^function\s+(.+)\s+called\s+(\d+)\s+/) { next if (!$func_coverage); if ($exclude_line) { next; } push(@functions, $2, $1); } elsif (/^call/) { # Function call return data } elsif (/^\s*([^:]+):\s*([^:]+):(.*)$/) { my ($count, $line, $code) = ($1, $2, $3); # Skip instance-specific counts next if ($line <= (scalar(@result) / 3)); $last_line = $line; $last_block = $UNNAMED_BLOCK; # Check for exclusion markers if (!$no_markers) { if (/$EXCL_STOP/) { $exclude_flag = 0; } elsif (/$EXCL_START/) { $exclude_flag = 1; } if (/$excl_line/ || $exclude_flag) { $exclude_line = 1; } else { $exclude_line = 0; } } # Check for exclusion markers (branch exclude) if (!$no_markers) { if (/$EXCL_BR_STOP/) { $exclude_br_flag = 0; } elsif (/$EXCL_BR_START/) { $exclude_br_flag = 1; } if (/$excl_br_line/ || $exclude_br_flag) { $exclude_branch = 1; } else { $exclude_branch = 0; } } # Strip unexecuted basic block marker $count =~ s/\*$//; # :: if ($line eq "0") { # Extra data } elsif ($count eq "-") { # Uninstrumented line push(@result, 0); push(@result, 0); push(@result, $code); } else { if ($exclude_line) { push(@result, 0); push(@result, 0); } else { # Check for zero count if ($count =~ /^[#=]/) { $count = 0; } push(@result, 1); push(@result, $count); } push(@result, $code); } } } } close(INPUT); if ($exclude_flag || $exclude_br_flag) { warn("WARNING: unterminated exclusion section in $filename\n"); } return(\@result, $branches, \@functions); } # Map LLVM versions to the version of GCC gcov which they emulate. sub map_llvm_version($) { my ($ver) = @_; return 0x040200 if ($ver >= 0x030400); warn("WARNING: This version of LLVM's gcov is unknown. ". "Assuming it emulates GCC gcov version 4.2.\n"); return 0x040200; } # Return a readable version of encoded gcov version. sub version_to_str($) { my ($ver) = @_; my ($a, $b, $c); $a = $ver >> 16 & 0xff; $b = $ver >> 8 & 0xff; $c = $ver & 0xff; return "$a.$b.$c"; } # # Get the GCOV tool version. Return an integer number which represents the # GCOV version. Version numbers can be compared using standard integer # operations. # sub get_gcov_version() { local *HANDLE; my $version_string; my $result; my ($a, $b, $c) = (4, 2, 0); # Fallback version # Examples for gcov version output: # # gcov (GCC) 4.4.7 20120313 (Red Hat 4.4.7-3) # # gcov (crosstool-NG 1.18.0) 4.7.2 # # LLVM (http://llvm.org/): # LLVM version 3.4svn # # Apple LLVM version 8.0.0 (clang-800.0.38) # Optimized build. # Default target: x86_64-apple-darwin16.0.0 # Host CPU: haswell open(GCOV_PIPE, "-|", "$gcov_tool --version") or die("ERROR: cannot retrieve gcov version!\n"); local $/; $version_string = ; close(GCOV_PIPE); # Remove all bracketed information $version_string =~ s/\([^\)]*\)//g; if ($version_string =~ /(\d+)\.(\d+)(\.(\d+))?/) { ($a, $b, $c) = ($1, $2, $4); $c = 0 if (!defined($c)); } else { warn("WARNING: cannot determine gcov version - ". "assuming $a.$b.$c\n"); } $result = $a << 16 | $b << 8 | $c; if ($version_string =~ /LLVM/) { $result = map_llvm_version($result); info("Found LLVM gcov version $a.$b.$c, which emulates gcov ". "version ".version_to_str($result)."\n"); } else { info("Found gcov version: ".version_to_str($result)."\n"); } return ($result, $version_string); } # # info(printf_parameter) # # Use printf to write PRINTF_PARAMETER to stdout only when the $quiet flag # is not set. # sub info(@) { if (!$quiet) { # Print info string if (defined($output_filename) && ($output_filename eq "-")) { # Don't interfere with the .info output to STDOUT printf(STDERR @_); } else { printf(@_); } } } # # int_handler() # # Called when the script was interrupted by an INT signal (e.g. CTRl-C) # sub int_handler() { if ($cwd) { chdir($cwd); } info("Aborted.\n"); exit(1); } # # system_no_output(mode, parameters) # # Call an external program using PARAMETERS while suppressing depending on # the value of MODE: # # MODE & 1: suppress STDOUT # MODE & 2: suppress STDERR # # Return 0 on success, non-zero otherwise. # sub system_no_output($@) { my $mode = shift; my $result; local *OLD_STDERR; local *OLD_STDOUT; # Save old stdout and stderr handles ($mode & 1) && open(OLD_STDOUT, ">>&", "STDOUT"); ($mode & 2) && open(OLD_STDERR, ">>&", "STDERR"); # Redirect to /dev/null ($mode & 1) && open(STDOUT, ">", "/dev/null"); ($mode & 2) && open(STDERR, ">", "/dev/null"); debug("system(".join(' ', @_).")\n"); system(@_); $result = $?; # Close redirected handles ($mode & 1) && close(STDOUT); ($mode & 2) && close(STDERR); # Restore old handles ($mode & 1) && open(STDOUT, ">>&", "OLD_STDOUT"); ($mode & 2) && open(STDERR, ">>&", "OLD_STDERR"); return $result; } # # read_config(filename) # # Read configuration file FILENAME and return a reference to a hash containing # all valid key=value pairs found. # sub read_config($) { my $filename = $_[0]; my %result; my $key; my $value; local *HANDLE; if (!open(HANDLE, "<", $filename)) { warn("WARNING: cannot read configuration file $filename\n"); return undef; } while () { chomp; # Skip comments s/#.*//; # Remove leading blanks s/^\s+//; # Remove trailing blanks s/\s+$//; next unless length; ($key, $value) = split(/\s*=\s*/, $_, 2); if (defined($key) && defined($value)) { $result{$key} = $value; } else { warn("WARNING: malformed statement in line $. ". "of configuration file $filename\n"); } } close(HANDLE); return \%result; } # # apply_config(REF) # # REF is a reference to a hash containing the following mapping: # # key_string => var_ref # # where KEY_STRING is a keyword and VAR_REF is a reference to an associated # variable. If the global configuration hashes CONFIG or OPT_RC contain a value # for keyword KEY_STRING, VAR_REF will be assigned the value for that keyword. # sub apply_config($) { my $ref = $_[0]; foreach (keys(%{$ref})) { if (defined($opt_rc{$_})) { ${$ref->{$_}} = $opt_rc{$_}; } elsif (defined($config->{$_})) { ${$ref->{$_}} = $config->{$_}; } } } # # get_exclusion_data(filename) # # Scan specified source code file for exclusion markers and return # linenumber -> 1 # for all lines which should be excluded. # sub get_exclusion_data($) { my ($filename) = @_; my %list; my $flag = 0; local *HANDLE; if (!open(HANDLE, "<", $filename)) { warn("WARNING: could not open $filename\n"); return undef; } while () { if (/$EXCL_STOP/) { $flag = 0; } elsif (/$EXCL_START/) { $flag = 1; } if (/$excl_line/ || $flag) { $list{$.} = 1; } } close(HANDLE); if ($flag) { warn("WARNING: unterminated exclusion section in $filename\n"); } return \%list; } # # apply_exclusion_data(instr, graph) # # Remove lines from instr and graph data structures which are marked # for exclusion in the source code file. # # Return adjusted (instr, graph). # # graph : file name -> function data # function data : function name -> line data # line data : [ line1, line2, ... ] # # instr : filename -> line data # line data : [ line1, line2, ... ] # sub apply_exclusion_data($$) { my ($instr, $graph) = @_; my $filename; my %excl_data; my $excl_read_failed = 0; # Collect exclusion marker data foreach $filename (sort_uniq_lex(keys(%{$graph}), keys(%{$instr}))) { my $excl = get_exclusion_data($filename); # Skip and note if file could not be read if (!defined($excl)) { $excl_read_failed = 1; next; } # Add to collection if there are markers $excl_data{$filename} = $excl if (keys(%{$excl}) > 0); } # Warn if not all source files could be read if ($excl_read_failed) { warn("WARNING: some exclusion markers may be ignored\n"); } # Skip if no markers were found return ($instr, $graph) if (keys(%excl_data) == 0); # Apply exclusion marker data to graph foreach $filename (keys(%excl_data)) { my $function_data = $graph->{$filename}; my $excl = $excl_data{$filename}; my $function; next if (!defined($function_data)); foreach $function (keys(%{$function_data})) { my $line_data = $function_data->{$function}; my $line; my @new_data; # To be consistent with exclusion parser in non-initial # case we need to remove a function if the first line # was excluded if ($excl->{$line_data->[0]}) { delete($function_data->{$function}); next; } # Copy only lines which are not excluded foreach $line (@{$line_data}) { push(@new_data, $line) if (!$excl->{$line}); } # Store modified list if (scalar(@new_data) > 0) { $function_data->{$function} = \@new_data; } else { # All of this function was excluded delete($function_data->{$function}); } } # Check if all functions of this file were excluded if (keys(%{$function_data}) == 0) { delete($graph->{$filename}); } } # Apply exclusion marker data to instr foreach $filename (keys(%excl_data)) { my $line_data = $instr->{$filename}; my $excl = $excl_data{$filename}; my $line; my @new_data; next if (!defined($line_data)); # Copy only lines which are not excluded foreach $line (@{$line_data}) { push(@new_data, $line) if (!$excl->{$line}); } # Store modified list $instr->{$filename} = \@new_data; } return ($instr, $graph); } sub process_graphfile($$) { my ($file, $dir) = @_; my $graph_filename = $file; my $graph_dir; my $graph_basename; my $source_dir; my $base_dir; my $graph; my $instr; my $filename; local *INFO_HANDLE; info("Processing %s\n", abs2rel($file, $dir)); # Get path to data file in absolute and normalized form (begins with /, # contains no more ../ or ./) $graph_filename = solve_relative_path($cwd, $graph_filename); # Get directory and basename of data file ($graph_dir, $graph_basename) = split_filename($graph_filename); $source_dir = $graph_dir; if (is_compat($COMPAT_MODE_LIBTOOL)) { # Avoid files from .libs dirs $source_dir =~ s/\.libs$//; } # Construct base_dir for current file if ($base_directory) { $base_dir = $base_directory; } else { $base_dir = $source_dir; } # Ignore empty graph file (e.g. source file with no statement) if (-z $graph_filename) { warn("WARNING: empty $graph_filename (skipped)\n"); return; } if ($gcov_version < $GCOV_VERSION_3_4_0) { if (is_compat($COMPAT_MODE_HAMMER)) { ($instr, $graph) = read_bbg($graph_filename); } else { ($instr, $graph) = read_bb($graph_filename); } } else { ($instr, $graph) = read_gcno($graph_filename); } # Try to find base directory automatically if requested by user if ($rc_auto_base) { $base_dir = find_base_from_graph($base_dir, $instr, $graph); } ($instr, $graph) = adjust_graph_filenames($base_dir, $instr, $graph); if (!$no_markers) { # Apply exclusion marker data to graph file data ($instr, $graph) = apply_exclusion_data($instr, $graph); } # Check whether we're writing to a single file if ($output_filename) { if ($output_filename eq "-") { *INFO_HANDLE = *STDOUT; } else { # Append to output file open(INFO_HANDLE, ">>", $output_filename) or die("ERROR: cannot write to ". "$output_filename!\n"); } } else { # Open .info file for output open(INFO_HANDLE, ">", "$graph_filename.info") or die("ERROR: cannot create $graph_filename.info!\n"); } # Write test name printf(INFO_HANDLE "TN:%s\n", $test_name); foreach $filename (sort(keys(%{$instr}))) { my $funcdata = $graph->{$filename}; my $line; my $linedata; # Skip external files if requested if (!$opt_external) { if (is_external($filename)) { info(" ignoring data for external file ". "$filename\n"); next; } } print(INFO_HANDLE "SF:$filename\n"); if (defined($funcdata) && $func_coverage) { my @functions = sort {$funcdata->{$a}->[0] <=> $funcdata->{$b}->[0]} keys(%{$funcdata}); my $func; # Gather list of instrumented lines and functions foreach $func (@functions) { $linedata = $funcdata->{$func}; # Print function name and starting line print(INFO_HANDLE "FN:".$linedata->[0]. ",".filter_fn_name($func)."\n"); } # Print zero function coverage data foreach $func (@functions) { print(INFO_HANDLE "FNDA:0,". filter_fn_name($func)."\n"); } # Print function summary print(INFO_HANDLE "FNF:".scalar(@functions)."\n"); print(INFO_HANDLE "FNH:0\n"); } # Print zero line coverage data foreach $line (@{$instr->{$filename}}) { print(INFO_HANDLE "DA:$line,0\n"); } # Print line summary print(INFO_HANDLE "LF:".scalar(@{$instr->{$filename}})."\n"); print(INFO_HANDLE "LH:0\n"); print(INFO_HANDLE "end_of_record\n"); } if (!($output_filename && ($output_filename eq "-"))) { close(INFO_HANDLE); } } sub filter_fn_name($) { my ($fn) = @_; # Remove characters used internally as function name delimiters $fn =~ s/[,=]/_/g; return $fn; } sub warn_handler($) { my ($msg) = @_; warn("$tool_name: $msg"); } sub die_handler($) { my ($msg) = @_; die("$tool_name: $msg"); } # # graph_error(filename, message) # # Print message about error in graph file. If ignore_graph_error is set, return. # Otherwise abort. # sub graph_error($$) { my ($filename, $msg) = @_; if ($ignore[$ERROR_GRAPH]) { warn("WARNING: $filename: $msg - skipping\n"); return; } die("ERROR: $filename: $msg\n"); } # # graph_expect(description) # # If debug is set to a non-zero value, print the specified description of what # is expected to be read next from the graph file. # sub graph_expect($) { my ($msg) = @_; if (!$debug || !defined($msg)) { return; } print(STDERR "DEBUG: expecting $msg\n"); } # # graph_read(handle, bytes[, description, peek]) # # Read and return the specified number of bytes from handle. Return undef # if the number of bytes could not be read. If PEEK is non-zero, reset # file position after read. # sub graph_read(*$;$$) { my ($handle, $length, $desc, $peek) = @_; my $data; my $result; my $pos; graph_expect($desc); if ($peek) { $pos = tell($handle); if ($pos == -1) { warn("Could not get current file position: $!\n"); return undef; } } $result = read($handle, $data, $length); if ($debug) { my $op = $peek ? "peek" : "read"; my $ascii = ""; my $hex = ""; my $i; print(STDERR "DEBUG: $op($length)=$result: "); for ($i = 0; $i < length($data); $i++) { my $c = substr($data, $i, 1);; my $n = ord($c); $hex .= sprintf("%02x ", $n); if ($n >= 32 && $n <= 127) { $ascii .= $c; } else { $ascii .= "."; } } print(STDERR "$hex |$ascii|"); print(STDERR "\n"); } if ($peek) { if (!seek($handle, $pos, 0)) { warn("Could not set file position: $!\n"); return undef; } } if ($result != $length) { return undef; } return $data; } # # graph_skip(handle, bytes[, description]) # # Read and discard the specified number of bytes from handle. Return non-zero # if bytes could be read, zero otherwise. # sub graph_skip(*$;$) { my ($handle, $length, $desc) = @_; if (defined(graph_read($handle, $length, $desc))) { return 1; } return 0; } # # uniq(list) # # Return list without duplicate entries. # sub uniq(@) { my (@list) = @_; my @new_list; my %known; foreach my $item (@list) { next if ($known{$item}); $known{$item} = 1; push(@new_list, $item); } return @new_list; } # # sort_uniq(list) # # Return list in numerically ascending order and without duplicate entries. # sub sort_uniq(@) { my (@list) = @_; my %hash; foreach (@list) { $hash{$_} = 1; } return sort { $a <=> $b } keys(%hash); } # # sort_uniq_lex(list) # # Return list in lexically ascending order and without duplicate entries. # sub sort_uniq_lex(@) { my (@list) = @_; my %hash; foreach (@list) { $hash{$_} = 1; } return sort keys(%hash); } # # parent_dir(dir) # # Return parent directory for DIR. DIR must not contain relative path # components. # sub parent_dir($) { my ($dir) = @_; my ($v, $d, $f) = splitpath($dir, 1); my @dirs = splitdir($d); pop(@dirs); return catpath($v, catdir(@dirs), $f); } # # find_base_from_graph(base_dir, instr, graph) # # Try to determine the base directory of the graph file specified by INSTR # and GRAPH. The base directory is the base for all relative filenames in # the graph file. It is defined by the current working directory at time # of compiling the source file. # # This function implements a heuristic which relies on the following # assumptions: # - all files used for compilation are still present at their location # - the base directory is either BASE_DIR or one of its parent directories # - files by the same name are not present in multiple parent directories # sub find_base_from_graph($$$) { my ($base_dir, $instr, $graph) = @_; my $old_base; my $best_miss; my $best_base; my %rel_files; # Determine list of relative paths foreach my $filename (keys(%{$instr}), keys(%{$graph})) { next if (file_name_is_absolute($filename)); $rel_files{$filename} = 1; } # Early exit if there are no relative paths return $base_dir if (!%rel_files); do { my $miss = 0; foreach my $filename (keys(%rel_files)) { if (!-e solve_relative_path($base_dir, $filename)) { $miss++; } } debug("base_dir=$base_dir miss=$miss\n"); # Exit if we find an exact match with no misses return $base_dir if ($miss == 0); # No exact match, aim for the one with the least source file # misses if (!defined($best_base) || $miss < $best_miss) { $best_base = $base_dir; $best_miss = $miss; } # Repeat until there's no more parent directory $old_base = $base_dir; $base_dir = parent_dir($base_dir); } while ($old_base ne $base_dir); return $best_base; } # # adjust_graph_filenames(base_dir, instr, graph) # # Make relative paths in INSTR and GRAPH absolute and apply # geninfo_adjust_src_path setting to graph file data. # sub adjust_graph_filenames($$$) { my ($base_dir, $instr, $graph) = @_; foreach my $filename (keys(%{$instr})) { my $old_filename = $filename; # Convert to absolute canonical form $filename = solve_relative_path($base_dir, $filename); # Apply adjustment if (defined($adjust_src_pattern)) { $filename =~ s/$adjust_src_pattern/$adjust_src_replace/g; } if ($filename ne $old_filename) { $instr->{$filename} = delete($instr->{$old_filename}); } } foreach my $filename (keys(%{$graph})) { my $old_filename = $filename; # Make absolute # Convert to absolute canonical form $filename = solve_relative_path($base_dir, $filename); # Apply adjustment if (defined($adjust_src_pattern)) { $filename =~ s/$adjust_src_pattern/$adjust_src_replace/g; } if ($filename ne $old_filename) { $graph->{$filename} = delete($graph->{$old_filename}); } } return ($instr, $graph); } # # graph_cleanup(graph) # # Remove entries for functions with no lines. Remove duplicate line numbers. # Sort list of line numbers numerically ascending. # sub graph_cleanup($) { my ($graph) = @_; my $filename; foreach $filename (keys(%{$graph})) { my $per_file = $graph->{$filename}; my $function; foreach $function (keys(%{$per_file})) { my $lines = $per_file->{$function}; if (scalar(@$lines) == 0) { # Remove empty function delete($per_file->{$function}); next; } # Normalize list $per_file->{$function} = [ uniq(@$lines) ]; } if (scalar(keys(%{$per_file})) == 0) { # Remove empty file delete($graph->{$filename}); } } } # # graph_find_base(bb) # # Try to identify the filename which is the base source file for the # specified bb data. # sub graph_find_base($) { my ($bb) = @_; my %file_count; my $basefile; my $file; my $func; my $filedata; my $count; my $num; # Identify base name for this bb data. foreach $func (keys(%{$bb})) { $filedata = $bb->{$func}; foreach $file (keys(%{$filedata})) { $count = $file_count{$file}; # Count file occurrence $file_count{$file} = defined($count) ? $count + 1 : 1; } } $count = 0; $num = 0; foreach $file (keys(%file_count)) { if ($file_count{$file} > $count) { # The file that contains code for the most functions # is likely the base file $count = $file_count{$file}; $num = 1; $basefile = $file; } elsif ($file_count{$file} == $count) { # If more than one file could be the basefile, we # don't have a basefile $basefile = undef; } } return $basefile; } # # graph_from_bb(bb, fileorder, bb_filename, fileorder_first) # # Convert data from bb to the graph format and list of instrumented lines. # # If FILEORDER_FIRST is set, use fileorder data to determine a functions # base source file. # # Returns (instr, graph). # # bb : function name -> file data # : undef -> file order # file data : filename -> line data # line data : [ line1, line2, ... ] # # file order : function name -> [ filename1, filename2, ... ] # # graph : file name -> function data # function data : function name -> line data # line data : [ line1, line2, ... ] # # instr : filename -> line data # line data : [ line1, line2, ... ] # sub graph_from_bb($$$$) { my ($bb, $fileorder, $bb_filename, $fileorder_first) = @_; my $graph = {}; my $instr = {}; my $basefile; my $file; my $func; my $filedata; my $linedata; my $order; $basefile = graph_find_base($bb); # Create graph structure foreach $func (keys(%{$bb})) { $filedata = $bb->{$func}; $order = $fileorder->{$func}; # Account for lines in functions if (defined($basefile) && defined($filedata->{$basefile}) && !$fileorder_first) { # If the basefile contributes to this function, # account this function to the basefile. $graph->{$basefile}->{$func} = $filedata->{$basefile}; } else { # If the basefile does not contribute to this function, # account this function to the first file contributing # lines. $graph->{$order->[0]}->{$func} = $filedata->{$order->[0]}; } foreach $file (keys(%{$filedata})) { # Account for instrumented lines $linedata = $filedata->{$file}; push(@{$instr->{$file}}, @$linedata); } } # Clean up array of instrumented lines foreach $file (keys(%{$instr})) { $instr->{$file} = [ sort_uniq(@{$instr->{$file}}) ]; } return ($instr, $graph); } # # graph_add_order(fileorder, function, filename) # # Add an entry for filename to the fileorder data set for function. # sub graph_add_order($$$) { my ($fileorder, $function, $filename) = @_; my $item; my $list; $list = $fileorder->{$function}; foreach $item (@$list) { if ($item eq $filename) { return; } } push(@$list, $filename); $fileorder->{$function} = $list; } # # read_bb_word(handle[, description]) # # Read and return a word in .bb format from handle. # sub read_bb_word(*;$) { my ($handle, $desc) = @_; return graph_read($handle, 4, $desc); } # # read_bb_value(handle[, description]) # # Read a word in .bb format from handle and return the word and its integer # value. # sub read_bb_value(*;$) { my ($handle, $desc) = @_; my $word; $word = read_bb_word($handle, $desc); return undef if (!defined($word)); return ($word, unpack("V", $word)); } # # read_bb_string(handle, delimiter) # # Read and return a string in .bb format from handle up to the specified # delimiter value. # sub read_bb_string(*$) { my ($handle, $delimiter) = @_; my $word; my $value; my $string = ""; graph_expect("string"); do { ($word, $value) = read_bb_value($handle, "string or delimiter"); return undef if (!defined($value)); if ($value != $delimiter) { $string .= $word; } } while ($value != $delimiter); $string =~ s/\0//g; return $string; } # # read_bb(filename) # # Read the contents of the specified .bb file and return (instr, graph), where: # # instr : filename -> line data # line data : [ line1, line2, ... ] # # graph : filename -> file_data # file_data : function name -> line_data # line_data : [ line1, line2, ... ] # # See the gcov info pages of gcc 2.95 for a description of the .bb file format. # sub read_bb($) { my ($bb_filename) = @_; my $minus_one = 0x80000001; my $minus_two = 0x80000002; my $value; my $filename; my $function; my $bb = {}; my $fileorder = {}; my $instr; my $graph; local *HANDLE; open(HANDLE, "<", $bb_filename) or goto open_error; binmode(HANDLE); while (!eof(HANDLE)) { $value = read_bb_value(*HANDLE, "data word"); goto incomplete if (!defined($value)); if ($value == $minus_one) { # Source file name graph_expect("filename"); $filename = read_bb_string(*HANDLE, $minus_one); goto incomplete if (!defined($filename)); } elsif ($value == $minus_two) { # Function name graph_expect("function name"); $function = read_bb_string(*HANDLE, $minus_two); goto incomplete if (!defined($function)); } elsif ($value > 0) { # Line number if (!defined($filename) || !defined($function)) { warn("WARNING: unassigned line number ". "$value\n"); next; } push(@{$bb->{$function}->{$filename}}, $value); graph_add_order($fileorder, $function, $filename); } } close(HANDLE); ($instr, $graph) = graph_from_bb($bb, $fileorder, $bb_filename, 0); graph_cleanup($graph); return ($instr, $graph); open_error: graph_error($bb_filename, "could not open file"); return undef; incomplete: graph_error($bb_filename, "reached unexpected end of file"); return undef; } # # read_bbg_word(handle[, description]) # # Read and return a word in .bbg format. # sub read_bbg_word(*;$) { my ($handle, $desc) = @_; return graph_read($handle, 4, $desc); } # # read_bbg_value(handle[, description]) # # Read a word in .bbg format from handle and return its integer value. # sub read_bbg_value(*;$) { my ($handle, $desc) = @_; my $word; $word = read_bbg_word($handle, $desc); return undef if (!defined($word)); return unpack("N", $word); } # # read_bbg_string(handle) # # Read and return a string in .bbg format. # sub read_bbg_string(*) { my ($handle, $desc) = @_; my $length; my $string; graph_expect("string"); # Read string length $length = read_bbg_value($handle, "string length"); return undef if (!defined($length)); if ($length == 0) { return ""; } # Read string $string = graph_read($handle, $length, "string"); return undef if (!defined($string)); # Skip padding graph_skip($handle, 4 - $length % 4, "string padding") or return undef; return $string; } # # read_bbg_lines_record(handle, bbg_filename, bb, fileorder, filename, # function) # # Read a bbg format lines record from handle and add the relevant data to # bb and fileorder. Return filename on success, undef on error. # sub read_bbg_lines_record(*$$$$$) { my ($handle, $bbg_filename, $bb, $fileorder, $filename, $function) = @_; my $string; my $lineno; graph_expect("lines record"); # Skip basic block index graph_skip($handle, 4, "basic block index") or return undef; while (1) { # Read line number $lineno = read_bbg_value($handle, "line number"); return undef if (!defined($lineno)); if ($lineno == 0) { # Got a marker for a new filename graph_expect("filename"); $string = read_bbg_string($handle); return undef if (!defined($string)); # Check for end of record if ($string eq "") { return $filename; } $filename = $string; if (!exists($bb->{$function}->{$filename})) { $bb->{$function}->{$filename} = []; } next; } # Got an actual line number if (!defined($filename)) { warn("WARNING: unassigned line number in ". "$bbg_filename\n"); next; } push(@{$bb->{$function}->{$filename}}, $lineno); graph_add_order($fileorder, $function, $filename); } } # # read_bbg(filename) # # Read the contents of the specified .bbg file and return the following mapping: # graph: filename -> file_data # file_data: function name -> line_data # line_data: [ line1, line2, ... ] # # See the gcov-io.h file in the SLES 9 gcc 3.3.3 source code for a description # of the .bbg format. # sub read_bbg($) { my ($bbg_filename) = @_; my $file_magic = 0x67626267; my $tag_function = 0x01000000; my $tag_lines = 0x01450000; my $word; my $tag; my $length; my $function; my $filename; my $bb = {}; my $fileorder = {}; my $instr; my $graph; local *HANDLE; open(HANDLE, "<", $bbg_filename) or goto open_error; binmode(HANDLE); # Read magic $word = read_bbg_value(*HANDLE, "file magic"); goto incomplete if (!defined($word)); # Check magic if ($word != $file_magic) { goto magic_error; } # Skip version graph_skip(*HANDLE, 4, "version") or goto incomplete; while (!eof(HANDLE)) { # Read record tag $tag = read_bbg_value(*HANDLE, "record tag"); goto incomplete if (!defined($tag)); # Read record length $length = read_bbg_value(*HANDLE, "record length"); goto incomplete if (!defined($tag)); if ($tag == $tag_function) { graph_expect("function record"); # Read function name graph_expect("function name"); $function = read_bbg_string(*HANDLE); goto incomplete if (!defined($function)); $filename = undef; # Skip function checksum graph_skip(*HANDLE, 4, "function checksum") or goto incomplete; } elsif ($tag == $tag_lines) { # Read lines record $filename = read_bbg_lines_record(HANDLE, $bbg_filename, $bb, $fileorder, $filename, $function); goto incomplete if (!defined($filename)); } else { # Skip record contents graph_skip(*HANDLE, $length, "unhandled record") or goto incomplete; } } close(HANDLE); ($instr, $graph) = graph_from_bb($bb, $fileorder, $bbg_filename, 0); graph_cleanup($graph); return ($instr, $graph); open_error: graph_error($bbg_filename, "could not open file"); return undef; incomplete: graph_error($bbg_filename, "reached unexpected end of file"); return undef; magic_error: graph_error($bbg_filename, "found unrecognized bbg file magic"); return undef; } # # read_gcno_word(handle[, description, peek]) # # Read and return a word in .gcno format. # sub read_gcno_word(*;$$) { my ($handle, $desc, $peek) = @_; return graph_read($handle, 4, $desc, $peek); } # # read_gcno_value(handle, big_endian[, description, peek]) # # Read a word in .gcno format from handle and return its integer value # according to the specified endianness. If PEEK is non-zero, reset file # position after read. # sub read_gcno_value(*$;$$) { my ($handle, $big_endian, $desc, $peek) = @_; my $word; my $pos; $word = read_gcno_word($handle, $desc, $peek); return undef if (!defined($word)); if ($big_endian) { return unpack("N", $word); } else { return unpack("V", $word); } } # # read_gcno_string(handle, big_endian) # # Read and return a string in .gcno format. # sub read_gcno_string(*$) { my ($handle, $big_endian) = @_; my $length; my $string; graph_expect("string"); # Read string length $length = read_gcno_value($handle, $big_endian, "string length"); return undef if (!defined($length)); if ($length == 0) { return ""; } $length *= 4; # Read string $string = graph_read($handle, $length, "string and padding"); return undef if (!defined($string)); $string =~ s/\0//g; return $string; } # # read_gcno_lines_record(handle, gcno_filename, bb, fileorder, filename, # function, big_endian) # # Read a gcno format lines record from handle and add the relevant data to # bb and fileorder. Return filename on success, undef on error. # sub read_gcno_lines_record(*$$$$$$) { my ($handle, $gcno_filename, $bb, $fileorder, $filename, $function, $big_endian) = @_; my $string; my $lineno; graph_expect("lines record"); # Skip basic block index graph_skip($handle, 4, "basic block index") or return undef; while (1) { # Read line number $lineno = read_gcno_value($handle, $big_endian, "line number"); return undef if (!defined($lineno)); if ($lineno == 0) { # Got a marker for a new filename graph_expect("filename"); $string = read_gcno_string($handle, $big_endian); return undef if (!defined($string)); # Check for end of record if ($string eq "") { return $filename; } $filename = $string; if (!exists($bb->{$function}->{$filename})) { $bb->{$function}->{$filename} = []; } next; } # Got an actual line number if (!defined($filename)) { warn("WARNING: unassigned line number in ". "$gcno_filename\n"); next; } # Add to list push(@{$bb->{$function}->{$filename}}, $lineno); graph_add_order($fileorder, $function, $filename); } } # # determine_gcno_split_crc(handle, big_endian, rec_length, version) # # Determine if HANDLE refers to a .gcno file with a split checksum function # record format. Return non-zero in case of split checksum format, zero # otherwise, undef in case of read error. # sub determine_gcno_split_crc($$$$) { my ($handle, $big_endian, $rec_length, $version) = @_; my $strlen; my $overlong_string; return 1 if ($version >= $GCOV_VERSION_4_7_0); return 1 if (is_compat($COMPAT_MODE_SPLIT_CRC)); # Heuristic: # Decide format based on contents of next word in record: # - pre-gcc 4.7 # This is the function name length / 4 which should be # less than the remaining record length # - gcc 4.7 # This is a checksum, likely with high-order bits set, # resulting in a large number $strlen = read_gcno_value($handle, $big_endian, undef, 1); return undef if (!defined($strlen)); $overlong_string = 1 if ($strlen * 4 >= $rec_length - 12); if ($overlong_string) { if (is_compat_auto($COMPAT_MODE_SPLIT_CRC)) { info("Auto-detected compatibility mode for split ". "checksum .gcno file format\n"); return 1; } else { # Sanity check warn("Found overlong string in function record: ". "try '--compat split_crc'\n"); } } return 0; } # # read_gcno_function_record(handle, graph, big_endian, rec_length, version) # # Read a gcno format function record from handle and add the relevant data # to graph. Return (filename, function, artificial) on success, undef on error. # sub read_gcno_function_record(*$$$$$) { my ($handle, $bb, $fileorder, $big_endian, $rec_length, $version) = @_; my $filename; my $function; my $lineno; my $lines; my $artificial; graph_expect("function record"); # Skip ident and checksum graph_skip($handle, 8, "function ident and checksum") or return undef; # Determine if this is a function record with split checksums if (!defined($gcno_split_crc)) { $gcno_split_crc = determine_gcno_split_crc($handle, $big_endian, $rec_length, $version); return undef if (!defined($gcno_split_crc)); } # Skip cfg checksum word in case of split checksums graph_skip($handle, 4, "function cfg checksum") if ($gcno_split_crc); # Read function name graph_expect("function name"); $function = read_gcno_string($handle, $big_endian); return undef if (!defined($function)); if ($version >= $GCOV_VERSION_8_0_0) { $artificial = read_gcno_value($handle, $big_endian, "compiler-generated entity flag"); return undef if (!defined($artificial)); } # Read filename graph_expect("filename"); $filename = read_gcno_string($handle, $big_endian); return undef if (!defined($filename)); # Read first line number $lineno = read_gcno_value($handle, $big_endian, "initial line number"); return undef if (!defined($lineno)); # Skip column and ending line number if ($version >= $GCOV_VERSION_8_0_0) { graph_skip($handle, 4, "column number") or return undef; graph_skip($handle, 4, "ending line number") or return undef; } # Add to list push(@{$bb->{$function}->{$filename}}, $lineno); graph_add_order($fileorder, $function, $filename); return ($filename, $function, $artificial); } # # map_gcno_version # # Map version number as found in .gcno files to the format used in geninfo. # sub map_gcno_version($) { my ($version) = @_; my ($a, $b, $c); my ($major, $minor); $a = $version >> 24; $b = $version >> 16 & 0xff; $c = $version >> 8 & 0xff; if ($a < ord('A')) { $major = $a - ord('0'); $minor = ($b - ord('0')) * 10 + $c - ord('0'); } else { $major = ($a - ord('A')) * 10 + $b - ord('0'); $minor = $c - ord('0'); } return $major << 16 | $minor << 8; } sub remove_fn_from_hash($$) { my ($hash, $fns) = @_; foreach my $fn (@$fns) { delete($hash->{$fn}); } } # # read_gcno(filename) # # Read the contents of the specified .gcno file and return the following # mapping: # graph: filename -> file_data # file_data: function name -> line_data # line_data: [ line1, line2, ... ] # # See the gcov-io.h file in the gcc 3.3 source code for a description of # the .gcno format. # sub read_gcno($) { my ($gcno_filename) = @_; my $file_magic = 0x67636e6f; my $tag_function = 0x01000000; my $tag_lines = 0x01450000; my $big_endian; my $word; my $tag; my $length; my $filename; my $function; my $bb = {}; my $fileorder = {}; my $instr; my $graph; my $filelength; my $version; my $artificial; my @artificial_fns; local *HANDLE; open(HANDLE, "<", $gcno_filename) or goto open_error; $filelength = (stat(HANDLE))[7]; binmode(HANDLE); # Read magic $word = read_gcno_word(*HANDLE, "file magic"); goto incomplete if (!defined($word)); # Determine file endianness if (unpack("N", $word) == $file_magic) { $big_endian = 1; } elsif (unpack("V", $word) == $file_magic) { $big_endian = 0; } else { goto magic_error; } # Read version $version = read_gcno_value(*HANDLE, $big_endian, "compiler version"); $version = map_gcno_version($version); debug(sprintf("found version 0x%08x\n", $version)); # Skip stamp graph_skip(*HANDLE, 4, "file timestamp") or goto incomplete; if ($version >= $GCOV_VERSION_8_0_0) { graph_skip(*HANDLE, 4, "support unexecuted blocks flag") or goto incomplete; } while (!eof(HANDLE)) { my $next_pos; my $curr_pos; # Read record tag $tag = read_gcno_value(*HANDLE, $big_endian, "record tag"); goto incomplete if (!defined($tag)); # Read record length $length = read_gcno_value(*HANDLE, $big_endian, "record length"); goto incomplete if (!defined($length)); # Convert length to bytes $length *= 4; # Calculate start of next record $next_pos = tell(HANDLE); goto tell_error if ($next_pos == -1); $next_pos += $length; # Catch garbage at the end of a gcno file if ($next_pos > $filelength) { debug("Overlong record: file_length=$filelength ". "rec_length=$length\n"); warn("WARNING: $gcno_filename: Overlong record at end ". "of file!\n"); last; } # Process record if ($tag == $tag_function) { ($filename, $function, $artificial) = read_gcno_function_record( *HANDLE, $bb, $fileorder, $big_endian, $length, $version); goto incomplete if (!defined($function)); push(@artificial_fns, $function) if ($artificial); } elsif ($tag == $tag_lines) { # Read lines record $filename = read_gcno_lines_record(*HANDLE, $gcno_filename, $bb, $fileorder, $filename, $function, $big_endian); goto incomplete if (!defined($filename)); } else { # Skip record contents graph_skip(*HANDLE, $length, "unhandled record") or goto incomplete; } # Ensure that we are at the start of the next record $curr_pos = tell(HANDLE); goto tell_error if ($curr_pos == -1); next if ($curr_pos == $next_pos); goto record_error if ($curr_pos > $next_pos); graph_skip(*HANDLE, $next_pos - $curr_pos, "unhandled record content") or goto incomplete; } close(HANDLE); # Remove artificial functions from result data remove_fn_from_hash($bb, \@artificial_fns); remove_fn_from_hash($fileorder, \@artificial_fns); ($instr, $graph) = graph_from_bb($bb, $fileorder, $gcno_filename, 1); graph_cleanup($graph); return ($instr, $graph); open_error: graph_error($gcno_filename, "could not open file"); return undef; incomplete: graph_error($gcno_filename, "reached unexpected end of file"); return undef; magic_error: graph_error($gcno_filename, "found unrecognized gcno file magic"); return undef; tell_error: graph_error($gcno_filename, "could not determine file position"); return undef; record_error: graph_error($gcno_filename, "found unrecognized record format"); return undef; } sub debug($) { my ($msg) = @_; return if (!$debug); print(STDERR "DEBUG: $msg"); } # # get_gcov_capabilities # # Determine the list of available gcov options. # sub get_gcov_capabilities() { my $help = `$gcov_tool --help`; my %capabilities; my %short_option_translations = ( 'a' => 'all-blocks', 'b' => 'branch-probabilities', 'c' => 'branch-counts', 'f' => 'function-summaries', 'h' => 'help', 'l' => 'long-file-names', 'n' => 'no-output', 'o' => 'object-directory', 'p' => 'preserve-paths', 'u' => 'unconditional-branches', 'v' => 'version', 'x' => 'hash-filenames', ); foreach (split(/\n/, $help)) { my $capability; if (/--(\S+)/) { $capability = $1; } else { # If the line provides a short option, translate it. next if (!/^\s*-(\S)\s/); $capability = $short_option_translations{$1}; next if not defined($capability); } next if ($capability eq 'help'); next if ($capability eq 'version'); next if ($capability eq 'object-directory'); $capabilities{$capability} = 1; debug("gcov has capability '$capability'\n"); } return \%capabilities; } # # parse_ignore_errors(@ignore_errors) # # Parse user input about which errors to ignore. # sub parse_ignore_errors(@) { my (@ignore_errors) = @_; my @items; my $item; return if (!@ignore_errors); foreach $item (@ignore_errors) { $item =~ s/\s//g; if ($item =~ /,/) { # Split and add comma-separated parameters push(@items, split(/,/, $item)); } else { # Add single parameter push(@items, $item); } } foreach $item (@items) { my $item_id = $ERROR_ID{lc($item)}; if (!defined($item_id)) { die("ERROR: unknown argument for --ignore-errors: ". "$item\n"); } $ignore[$item_id] = 1; } } # # is_external(filename) # # Determine if a file is located outside of the specified data directories. # sub is_external($) { my ($filename) = @_; my $dir; foreach $dir (@internal_dirs) { return 0 if ($filename =~ /^\Q$dir\/\E/); } return 1; } # # compat_name(mode) # # Return the name of compatibility mode MODE. # sub compat_name($) { my ($mode) = @_; my $name = $COMPAT_MODE_TO_NAME{$mode}; return $name if (defined($name)); return ""; } # # parse_compat_modes(opt) # # Determine compatibility mode settings. # sub parse_compat_modes($) { my ($opt) = @_; my @opt_list; my %specified; # Initialize with defaults %compat_value = %COMPAT_MODE_DEFAULTS; # Add old style specifications if (defined($opt_compat_libtool)) { $compat_value{$COMPAT_MODE_LIBTOOL} = $opt_compat_libtool ? $COMPAT_VALUE_ON : $COMPAT_VALUE_OFF; } # Parse settings if (defined($opt)) { @opt_list = split(/\s*,\s*/, $opt); } foreach my $directive (@opt_list) { my ($mode, $value); # Either # mode=off|on|auto or # mode (implies on) if ($directive !~ /^(\w+)=(\w+)$/ && $directive !~ /^(\w+)$/) { die("ERROR: Unknown compatibility mode specification: ". "$directive!\n"); } # Determine mode $mode = $COMPAT_NAME_TO_MODE{lc($1)}; if (!defined($mode)) { die("ERROR: Unknown compatibility mode '$1'!\n"); } $specified{$mode} = 1; # Determine value if (defined($2)) { $value = $COMPAT_NAME_TO_VALUE{lc($2)}; if (!defined($value)) { die("ERROR: Unknown compatibility mode ". "value '$2'!\n"); } } else { $value = $COMPAT_VALUE_ON; } $compat_value{$mode} = $value; } # Perform auto-detection foreach my $mode (sort(keys(%compat_value))) { my $value = $compat_value{$mode}; my $is_autodetect = ""; my $name = compat_name($mode); if ($value == $COMPAT_VALUE_AUTO) { my $autodetect = $COMPAT_MODE_AUTO{$mode}; if (!defined($autodetect)) { die("ERROR: No auto-detection for ". "mode '$name' available!\n"); } if (ref($autodetect) eq "CODE") { $value = &$autodetect(); $compat_value{$mode} = $value; $is_autodetect = " (auto-detected)"; } } if ($specified{$mode}) { if ($value == $COMPAT_VALUE_ON) { info("Enabling compatibility mode ". "'$name'$is_autodetect\n"); } elsif ($value == $COMPAT_VALUE_OFF) { info("Disabling compatibility mode ". "'$name'$is_autodetect\n"); } else { info("Using delayed auto-detection for ". "compatibility mode ". "'$name'\n"); } } } } sub compat_hammer_autodetect() { if ($gcov_version_string =~ /suse/i && $gcov_version == 0x30303 || $gcov_version_string =~ /mandrake/i && $gcov_version == 0x30302) { info("Auto-detected compatibility mode for GCC 3.3 (hammer)\n"); return $COMPAT_VALUE_ON; } return $COMPAT_VALUE_OFF; } # # is_compat(mode) # # Return non-zero if compatibility mode MODE is enabled. # sub is_compat($) { my ($mode) = @_; return 1 if ($compat_value{$mode} == $COMPAT_VALUE_ON); return 0; } # # is_compat_auto(mode) # # Return non-zero if compatibility mode MODE is set to auto-detect. # sub is_compat_auto($) { my ($mode) = @_; return 1 if ($compat_value{$mode} == $COMPAT_VALUE_AUTO); return 0; } lcov-1.14/bin/genhtml0000755000000000000000000042777213435751076013272 0ustar rootroot#!/usr/bin/env perl # # Copyright (c) International Business Machines Corp., 2002,2012 # # This program is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 2 of the License, or (at # your option) any later version. # # This program is distributed in the hope that it will be useful, but # WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU # General Public License for more details. # # You should have received a copy of the GNU General Public License # along with this program; if not, write to the Free Software # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA # # # genhtml # # This script generates HTML output from .info files as created by the # geninfo script. Call it with --help and refer to the genhtml man page # to get information on usage and available options. # # # History: # 2002-08-23 created by Peter Oberparleiter # IBM Lab Boeblingen # based on code by Manoj Iyer and # Megan Bock # IBM Austin # 2002-08-27 / Peter Oberparleiter: implemented frame view # 2002-08-29 / Peter Oberparleiter: implemented test description filtering # so that by default only descriptions for test cases which # actually hit some source lines are kept # 2002-09-05 / Peter Oberparleiter: implemented --no-sourceview # 2002-09-05 / Mike Kobler: One of my source file paths includes a "+" in # the directory name. I found that genhtml.pl died when it # encountered it. I was able to fix the problem by modifying # the string with the escape character before parsing it. # 2002-10-26 / Peter Oberparleiter: implemented --num-spaces # 2003-04-07 / Peter Oberparleiter: fixed bug which resulted in an error # when trying to combine .info files containing data without # a test name # 2003-04-10 / Peter Oberparleiter: extended fix by Mike to also cover # other special characters # 2003-04-30 / Peter Oberparleiter: made info write to STDERR, not STDOUT # 2003-07-10 / Peter Oberparleiter: added line checksum support # 2004-08-09 / Peter Oberparleiter: added configuration file support # 2005-03-04 / Cal Pierog: added legend to HTML output, fixed coloring of # "good coverage" background # 2006-03-18 / Marcus Boerger: added --custom-intro, --custom-outro and # overwrite --no-prefix if --prefix is present # 2006-03-20 / Peter Oberparleiter: changes to custom_* function (rename # to html_prolog/_epilog, minor modifications to implementation), # changed prefix/noprefix handling to be consistent with current # logic # 2006-03-20 / Peter Oberparleiter: added --html-extension option # 2008-07-14 / Tom Zoerner: added --function-coverage command line option; # added function table to source file page # 2008-08-13 / Peter Oberparleiter: modified function coverage # implementation (now enabled per default), # introduced sorting option (enabled per default) # use strict; use warnings; use File::Basename; use File::Temp qw(tempfile); use Getopt::Long; use Digest::MD5 qw(md5_base64); use Cwd qw/abs_path cwd/; # Global constants our $title = "LCOV - code coverage report"; our $tool_dir = abs_path(dirname($0)); our $lcov_version = "LCOV version 1.14"; our $lcov_url = "http://ltp.sourceforge.net/coverage/lcov.php"; our $tool_name = basename($0); # Specify coverage rate default precision our $default_precision = 1; # Specify coverage rate limits (in %) for classifying file entries # HI: $hi_limit <= rate <= 100 graph color: green # MED: $med_limit <= rate < $hi_limit graph color: orange # LO: 0 <= rate < $med_limit graph color: red # For line coverage/all coverage types if not specified our $hi_limit = 90; our $med_limit = 75; # For function coverage our $fn_hi_limit; our $fn_med_limit; # For branch coverage our $br_hi_limit; our $br_med_limit; # Width of overview image our $overview_width = 80; # Resolution of overview navigation: this number specifies the maximum # difference in lines between the position a user selected from the overview # and the position the source code window is scrolled to. our $nav_resolution = 4; # Clicking a line in the overview image should show the source code view at # a position a bit further up so that the requested line is not the first # line in the window. This number specifies that offset in lines. our $nav_offset = 10; # Clicking on a function name should show the source code at a position a # few lines before the first line of code of that function. This number # specifies that offset in lines. our $func_offset = 2; our $overview_title = "top level"; # Width for line coverage information in the source code view our $line_field_width = 12; # Width for branch coverage information in the source code view our $br_field_width = 16; # Internal Constants # Header types our $HDR_DIR = 0; our $HDR_FILE = 1; our $HDR_SOURCE = 2; our $HDR_TESTDESC = 3; our $HDR_FUNC = 4; # Sort types our $SORT_FILE = 0; our $SORT_LINE = 1; our $SORT_FUNC = 2; our $SORT_BRANCH = 3; # Fileview heading types our $HEAD_NO_DETAIL = 1; our $HEAD_DETAIL_HIDDEN = 2; our $HEAD_DETAIL_SHOWN = 3; # Additional offsets used when converting branch coverage data to HTML our $BR_LEN = 3; our $BR_OPEN = 4; our $BR_CLOSE = 5; # Branch data combination types our $BR_SUB = 0; our $BR_ADD = 1; # Error classes which users may specify to ignore during processing our $ERROR_SOURCE = 0; our %ERROR_ID = ( "source" => $ERROR_SOURCE, ); # Data related prototypes sub print_usage(*); sub gen_html(); sub html_create($$); sub process_dir($); sub process_file($$$); sub info(@); sub read_info_file($); sub get_info_entry($); sub set_info_entry($$$$$$$$$;$$$$$$); sub get_prefix($@); sub shorten_prefix($); sub get_dir_list(@); sub get_relative_base_path($); sub read_testfile($); sub get_date_string(); sub create_sub_dir($); sub subtract_counts($$); sub add_counts($$); sub apply_baseline($$); sub remove_unused_descriptions(); sub get_found_and_hit($); sub get_affecting_tests($$$); sub combine_info_files($$); sub merge_checksums($$$); sub combine_info_entries($$$); sub apply_prefix($@); sub system_no_output($@); sub read_config($); sub apply_config($); sub get_html_prolog($); sub get_html_epilog($); sub write_dir_page($$$$$$$$$$$$$$$$$); sub classify_rate($$$$); sub combine_brcount($$$;$); sub get_br_found_and_hit($); sub warn_handler($); sub die_handler($); sub parse_ignore_errors(@); sub parse_dir_prefix(@); sub rate($$;$$$); # HTML related prototypes sub escape_html($); sub get_bar_graph_code($$$); sub write_png_files(); sub write_htaccess_file(); sub write_css_file(); sub write_description_file($$$$$$$); sub write_function_table(*$$$$$$$$$$); sub write_html(*$); sub write_html_prolog(*$$); sub write_html_epilog(*$;$); sub write_header(*$$$$$$$$$$); sub write_header_prolog(*$); sub write_header_line(*@); sub write_header_epilog(*$); sub write_file_table(*$$$$$$$); sub write_file_table_prolog(*$@); sub write_file_table_entry(*$$$@); sub write_file_table_detail_entry(*$@); sub write_file_table_epilog(*); sub write_test_table_prolog(*$); sub write_test_table_entry(*$$); sub write_test_table_epilog(*); sub write_source($$$$$$$); sub write_source_prolog(*); sub write_source_line(*$$$$$); sub write_source_epilog(*); sub write_frameset(*$$$); sub write_overview_line(*$$$); sub write_overview(*$$$$); # External prototype (defined in genpng) sub gen_png($$$@); # Global variables & initialization our %info_data; # Hash containing all data from .info file our @opt_dir_prefix; # Array of prefixes to remove from all sub directories our @dir_prefix; our %test_description; # Hash containing test descriptions if available our $date = get_date_string(); our @info_filenames; # List of .info files to use as data source our $test_title; # Title for output as written to each page header our $output_directory; # Name of directory in which to store output our $base_filename; # Optional name of file containing baseline data our $desc_filename; # Name of file containing test descriptions our $css_filename; # Optional name of external stylesheet file to use our $quiet; # If set, suppress information messages our $help; # Help option flag our $version; # Version option flag our $show_details; # If set, generate detailed directory view our $no_prefix; # If set, do not remove filename prefix our $func_coverage; # If set, generate function coverage statistics our $no_func_coverage; # Disable func_coverage our $br_coverage; # If set, generate branch coverage statistics our $no_br_coverage; # Disable br_coverage our $sort = 1; # If set, provide directory listings with sorted entries our $no_sort; # Disable sort our $frames; # If set, use frames for source code view our $keep_descriptions; # If set, do not remove unused test case descriptions our $no_sourceview; # If set, do not create a source code view for each file our $highlight; # If set, highlight lines covered by converted data only our $legend; # If set, include legend in output our $tab_size = 8; # Number of spaces to use in place of tab our $config; # Configuration file contents our $html_prolog_file; # Custom HTML prolog file (up to and including ) our $html_epilog_file; # Custom HTML epilog file (from onwards) our $html_prolog; # Actual HTML prolog our $html_epilog; # Actual HTML epilog our $html_ext = "html"; # Extension for generated HTML files our $html_gzip = 0; # Compress with gzip our $demangle_cpp = 0; # Demangle C++ function names our @opt_ignore_errors; # Ignore certain error classes during processing our @ignore; our $opt_config_file; # User-specified configuration file location our %opt_rc; our $opt_missed; # List/sort lines by missed counts our $charset = "UTF-8"; # Default charset for HTML pages our @fileview_sortlist; our @fileview_sortname = ("", "-sort-l", "-sort-f", "-sort-b"); our @funcview_sortlist; our @rate_name = ("Lo", "Med", "Hi"); our @rate_png = ("ruby.png", "amber.png", "emerald.png"); our $lcov_func_coverage = 1; our $lcov_branch_coverage = 0; our $rc_desc_html = 0; # lcovrc: genhtml_desc_html our $cwd = cwd(); # Current working directory # # Code entry point # $SIG{__WARN__} = \&warn_handler; $SIG{__DIE__} = \&die_handler; # Check command line for a configuration file name Getopt::Long::Configure("pass_through", "no_auto_abbrev"); GetOptions("config-file=s" => \$opt_config_file, "rc=s%" => \%opt_rc); Getopt::Long::Configure("default"); { # Remove spaces around rc options my %new_opt_rc; while (my ($key, $value) = each(%opt_rc)) { $key =~ s/^\s+|\s+$//g; $value =~ s/^\s+|\s+$//g; $new_opt_rc{$key} = $value; } %opt_rc = %new_opt_rc; } # Read configuration file if available if (defined($opt_config_file)) { $config = read_config($opt_config_file); } elsif (defined($ENV{"HOME"}) && (-r $ENV{"HOME"}."/.lcovrc")) { $config = read_config($ENV{"HOME"}."/.lcovrc"); } elsif (-r "/etc/lcovrc") { $config = read_config("/etc/lcovrc"); } elsif (-r "/usr/local/etc/lcovrc") { $config = read_config("/usr/local/etc/lcovrc"); } if ($config || %opt_rc) { # Copy configuration file and --rc values to variables apply_config({ "genhtml_css_file" => \$css_filename, "genhtml_hi_limit" => \$hi_limit, "genhtml_med_limit" => \$med_limit, "genhtml_line_field_width" => \$line_field_width, "genhtml_overview_width" => \$overview_width, "genhtml_nav_resolution" => \$nav_resolution, "genhtml_nav_offset" => \$nav_offset, "genhtml_keep_descriptions" => \$keep_descriptions, "genhtml_no_prefix" => \$no_prefix, "genhtml_no_source" => \$no_sourceview, "genhtml_num_spaces" => \$tab_size, "genhtml_highlight" => \$highlight, "genhtml_legend" => \$legend, "genhtml_html_prolog" => \$html_prolog_file, "genhtml_html_epilog" => \$html_epilog_file, "genhtml_html_extension" => \$html_ext, "genhtml_html_gzip" => \$html_gzip, "genhtml_precision" => \$default_precision, "genhtml_function_hi_limit" => \$fn_hi_limit, "genhtml_function_med_limit" => \$fn_med_limit, "genhtml_function_coverage" => \$func_coverage, "genhtml_branch_hi_limit" => \$br_hi_limit, "genhtml_branch_med_limit" => \$br_med_limit, "genhtml_branch_coverage" => \$br_coverage, "genhtml_branch_field_width" => \$br_field_width, "genhtml_sort" => \$sort, "genhtml_charset" => \$charset, "genhtml_desc_html" => \$rc_desc_html, "genhtml_demangle_cpp" => \$demangle_cpp, "genhtml_missed" => \$opt_missed, "lcov_function_coverage" => \$lcov_func_coverage, "lcov_branch_coverage" => \$lcov_branch_coverage, }); } # Copy related values if not specified $fn_hi_limit = $hi_limit if (!defined($fn_hi_limit)); $fn_med_limit = $med_limit if (!defined($fn_med_limit)); $br_hi_limit = $hi_limit if (!defined($br_hi_limit)); $br_med_limit = $med_limit if (!defined($br_med_limit)); $func_coverage = $lcov_func_coverage if (!defined($func_coverage)); $br_coverage = $lcov_branch_coverage if (!defined($br_coverage)); # Parse command line options if (!GetOptions("output-directory|o=s" => \$output_directory, "title|t=s" => \$test_title, "description-file|d=s" => \$desc_filename, "keep-descriptions|k" => \$keep_descriptions, "css-file|c=s" => \$css_filename, "baseline-file|b=s" => \$base_filename, "prefix|p=s" => \@opt_dir_prefix, "num-spaces=i" => \$tab_size, "no-prefix" => \$no_prefix, "no-sourceview" => \$no_sourceview, "show-details|s" => \$show_details, "frames|f" => \$frames, "highlight" => \$highlight, "legend" => \$legend, "quiet|q" => \$quiet, "help|h|?" => \$help, "version|v" => \$version, "html-prolog=s" => \$html_prolog_file, "html-epilog=s" => \$html_epilog_file, "html-extension=s" => \$html_ext, "html-gzip" => \$html_gzip, "function-coverage" => \$func_coverage, "no-function-coverage" => \$no_func_coverage, "branch-coverage" => \$br_coverage, "no-branch-coverage" => \$no_br_coverage, "sort" => \$sort, "no-sort" => \$no_sort, "demangle-cpp" => \$demangle_cpp, "ignore-errors=s" => \@opt_ignore_errors, "config-file=s" => \$opt_config_file, "rc=s%" => \%opt_rc, "precision=i" => \$default_precision, "missed" => \$opt_missed, )) { print(STDERR "Use $tool_name --help to get usage information\n"); exit(1); } else { # Merge options if ($no_func_coverage) { $func_coverage = 0; } if ($no_br_coverage) { $br_coverage = 0; } # Merge sort options if ($no_sort) { $sort = 0; } } @info_filenames = @ARGV; # Check for help option if ($help) { print_usage(*STDOUT); exit(0); } # Check for version option if ($version) { print("$tool_name: $lcov_version\n"); exit(0); } # Determine which errors the user wants us to ignore parse_ignore_errors(@opt_ignore_errors); # Split the list of prefixes if needed parse_dir_prefix(@opt_dir_prefix); # Check for info filename if (!@info_filenames) { die("No filename specified\n". "Use $tool_name --help to get usage information\n"); } # Generate a title if none is specified if (!$test_title) { if (scalar(@info_filenames) == 1) { # Only one filename specified, use it as title $test_title = basename($info_filenames[0]); } else { # More than one filename specified, used default title $test_title = "unnamed"; } } # Make sure css_filename is an absolute path (in case we're changing # directories) if ($css_filename) { if (!($css_filename =~ /^\/(.*)$/)) { $css_filename = $cwd."/".$css_filename; } } # Make sure tab_size is within valid range if ($tab_size < 1) { print(STDERR "ERROR: invalid number of spaces specified: ". "$tab_size!\n"); exit(1); } # Get HTML prolog and epilog $html_prolog = get_html_prolog($html_prolog_file); $html_epilog = get_html_epilog($html_epilog_file); # Issue a warning if --no-sourceview is enabled together with --frames if ($no_sourceview && defined($frames)) { warn("WARNING: option --frames disabled because --no-sourceview ". "was specified!\n"); $frames = undef; } # Issue a warning if --no-prefix is enabled together with --prefix if ($no_prefix && @dir_prefix) { warn("WARNING: option --prefix disabled because --no-prefix was ". "specified!\n"); @dir_prefix = undef; } @fileview_sortlist = ($SORT_FILE); @funcview_sortlist = ($SORT_FILE); if ($sort) { push(@fileview_sortlist, $SORT_LINE); push(@fileview_sortlist, $SORT_FUNC) if ($func_coverage); push(@fileview_sortlist, $SORT_BRANCH) if ($br_coverage); push(@funcview_sortlist, $SORT_LINE); } if ($frames) { # Include genpng code needed for overview image generation do("$tool_dir/genpng"); } # Ensure that the c++filt tool is available when using --demangle-cpp if ($demangle_cpp) { if (system_no_output(3, "c++filt", "--version")) { die("ERROR: could not find c++filt tool needed for ". "--demangle-cpp\n"); } } # Make sure precision is within valid range if ($default_precision < 1 || $default_precision > 4) { die("ERROR: specified precision is out of range (1 to 4)\n"); } # Make sure output_directory exists, create it if necessary if ($output_directory) { stat($output_directory); if (! -e _) { create_sub_dir($output_directory); } } # Do something gen_html(); exit(0); # # print_usage(handle) # # Print usage information. # sub print_usage(*) { local *HANDLE = $_[0]; print(HANDLE <{$filename}; my $funcdata = $data->{"func"}; my $sumfnccount = $data->{"sumfnc"}; if (defined($funcdata)) { foreach my $func_name (keys(%{$funcdata})) { $fns{$func_name} = 1; } } if (defined($sumfnccount)) { foreach my $func_name (keys(%{$sumfnccount})) { $fns{$func_name} = 1; } } } @result = keys(%fns); return \@result; } # # rename_functions(info, conv) # # Rename all function names in INFO according to CONV: OLD_NAME -> NEW_NAME. # In case two functions demangle to the same name, assume that they are # different object code implementations for the same source function. # sub rename_functions($$) { my ($info, $conv) = @_; foreach my $filename (keys(%{$info})) { my $data = $info->{$filename}; my $funcdata; my $testfncdata; my $sumfnccount; my %newfuncdata; my %newsumfnccount; my $f_found; my $f_hit; # funcdata: function name -> line number $funcdata = $data->{"func"}; foreach my $fn (keys(%{$funcdata})) { my $cn = $conv->{$fn}; # Abort if two functions on different lines map to the # same demangled name. if (defined($newfuncdata{$cn}) && $newfuncdata{$cn} != $funcdata->{$fn}) { die("ERROR: Demangled function name $cn ". "maps to different lines (". $newfuncdata{$cn}." vs ". $funcdata->{$fn}.") in $filename\n"); } $newfuncdata{$cn} = $funcdata->{$fn}; } $data->{"func"} = \%newfuncdata; # testfncdata: test name -> testfnccount # testfnccount: function name -> execution count $testfncdata = $data->{"testfnc"}; foreach my $tn (keys(%{$testfncdata})) { my $testfnccount = $testfncdata->{$tn}; my %newtestfnccount; foreach my $fn (keys(%{$testfnccount})) { my $cn = $conv->{$fn}; # Add counts for different functions that map # to the same name. $newtestfnccount{$cn} += $testfnccount->{$fn}; } $testfncdata->{$tn} = \%newtestfnccount; } # sumfnccount: function name -> execution count $sumfnccount = $data->{"sumfnc"}; foreach my $fn (keys(%{$sumfnccount})) { my $cn = $conv->{$fn}; # Add counts for different functions that map # to the same name. $newsumfnccount{$cn} += $sumfnccount->{$fn}; } $data->{"sumfnc"} = \%newsumfnccount; # Update function found and hit counts since they may have # changed $f_found = 0; $f_hit = 0; foreach my $fn (keys(%newsumfnccount)) { $f_found++; $f_hit++ if ($newsumfnccount{$fn} > 0); } $data->{"f_found"} = $f_found; $data->{"f_hit"} = $f_hit; } } # # gen_html() # # Generate a set of HTML pages from contents of .info file INFO_FILENAME. # Files will be written to the current directory. If provided, test case # descriptions will be read from .tests file TEST_FILENAME and included # in ouput. # # Die on error. # sub gen_html() { local *HTML_HANDLE; my %overview; my %base_data; my $lines_found; my $lines_hit; my $fn_found; my $fn_hit; my $br_found; my $br_hit; my $overall_found = 0; my $overall_hit = 0; my $total_fn_found = 0; my $total_fn_hit = 0; my $total_br_found = 0; my $total_br_hit = 0; my $dir_name; my $link_name; my @dir_list; my %new_info; # Read in all specified .info files foreach (@info_filenames) { %new_info = %{read_info_file($_)}; # Combine %new_info with %info_data %info_data = %{combine_info_files(\%info_data, \%new_info)}; } info("Found %d entries.\n", scalar(keys(%info_data))); # Read and apply baseline data if specified if ($base_filename) { # Read baseline file info("Reading baseline file $base_filename\n"); %base_data = %{read_info_file($base_filename)}; info("Found %d entries.\n", scalar(keys(%base_data))); # Apply baseline info("Subtracting baseline data.\n"); %info_data = %{apply_baseline(\%info_data, \%base_data)}; } @dir_list = get_dir_list(keys(%info_data)); if ($no_prefix) { # User requested that we leave filenames alone info("User asked not to remove filename prefix\n"); } elsif (! @dir_prefix) { # Get prefix common to most directories in list my $prefix = get_prefix(1, keys(%info_data)); if ($prefix) { info("Found common filename prefix \"$prefix\"\n"); $dir_prefix[0] = $prefix; } else { info("No common filename prefix found!\n"); $no_prefix=1; } } else { my $msg = "Using user-specified filename prefix "; for my $i (0 .. $#dir_prefix) { $dir_prefix[$i] =~ s/\/+$//; $msg .= ", " unless 0 == $i; $msg .= "\"" . $dir_prefix[$i] . "\""; } info($msg . "\n"); } # Read in test description file if specified if ($desc_filename) { info("Reading test description file $desc_filename\n"); %test_description = %{read_testfile($desc_filename)}; # Remove test descriptions which are not referenced # from %info_data if user didn't tell us otherwise if (!$keep_descriptions) { remove_unused_descriptions(); } } # Change to output directory if specified if ($output_directory) { chdir($output_directory) or die("ERROR: cannot change to directory ". "$output_directory!\n"); } info("Writing .css and .png files.\n"); write_css_file(); write_png_files(); if ($html_gzip) { info("Writing .htaccess file.\n"); write_htaccess_file(); } info("Generating output.\n"); # Process each subdirectory and collect overview information foreach $dir_name (@dir_list) { ($lines_found, $lines_hit, $fn_found, $fn_hit, $br_found, $br_hit) = process_dir($dir_name); # Handle files in root directory gracefully $dir_name = "root" if ($dir_name eq ""); # Remove prefix if applicable if (!$no_prefix && @dir_prefix) { # Match directory names beginning with one of @dir_prefix $dir_name = apply_prefix($dir_name,@dir_prefix); } # Generate name for directory overview HTML page if ($dir_name =~ /^\/(.*)$/) { $link_name = substr($dir_name, 1)."/index.$html_ext"; } else { $link_name = $dir_name."/index.$html_ext"; } $overview{$dir_name} = [$lines_found, $lines_hit, $fn_found, $fn_hit, $br_found, $br_hit, $link_name, get_rate($lines_found, $lines_hit), get_rate($fn_found, $fn_hit), get_rate($br_found, $br_hit)]; $overall_found += $lines_found; $overall_hit += $lines_hit; $total_fn_found += $fn_found; $total_fn_hit += $fn_hit; $total_br_found += $br_found; $total_br_hit += $br_hit; } # Generate overview page info("Writing directory view page.\n"); # Create sorted pages foreach (@fileview_sortlist) { write_dir_page($fileview_sortname[$_], ".", "", $test_title, undef, $overall_found, $overall_hit, $total_fn_found, $total_fn_hit, $total_br_found, $total_br_hit, \%overview, {}, {}, {}, 0, $_); } # Check if there are any test case descriptions to write out if (%test_description) { info("Writing test case description file.\n"); write_description_file( \%test_description, $overall_found, $overall_hit, $total_fn_found, $total_fn_hit, $total_br_found, $total_br_hit); } print_overall_rate(1, $overall_found, $overall_hit, $func_coverage, $total_fn_found, $total_fn_hit, $br_coverage, $total_br_found, $total_br_hit); chdir($cwd); } # # html_create(handle, filename) # sub html_create($$) { my $handle = $_[0]; my $filename = $_[1]; if ($html_gzip) { open($handle, "|-", "gzip -c >'$filename'") or die("ERROR: cannot open $filename for writing ". "(gzip)!\n"); } else { open($handle, ">", $filename) or die("ERROR: cannot open $filename for writing!\n"); } } sub write_dir_page($$$$$$$$$$$$$$$$$) { my ($name, $rel_dir, $base_dir, $title, $trunc_dir, $overall_found, $overall_hit, $total_fn_found, $total_fn_hit, $total_br_found, $total_br_hit, $overview, $testhash, $testfnchash, $testbrhash, $view_type, $sort_type) = @_; # Generate directory overview page including details html_create(*HTML_HANDLE, "$rel_dir/index$name.$html_ext"); if (!defined($trunc_dir)) { $trunc_dir = ""; } $title .= " - " if ($trunc_dir ne ""); write_html_prolog(*HTML_HANDLE, $base_dir, "LCOV - $title$trunc_dir"); write_header(*HTML_HANDLE, $view_type, $trunc_dir, $rel_dir, $overall_found, $overall_hit, $total_fn_found, $total_fn_hit, $total_br_found, $total_br_hit, $sort_type); write_file_table(*HTML_HANDLE, $base_dir, $overview, $testhash, $testfnchash, $testbrhash, $view_type, $sort_type); write_html_epilog(*HTML_HANDLE, $base_dir); close(*HTML_HANDLE); } # # process_dir(dir_name) # sub process_dir($) { my $abs_dir = $_[0]; my $trunc_dir; my $rel_dir = $abs_dir; my $base_dir; my $filename; my %overview; my $lines_found; my $lines_hit; my $fn_found; my $fn_hit; my $br_found; my $br_hit; my $overall_found=0; my $overall_hit=0; my $total_fn_found=0; my $total_fn_hit=0; my $total_br_found = 0; my $total_br_hit = 0; my $base_name; my $extension; my $testdata; my %testhash; my $testfncdata; my %testfnchash; my $testbrdata; my %testbrhash; my @sort_list; local *HTML_HANDLE; # Remove prefix if applicable if (!$no_prefix) { # Match directory name beginning with one of @dir_prefix $rel_dir = apply_prefix($rel_dir,@dir_prefix); } $trunc_dir = $rel_dir; # Remove leading / if ($rel_dir =~ /^\/(.*)$/) { $rel_dir = substr($rel_dir, 1); } # Handle files in root directory gracefully $rel_dir = "root" if ($rel_dir eq ""); $trunc_dir = "root" if ($trunc_dir eq ""); $base_dir = get_relative_base_path($rel_dir); create_sub_dir($rel_dir); # Match filenames which specify files in this directory, not including # sub-directories foreach $filename (grep(/^\Q$abs_dir\E\/[^\/]*$/,keys(%info_data))) { my $page_link; my $func_link; ($lines_found, $lines_hit, $fn_found, $fn_hit, $br_found, $br_hit, $testdata, $testfncdata, $testbrdata) = process_file($trunc_dir, $rel_dir, $filename); $base_name = basename($filename); if ($no_sourceview) { $page_link = ""; } elsif ($frames) { # Link to frameset page $page_link = "$base_name.gcov.frameset.$html_ext"; } else { # Link directory to source code view page $page_link = "$base_name.gcov.$html_ext"; } $overview{$base_name} = [$lines_found, $lines_hit, $fn_found, $fn_hit, $br_found, $br_hit, $page_link, get_rate($lines_found, $lines_hit), get_rate($fn_found, $fn_hit), get_rate($br_found, $br_hit)]; $testhash{$base_name} = $testdata; $testfnchash{$base_name} = $testfncdata; $testbrhash{$base_name} = $testbrdata; $overall_found += $lines_found; $overall_hit += $lines_hit; $total_fn_found += $fn_found; $total_fn_hit += $fn_hit; $total_br_found += $br_found; $total_br_hit += $br_hit; } # Create sorted pages foreach (@fileview_sortlist) { # Generate directory overview page (without details) write_dir_page($fileview_sortname[$_], $rel_dir, $base_dir, $test_title, $trunc_dir, $overall_found, $overall_hit, $total_fn_found, $total_fn_hit, $total_br_found, $total_br_hit, \%overview, {}, {}, {}, 1, $_); if (!$show_details) { next; } # Generate directory overview page including details write_dir_page("-detail".$fileview_sortname[$_], $rel_dir, $base_dir, $test_title, $trunc_dir, $overall_found, $overall_hit, $total_fn_found, $total_fn_hit, $total_br_found, $total_br_hit, \%overview, \%testhash, \%testfnchash, \%testbrhash, 1, $_); } # Calculate resulting line counts return ($overall_found, $overall_hit, $total_fn_found, $total_fn_hit, $total_br_found, $total_br_hit); } # # get_converted_lines(testdata) # # Return hash of line numbers of those lines which were only covered in # converted data sets. # sub get_converted_lines($) { my $testdata = $_[0]; my $testcount; my %converted; my %nonconverted; my $hash; my $testcase; my $line; my %result; # Get a hash containing line numbers with positive counts both for # converted and original data sets foreach $testcase (keys(%{$testdata})) { # Check to see if this is a converted data set if ($testcase =~ /,diff$/) { $hash = \%converted; } else { $hash = \%nonconverted; } $testcount = $testdata->{$testcase}; # Add lines with a positive count to hash foreach $line (keys%{$testcount}) { if ($testcount->{$line} > 0) { $hash->{$line} = 1; } } } # Combine both hashes to resulting list foreach $line (keys(%converted)) { if (!defined($nonconverted{$line})) { $result{$line} = 1; } } return \%result; } sub write_function_page($$$$$$$$$$$$$$$$$$) { my ($base_dir, $rel_dir, $trunc_dir, $base_name, $title, $lines_found, $lines_hit, $fn_found, $fn_hit, $br_found, $br_hit, $sumcount, $funcdata, $sumfnccount, $testfncdata, $sumbrcount, $testbrdata, $sort_type) = @_; my $pagetitle; my $filename; # Generate function table for this file if ($sort_type == 0) { $filename = "$rel_dir/$base_name.func.$html_ext"; } else { $filename = "$rel_dir/$base_name.func-sort-c.$html_ext"; } html_create(*HTML_HANDLE, $filename); $pagetitle = "LCOV - $title - $trunc_dir/$base_name - functions"; write_html_prolog(*HTML_HANDLE, $base_dir, $pagetitle); write_header(*HTML_HANDLE, 4, "$trunc_dir/$base_name", "$rel_dir/$base_name", $lines_found, $lines_hit, $fn_found, $fn_hit, $br_found, $br_hit, $sort_type); write_function_table(*HTML_HANDLE, "$base_name.gcov.$html_ext", $sumcount, $funcdata, $sumfnccount, $testfncdata, $sumbrcount, $testbrdata, $base_name, $base_dir, $sort_type); write_html_epilog(*HTML_HANDLE, $base_dir, 1); close(*HTML_HANDLE); } # # process_file(trunc_dir, rel_dir, filename) # sub process_file($$$) { info("Processing file ".apply_prefix($_[2], @dir_prefix)."\n"); my $trunc_dir = $_[0]; my $rel_dir = $_[1]; my $filename = $_[2]; my $base_name = basename($filename); my $base_dir = get_relative_base_path($rel_dir); my $testdata; my $testcount; my $sumcount; my $funcdata; my $checkdata; my $testfncdata; my $sumfnccount; my $testbrdata; my $sumbrcount; my $lines_found; my $lines_hit; my $fn_found; my $fn_hit; my $br_found; my $br_hit; my $converted; my @source; my $pagetitle; local *HTML_HANDLE; ($testdata, $sumcount, $funcdata, $checkdata, $testfncdata, $sumfnccount, $testbrdata, $sumbrcount, $lines_found, $lines_hit, $fn_found, $fn_hit, $br_found, $br_hit) = get_info_entry($info_data{$filename}); # Return after this point in case user asked us not to generate # source code view if ($no_sourceview) { return ($lines_found, $lines_hit, $fn_found, $fn_hit, $br_found, $br_hit, $testdata, $testfncdata, $testbrdata); } $converted = get_converted_lines($testdata); # Generate source code view for this file html_create(*HTML_HANDLE, "$rel_dir/$base_name.gcov.$html_ext"); $pagetitle = "LCOV - $test_title - $trunc_dir/$base_name"; write_html_prolog(*HTML_HANDLE, $base_dir, $pagetitle); write_header(*HTML_HANDLE, 2, "$trunc_dir/$base_name", "$rel_dir/$base_name", $lines_found, $lines_hit, $fn_found, $fn_hit, $br_found, $br_hit, 0); @source = write_source(*HTML_HANDLE, $filename, $sumcount, $checkdata, $converted, $funcdata, $sumbrcount); write_html_epilog(*HTML_HANDLE, $base_dir, 1); close(*HTML_HANDLE); if ($func_coverage) { # Create function tables foreach (@funcview_sortlist) { write_function_page($base_dir, $rel_dir, $trunc_dir, $base_name, $test_title, $lines_found, $lines_hit, $fn_found, $fn_hit, $br_found, $br_hit, $sumcount, $funcdata, $sumfnccount, $testfncdata, $sumbrcount, $testbrdata, $_); } } # Additional files are needed in case of frame output if (!$frames) { return ($lines_found, $lines_hit, $fn_found, $fn_hit, $br_found, $br_hit, $testdata, $testfncdata, $testbrdata); } # Create overview png file gen_png("$rel_dir/$base_name.gcov.png", $overview_width, $tab_size, @source); # Create frameset page html_create(*HTML_HANDLE, "$rel_dir/$base_name.gcov.frameset.$html_ext"); write_frameset(*HTML_HANDLE, $base_dir, $base_name, $pagetitle); close(*HTML_HANDLE); # Write overview frame html_create(*HTML_HANDLE, "$rel_dir/$base_name.gcov.overview.$html_ext"); write_overview(*HTML_HANDLE, $base_dir, $base_name, $pagetitle, scalar(@source)); close(*HTML_HANDLE); return ($lines_found, $lines_hit, $fn_found, $fn_hit, $br_found, $br_hit, $testdata, $testfncdata, $testbrdata); } sub compress_brcount($) { my ($brcount) = @_; my $db; $db = brcount_to_db($brcount); return db_to_brcount($db, $brcount); } # # read_info_file(info_filename) # # Read in the contents of the .info file specified by INFO_FILENAME. Data will # be returned as a reference to a hash containing the following mappings: # # %result: for each filename found in file -> \%data # # %data: "test" -> \%testdata # "sum" -> \%sumcount # "func" -> \%funcdata # "found" -> $lines_found (number of instrumented lines found in file) # "hit" -> $lines_hit (number of executed lines in file) # "f_found" -> $fn_found (number of instrumented functions found in file) # "f_hit" -> $fn_hit (number of executed functions in file) # "b_found" -> $br_found (number of instrumented branches found in file) # "b_hit" -> $br_hit (number of executed branches in file) # "check" -> \%checkdata # "testfnc" -> \%testfncdata # "sumfnc" -> \%sumfnccount # "testbr" -> \%testbrdata # "sumbr" -> \%sumbrcount # # %testdata : name of test affecting this file -> \%testcount # %testfncdata: name of test affecting this file -> \%testfnccount # %testbrdata: name of test affecting this file -> \%testbrcount # # %testcount : line number -> execution count for a single test # %testfnccount: function name -> execution count for a single test # %testbrcount : line number -> branch coverage data for a single test # %sumcount : line number -> execution count for all tests # %sumfnccount : function name -> execution count for all tests # %sumbrcount : line number -> branch coverage data for all tests # %funcdata : function name -> line number # %checkdata : line number -> checksum of source code line # $brdata : vector of items: block, branch, taken # # Note that .info file sections referring to the same file and test name # will automatically be combined by adding all execution counts. # # Note that if INFO_FILENAME ends with ".gz", it is assumed that the file # is compressed using GZIP. If available, GUNZIP will be used to decompress # this file. # # Die on error. # sub read_info_file($) { my $tracefile = $_[0]; # Name of tracefile my %result; # Resulting hash: file -> data my $data; # Data handle for current entry my $testdata; # " " my $testcount; # " " my $sumcount; # " " my $funcdata; # " " my $checkdata; # " " my $testfncdata; my $testfnccount; my $sumfnccount; my $testbrdata; my $testbrcount; my $sumbrcount; my $line; # Current line read from .info file my $testname; # Current test name my $filename; # Current filename my $hitcount; # Count for lines hit my $count; # Execution count of current line my $negative; # If set, warn about negative counts my $changed_testname; # If set, warn about changed testname my $line_checksum; # Checksum of current line my $notified_about_relative_paths; local *INFO_HANDLE; # Filehandle for .info file info("Reading data file $tracefile\n"); # Check if file exists and is readable stat($_[0]); if (!(-r _)) { die("ERROR: cannot read file $_[0]!\n"); } # Check if this is really a plain file if (!(-f _)) { die("ERROR: not a plain file: $_[0]!\n"); } # Check for .gz extension if ($_[0] =~ /\.gz$/) { # Check for availability of GZIP tool system_no_output(1, "gunzip" ,"-h") and die("ERROR: gunzip command not available!\n"); # Check integrity of compressed file system_no_output(1, "gunzip", "-t", $_[0]) and die("ERROR: integrity check failed for ". "compressed file $_[0]!\n"); # Open compressed file open(INFO_HANDLE, "-|", "gunzip -c '$_[0]'") or die("ERROR: cannot start gunzip to decompress ". "file $_[0]!\n"); } else { # Open decompressed file open(INFO_HANDLE, "<", $_[0]) or die("ERROR: cannot read file $_[0]!\n"); } $testname = ""; while () { chomp($_); $line = $_; # Switch statement foreach ($line) { /^TN:([^,]*)(,diff)?/ && do { # Test name information found $testname = defined($1) ? $1 : ""; if ($testname =~ s/\W/_/g) { $changed_testname = 1; } $testname .= $2 if (defined($2)); last; }; /^[SK]F:(.*)/ && do { # Filename information found # Retrieve data for new entry $filename = File::Spec->rel2abs($1, $cwd); if (!File::Spec->file_name_is_absolute($1) && !$notified_about_relative_paths) { info("Resolved relative source file ". "path \"$1\" with CWD to ". "\"$filename\".\n"); $notified_about_relative_paths = 1; } $data = $result{$filename}; ($testdata, $sumcount, $funcdata, $checkdata, $testfncdata, $sumfnccount, $testbrdata, $sumbrcount) = get_info_entry($data); if (defined($testname)) { $testcount = $testdata->{$testname}; $testfnccount = $testfncdata->{$testname}; $testbrcount = $testbrdata->{$testname}; } else { $testcount = {}; $testfnccount = {}; $testbrcount = {}; } last; }; /^DA:(\d+),(-?\d+)(,[^,\s]+)?/ && do { # Fix negative counts $count = $2 < 0 ? 0 : $2; if ($2 < 0) { $negative = 1; } # Execution count found, add to structure # Add summary counts $sumcount->{$1} += $count; # Add test-specific counts if (defined($testname)) { $testcount->{$1} += $count; } # Store line checksum if available if (defined($3)) { $line_checksum = substr($3, 1); # Does it match a previous definition if (defined($checkdata->{$1}) && ($checkdata->{$1} ne $line_checksum)) { die("ERROR: checksum mismatch ". "at $filename:$1\n"); } $checkdata->{$1} = $line_checksum; } last; }; /^FN:(\d+),([^,]+)/ && do { last if (!$func_coverage); # Function data found, add to structure $funcdata->{$2} = $1; # Also initialize function call data if (!defined($sumfnccount->{$2})) { $sumfnccount->{$2} = 0; } if (defined($testname)) { if (!defined($testfnccount->{$2})) { $testfnccount->{$2} = 0; } } last; }; /^FNDA:(\d+),([^,]+)/ && do { last if (!$func_coverage); # Function call count found, add to structure # Add summary counts $sumfnccount->{$2} += $1; # Add test-specific counts if (defined($testname)) { $testfnccount->{$2} += $1; } last; }; /^BRDA:(\d+),(\d+),(\d+),(\d+|-)/ && do { # Branch coverage data found my ($line, $block, $branch, $taken) = ($1, $2, $3, $4); last if (!$br_coverage); $sumbrcount->{$line} .= "$block,$branch,$taken:"; # Add test-specific counts if (defined($testname)) { $testbrcount->{$line} .= "$block,$branch,$taken:"; } last; }; /^end_of_record/ && do { # Found end of section marker if ($filename) { # Store current section data if (defined($testname)) { $testdata->{$testname} = $testcount; $testfncdata->{$testname} = $testfnccount; $testbrdata->{$testname} = $testbrcount; } set_info_entry($data, $testdata, $sumcount, $funcdata, $checkdata, $testfncdata, $sumfnccount, $testbrdata, $sumbrcount); $result{$filename} = $data; last; } }; # default last; } } close(INFO_HANDLE); # Calculate lines_found and lines_hit for each file foreach $filename (keys(%result)) { $data = $result{$filename}; ($testdata, $sumcount, undef, undef, $testfncdata, $sumfnccount, $testbrdata, $sumbrcount) = get_info_entry($data); # Filter out empty files if (scalar(keys(%{$sumcount})) == 0) { delete($result{$filename}); next; } # Filter out empty test cases foreach $testname (keys(%{$testdata})) { if (!defined($testdata->{$testname}) || scalar(keys(%{$testdata->{$testname}})) == 0) { delete($testdata->{$testname}); delete($testfncdata->{$testname}); } } $data->{"found"} = scalar(keys(%{$sumcount})); $hitcount = 0; foreach (keys(%{$sumcount})) { if ($sumcount->{$_} > 0) { $hitcount++; } } $data->{"hit"} = $hitcount; # Get found/hit values for function call data $data->{"f_found"} = scalar(keys(%{$sumfnccount})); $hitcount = 0; foreach (keys(%{$sumfnccount})) { if ($sumfnccount->{$_} > 0) { $hitcount++; } } $data->{"f_hit"} = $hitcount; # Combine branch data for the same branches (undef, $data->{"b_found"}, $data->{"b_hit"}) = compress_brcount($sumbrcount); foreach $testname (keys(%{$testbrdata})) { compress_brcount($testbrdata->{$testname}); } } if (scalar(keys(%result)) == 0) { die("ERROR: no valid records found in tracefile $tracefile\n"); } if ($negative) { warn("WARNING: negative counts found in tracefile ". "$tracefile\n"); } if ($changed_testname) { warn("WARNING: invalid characters removed from testname in ". "tracefile $tracefile\n"); } return(\%result); } # # get_info_entry(hash_ref) # # Retrieve data from an entry of the structure generated by read_info_file(). # Return a list of references to hashes: # (test data hash ref, sum count hash ref, funcdata hash ref, checkdata hash # ref, testfncdata hash ref, sumfnccount hash ref, lines found, lines hit, # functions found, functions hit) # sub get_info_entry($) { my $testdata_ref = $_[0]->{"test"}; my $sumcount_ref = $_[0]->{"sum"}; my $funcdata_ref = $_[0]->{"func"}; my $checkdata_ref = $_[0]->{"check"}; my $testfncdata = $_[0]->{"testfnc"}; my $sumfnccount = $_[0]->{"sumfnc"}; my $testbrdata = $_[0]->{"testbr"}; my $sumbrcount = $_[0]->{"sumbr"}; my $lines_found = $_[0]->{"found"}; my $lines_hit = $_[0]->{"hit"}; my $fn_found = $_[0]->{"f_found"}; my $fn_hit = $_[0]->{"f_hit"}; my $br_found = $_[0]->{"b_found"}; my $br_hit = $_[0]->{"b_hit"}; return ($testdata_ref, $sumcount_ref, $funcdata_ref, $checkdata_ref, $testfncdata, $sumfnccount, $testbrdata, $sumbrcount, $lines_found, $lines_hit, $fn_found, $fn_hit, $br_found, $br_hit); } # # set_info_entry(hash_ref, testdata_ref, sumcount_ref, funcdata_ref, # checkdata_ref, testfncdata_ref, sumfcncount_ref, # testbrdata_ref, sumbrcount_ref[,lines_found, # lines_hit, f_found, f_hit, $b_found, $b_hit]) # # Update the hash referenced by HASH_REF with the provided data references. # sub set_info_entry($$$$$$$$$;$$$$$$) { my $data_ref = $_[0]; $data_ref->{"test"} = $_[1]; $data_ref->{"sum"} = $_[2]; $data_ref->{"func"} = $_[3]; $data_ref->{"check"} = $_[4]; $data_ref->{"testfnc"} = $_[5]; $data_ref->{"sumfnc"} = $_[6]; $data_ref->{"testbr"} = $_[7]; $data_ref->{"sumbr"} = $_[8]; if (defined($_[9])) { $data_ref->{"found"} = $_[9]; } if (defined($_[10])) { $data_ref->{"hit"} = $_[10]; } if (defined($_[11])) { $data_ref->{"f_found"} = $_[11]; } if (defined($_[12])) { $data_ref->{"f_hit"} = $_[12]; } if (defined($_[13])) { $data_ref->{"b_found"} = $_[13]; } if (defined($_[14])) { $data_ref->{"b_hit"} = $_[14]; } } # # add_counts(data1_ref, data2_ref) # # DATA1_REF and DATA2_REF are references to hashes containing a mapping # # line number -> execution count # # Return a list (RESULT_REF, LINES_FOUND, LINES_HIT) where RESULT_REF # is a reference to a hash containing the combined mapping in which # execution counts are added. # sub add_counts($$) { my $data1_ref = $_[0]; # Hash 1 my $data2_ref = $_[1]; # Hash 2 my %result; # Resulting hash my $line; # Current line iteration scalar my $data1_count; # Count of line in hash1 my $data2_count; # Count of line in hash2 my $found = 0; # Total number of lines found my $hit = 0; # Number of lines with a count > 0 foreach $line (keys(%$data1_ref)) { $data1_count = $data1_ref->{$line}; $data2_count = $data2_ref->{$line}; # Add counts if present in both hashes if (defined($data2_count)) { $data1_count += $data2_count; } # Store sum in %result $result{$line} = $data1_count; $found++; if ($data1_count > 0) { $hit++; } } # Add lines unique to data2_ref foreach $line (keys(%$data2_ref)) { # Skip lines already in data1_ref if (defined($data1_ref->{$line})) { next; } # Copy count from data2_ref $result{$line} = $data2_ref->{$line}; $found++; if ($result{$line} > 0) { $hit++; } } return (\%result, $found, $hit); } # # merge_checksums(ref1, ref2, filename) # # REF1 and REF2 are references to hashes containing a mapping # # line number -> checksum # # Merge checksum lists defined in REF1 and REF2 and return reference to # resulting hash. Die if a checksum for a line is defined in both hashes # but does not match. # sub merge_checksums($$$) { my $ref1 = $_[0]; my $ref2 = $_[1]; my $filename = $_[2]; my %result; my $line; foreach $line (keys(%{$ref1})) { if (defined($ref2->{$line}) && ($ref1->{$line} ne $ref2->{$line})) { die("ERROR: checksum mismatch at $filename:$line\n"); } $result{$line} = $ref1->{$line}; } foreach $line (keys(%{$ref2})) { $result{$line} = $ref2->{$line}; } return \%result; } # # merge_func_data(funcdata1, funcdata2, filename) # sub merge_func_data($$$) { my ($funcdata1, $funcdata2, $filename) = @_; my %result; my $func; if (defined($funcdata1)) { %result = %{$funcdata1}; } foreach $func (keys(%{$funcdata2})) { my $line1 = $result{$func}; my $line2 = $funcdata2->{$func}; if (defined($line1) && ($line1 != $line2)) { warn("WARNING: function data mismatch at ". "$filename:$line2\n"); next; } $result{$func} = $line2; } return \%result; } # # add_fnccount(fnccount1, fnccount2) # # Add function call count data. Return list (fnccount_added, f_found, f_hit) # sub add_fnccount($$) { my ($fnccount1, $fnccount2) = @_; my %result; my $fn_found; my $fn_hit; my $function; if (defined($fnccount1)) { %result = %{$fnccount1}; } foreach $function (keys(%{$fnccount2})) { $result{$function} += $fnccount2->{$function}; } $fn_found = scalar(keys(%result)); $fn_hit = 0; foreach $function (keys(%result)) { if ($result{$function} > 0) { $fn_hit++; } } return (\%result, $fn_found, $fn_hit); } # # add_testfncdata(testfncdata1, testfncdata2) # # Add function call count data for several tests. Return reference to # added_testfncdata. # sub add_testfncdata($$) { my ($testfncdata1, $testfncdata2) = @_; my %result; my $testname; foreach $testname (keys(%{$testfncdata1})) { if (defined($testfncdata2->{$testname})) { my $fnccount; # Function call count data for this testname exists # in both data sets: add ($fnccount) = add_fnccount( $testfncdata1->{$testname}, $testfncdata2->{$testname}); $result{$testname} = $fnccount; next; } # Function call count data for this testname is unique to # data set 1: copy $result{$testname} = $testfncdata1->{$testname}; } # Add count data for testnames unique to data set 2 foreach $testname (keys(%{$testfncdata2})) { if (!defined($result{$testname})) { $result{$testname} = $testfncdata2->{$testname}; } } return \%result; } # # brcount_to_db(brcount) # # Convert brcount data to the following format: # # db: line number -> block hash # block hash: block number -> branch hash # branch hash: branch number -> taken value # sub brcount_to_db($) { my ($brcount) = @_; my $line; my $db; # Add branches to database foreach $line (keys(%{$brcount})) { my $brdata = $brcount->{$line}; foreach my $entry (split(/:/, $brdata)) { my ($block, $branch, $taken) = split(/,/, $entry); my $old = $db->{$line}->{$block}->{$branch}; if (!defined($old) || $old eq "-") { $old = $taken; } elsif ($taken ne "-") { $old += $taken; } $db->{$line}->{$block}->{$branch} = $old; } } return $db; } # # db_to_brcount(db[, brcount]) # # Convert branch coverage data back to brcount format. If brcount is specified, # the converted data is directly inserted in brcount. # sub db_to_brcount($;$) { my ($db, $brcount) = @_; my $line; my $br_found = 0; my $br_hit = 0; # Convert database back to brcount format foreach $line (sort({$a <=> $b} keys(%{$db}))) { my $ldata = $db->{$line}; my $brdata; my $block; foreach $block (sort({$a <=> $b} keys(%{$ldata}))) { my $bdata = $ldata->{$block}; my $branch; foreach $branch (sort({$a <=> $b} keys(%{$bdata}))) { my $taken = $bdata->{$branch}; $br_found++; $br_hit++ if ($taken ne "-" && $taken > 0); $brdata .= "$block,$branch,$taken:"; } } $brcount->{$line} = $brdata; } return ($brcount, $br_found, $br_hit); } # # brcount_db_combine(db1, db2, op) # # db1 := db1 op db2, where # db1, db2: brcount data as returned by brcount_to_db # op: one of $BR_ADD and BR_SUB # sub brcount_db_combine($$$) { my ($db1, $db2, $op) = @_; foreach my $line (keys(%{$db2})) { my $ldata = $db2->{$line}; foreach my $block (keys(%{$ldata})) { my $bdata = $ldata->{$block}; foreach my $branch (keys(%{$bdata})) { my $taken = $bdata->{$branch}; my $new = $db1->{$line}->{$block}->{$branch}; if (!defined($new) || $new eq "-") { $new = $taken; } elsif ($taken ne "-") { if ($op == $BR_ADD) { $new += $taken; } elsif ($op == $BR_SUB) { $new -= $taken; $new = 0 if ($new < 0); } } $db1->{$line}->{$block}->{$branch} = $new; } } } } # # brcount_db_get_found_and_hit(db) # # Return (br_found, br_hit) for db. # sub brcount_db_get_found_and_hit($) { my ($db) = @_; my ($br_found , $br_hit) = (0, 0); foreach my $line (keys(%{$db})) { my $ldata = $db->{$line}; foreach my $block (keys(%{$ldata})) { my $bdata = $ldata->{$block}; foreach my $branch (keys(%{$bdata})) { my $taken = $bdata->{$branch}; $br_found++; $br_hit++ if ($taken ne "-" && $taken > 0); } } } return ($br_found, $br_hit); } # combine_brcount(brcount1, brcount2, type, inplace) # # If add is BR_ADD, add branch coverage data and return list brcount_added. # If add is BR_SUB, subtract the taken values of brcount2 from brcount1 and # return brcount_sub. If inplace is set, the result is inserted into brcount1. # sub combine_brcount($$$;$) { my ($brcount1, $brcount2, $type, $inplace) = @_; my ($db1, $db2); $db1 = brcount_to_db($brcount1); $db2 = brcount_to_db($brcount2); brcount_db_combine($db1, $db2, $type); return db_to_brcount($db1, $inplace ? $brcount1 : undef); } # # add_testbrdata(testbrdata1, testbrdata2) # # Add branch coverage data for several tests. Return reference to # added_testbrdata. # sub add_testbrdata($$) { my ($testbrdata1, $testbrdata2) = @_; my %result; my $testname; foreach $testname (keys(%{$testbrdata1})) { if (defined($testbrdata2->{$testname})) { my $brcount; # Branch coverage data for this testname exists # in both data sets: add ($brcount) = combine_brcount($testbrdata1->{$testname}, $testbrdata2->{$testname}, $BR_ADD); $result{$testname} = $brcount; next; } # Branch coverage data for this testname is unique to # data set 1: copy $result{$testname} = $testbrdata1->{$testname}; } # Add count data for testnames unique to data set 2 foreach $testname (keys(%{$testbrdata2})) { if (!defined($result{$testname})) { $result{$testname} = $testbrdata2->{$testname}; } } return \%result; } # # combine_info_entries(entry_ref1, entry_ref2, filename) # # Combine .info data entry hashes referenced by ENTRY_REF1 and ENTRY_REF2. # Return reference to resulting hash. # sub combine_info_entries($$$) { my $entry1 = $_[0]; # Reference to hash containing first entry my $testdata1; my $sumcount1; my $funcdata1; my $checkdata1; my $testfncdata1; my $sumfnccount1; my $testbrdata1; my $sumbrcount1; my $entry2 = $_[1]; # Reference to hash containing second entry my $testdata2; my $sumcount2; my $funcdata2; my $checkdata2; my $testfncdata2; my $sumfnccount2; my $testbrdata2; my $sumbrcount2; my %result; # Hash containing combined entry my %result_testdata; my $result_sumcount = {}; my $result_funcdata; my $result_testfncdata; my $result_sumfnccount; my $result_testbrdata; my $result_sumbrcount; my $lines_found; my $lines_hit; my $fn_found; my $fn_hit; my $br_found; my $br_hit; my $testname; my $filename = $_[2]; # Retrieve data ($testdata1, $sumcount1, $funcdata1, $checkdata1, $testfncdata1, $sumfnccount1, $testbrdata1, $sumbrcount1) = get_info_entry($entry1); ($testdata2, $sumcount2, $funcdata2, $checkdata2, $testfncdata2, $sumfnccount2, $testbrdata2, $sumbrcount2) = get_info_entry($entry2); # Merge checksums $checkdata1 = merge_checksums($checkdata1, $checkdata2, $filename); # Combine funcdata $result_funcdata = merge_func_data($funcdata1, $funcdata2, $filename); # Combine function call count data $result_testfncdata = add_testfncdata($testfncdata1, $testfncdata2); ($result_sumfnccount, $fn_found, $fn_hit) = add_fnccount($sumfnccount1, $sumfnccount2); # Combine branch coverage data $result_testbrdata = add_testbrdata($testbrdata1, $testbrdata2); ($result_sumbrcount, $br_found, $br_hit) = combine_brcount($sumbrcount1, $sumbrcount2, $BR_ADD); # Combine testdata foreach $testname (keys(%{$testdata1})) { if (defined($testdata2->{$testname})) { # testname is present in both entries, requires # combination ($result_testdata{$testname}) = add_counts($testdata1->{$testname}, $testdata2->{$testname}); } else { # testname only present in entry1, add to result $result_testdata{$testname} = $testdata1->{$testname}; } # update sum count hash ($result_sumcount, $lines_found, $lines_hit) = add_counts($result_sumcount, $result_testdata{$testname}); } foreach $testname (keys(%{$testdata2})) { # Skip testnames already covered by previous iteration if (defined($testdata1->{$testname})) { next; } # testname only present in entry2, add to result hash $result_testdata{$testname} = $testdata2->{$testname}; # update sum count hash ($result_sumcount, $lines_found, $lines_hit) = add_counts($result_sumcount, $result_testdata{$testname}); } # Calculate resulting sumcount # Store result set_info_entry(\%result, \%result_testdata, $result_sumcount, $result_funcdata, $checkdata1, $result_testfncdata, $result_sumfnccount, $result_testbrdata, $result_sumbrcount, $lines_found, $lines_hit, $fn_found, $fn_hit, $br_found, $br_hit); return(\%result); } # # combine_info_files(info_ref1, info_ref2) # # Combine .info data in hashes referenced by INFO_REF1 and INFO_REF2. Return # reference to resulting hash. # sub combine_info_files($$) { my %hash1 = %{$_[0]}; my %hash2 = %{$_[1]}; my $filename; foreach $filename (keys(%hash2)) { if ($hash1{$filename}) { # Entry already exists in hash1, combine them $hash1{$filename} = combine_info_entries($hash1{$filename}, $hash2{$filename}, $filename); } else { # Entry is unique in both hashes, simply add to # resulting hash $hash1{$filename} = $hash2{$filename}; } } return(\%hash1); } # # get_prefix(min_dir, filename_list) # # Search FILENAME_LIST for a directory prefix which is common to as many # list entries as possible, so that removing this prefix will minimize the # sum of the lengths of all resulting shortened filenames while observing # that no filename has less than MIN_DIR parent directories. # sub get_prefix($@) { my ($min_dir, @filename_list) = @_; my %prefix; # mapping: prefix -> sum of lengths my $current; # Temporary iteration variable # Find list of prefixes foreach (@filename_list) { # Need explicit assignment to get a copy of $_ so that # shortening the contained prefix does not affect the list $current = $_; while ($current = shorten_prefix($current)) { $current .= "/"; # Skip rest if the remaining prefix has already been # added to hash if (exists($prefix{$current})) { last; } # Initialize with 0 $prefix{$current}="0"; } } # Remove all prefixes that would cause filenames to have less than # the minimum number of parent directories foreach my $filename (@filename_list) { my $dir = dirname($filename); for (my $i = 0; $i < $min_dir; $i++) { delete($prefix{$dir."/"}); $dir = shorten_prefix($dir); } } # Check if any prefix remains return undef if (!%prefix); # Calculate sum of lengths for all prefixes foreach $current (keys(%prefix)) { foreach (@filename_list) { # Add original length $prefix{$current} += length($_); # Check whether prefix matches if (substr($_, 0, length($current)) eq $current) { # Subtract prefix length for this filename $prefix{$current} -= length($current); } } } # Find and return prefix with minimal sum $current = (keys(%prefix))[0]; foreach (keys(%prefix)) { if ($prefix{$_} < $prefix{$current}) { $current = $_; } } $current =~ s/\/$//; return($current); } # # shorten_prefix(prefix) # # Return PREFIX shortened by last directory component. # sub shorten_prefix($) { my @list = split("/", $_[0]); pop(@list); return join("/", @list); } # # get_dir_list(filename_list) # # Return sorted list of directories for each entry in given FILENAME_LIST. # sub get_dir_list(@) { my %result; foreach (@_) { $result{shorten_prefix($_)} = ""; } return(sort(keys(%result))); } # # get_relative_base_path(subdirectory) # # Return a relative path string which references the base path when applied # in SUBDIRECTORY. # # Example: get_relative_base_path("fs/mm") -> "../../" # sub get_relative_base_path($) { my $result = ""; my $index; # Make an empty directory path a special case if (!$_[0]) { return(""); } # Count number of /s in path $index = ($_[0] =~ s/\//\//g); # Add a ../ to $result for each / in the directory path + 1 for (; $index>=0; $index--) { $result .= "../"; } return $result; } # # read_testfile(test_filename) # # Read in file TEST_FILENAME which contains test descriptions in the format: # # TN: # TD: # # for each test case. Return a reference to a hash containing a mapping # # test name -> test description. # # Die on error. # sub read_testfile($) { my %result; my $test_name; my $changed_testname; local *TEST_HANDLE; open(TEST_HANDLE, "<", $_[0]) or die("ERROR: cannot open $_[0]!\n"); while () { chomp($_); # Match lines beginning with TN: if (/^TN:\s+(.*?)\s*$/) { # Store name for later use $test_name = $1; if ($test_name =~ s/\W/_/g) { $changed_testname = 1; } } # Match lines beginning with TD: if (/^TD:\s+(.*?)\s*$/) { if (!defined($test_name)) { die("ERROR: Found test description without prior test name in $_[0]:$.\n"); } # Check for empty line if ($1) { # Add description to hash $result{$test_name} .= " $1"; } else { # Add empty line $result{$test_name} .= "\n\n"; } } } close(TEST_HANDLE); if ($changed_testname) { warn("WARNING: invalid characters removed from testname in ". "descriptions file $_[0]\n"); } return \%result; } # # escape_html(STRING) # # Return a copy of STRING in which all occurrences of HTML special characters # are escaped. # sub escape_html($) { my $string = $_[0]; if (!$string) { return ""; } $string =~ s/&/&/g; # & -> & $string =~ s/ < $string =~ s/>/>/g; # > -> > $string =~ s/\"/"/g; # " -> " while ($string =~ /^([^\t]*)(\t)/) { my $replacement = " "x($tab_size - (length($1) % $tab_size)); $string =~ s/^([^\t]*)(\t)/$1$replacement/; } $string =~ s/\n/
/g; # \n ->
return $string; } # # get_date_string() # # Return the current date in the form: yyyy-mm-dd # sub get_date_string() { my $year; my $month; my $day; my $hour; my $min; my $sec; my @timeresult; if (defined $ENV{'SOURCE_DATE_EPOCH'}) { @timeresult = gmtime($ENV{'SOURCE_DATE_EPOCH'}); } else { @timeresult = localtime(); } ($year, $month, $day, $hour, $min, $sec) = @timeresult[5, 4, 3, 2, 1, 0]; return sprintf("%d-%02d-%02d %02d:%02d:%02d", $year+1900, $month+1, $day, $hour, $min, $sec); } # # create_sub_dir(dir_name) # # Create subdirectory DIR_NAME if it does not already exist, including all its # parent directories. # # Die on error. # sub create_sub_dir($) { my ($dir) = @_; system("mkdir", "-p" ,$dir) and die("ERROR: cannot create directory $dir!\n"); } # # write_description_file(descriptions, overall_found, overall_hit, # total_fn_found, total_fn_hit, total_br_found, # total_br_hit) # # Write HTML file containing all test case descriptions. DESCRIPTIONS is a # reference to a hash containing a mapping # # test case name -> test case description # # Die on error. # sub write_description_file($$$$$$$) { my %description = %{$_[0]}; my $found = $_[1]; my $hit = $_[2]; my $fn_found = $_[3]; my $fn_hit = $_[4]; my $br_found = $_[5]; my $br_hit = $_[6]; my $test_name; local *HTML_HANDLE; html_create(*HTML_HANDLE,"descriptions.$html_ext"); write_html_prolog(*HTML_HANDLE, "", "LCOV - test case descriptions"); write_header(*HTML_HANDLE, 3, "", "", $found, $hit, $fn_found, $fn_hit, $br_found, $br_hit, 0); write_test_table_prolog(*HTML_HANDLE, "Test case descriptions - alphabetical list"); foreach $test_name (sort(keys(%description))) { my $desc = $description{$test_name}; $desc = escape_html($desc) if (!$rc_desc_html); write_test_table_entry(*HTML_HANDLE, $test_name, $desc); } write_test_table_epilog(*HTML_HANDLE); write_html_epilog(*HTML_HANDLE, ""); close(*HTML_HANDLE); } # # write_png_files() # # Create all necessary .png files for the HTML-output in the current # directory. .png-files are used as bar graphs. # # Die on error. # sub write_png_files() { my %data; local *PNG_HANDLE; $data{"ruby.png"} = [0x89, 0x50, 0x4e, 0x47, 0x0d, 0x0a, 0x1a, 0x0a, 0x00, 0x00, 0x00, 0x0d, 0x49, 0x48, 0x44, 0x52, 0x00, 0x00, 0x00, 0x01, 0x00, 0x00, 0x00, 0x01, 0x01, 0x03, 0x00, 0x00, 0x00, 0x25, 0xdb, 0x56, 0xca, 0x00, 0x00, 0x00, 0x07, 0x74, 0x49, 0x4d, 0x45, 0x07, 0xd2, 0x07, 0x11, 0x0f, 0x18, 0x10, 0x5d, 0x57, 0x34, 0x6e, 0x00, 0x00, 0x00, 0x09, 0x70, 0x48, 0x59, 0x73, 0x00, 0x00, 0x0b, 0x12, 0x00, 0x00, 0x0b, 0x12, 0x01, 0xd2, 0xdd, 0x7e, 0xfc, 0x00, 0x00, 0x00, 0x04, 0x67, 0x41, 0x4d, 0x41, 0x00, 0x00, 0xb1, 0x8f, 0x0b, 0xfc, 0x61, 0x05, 0x00, 0x00, 0x00, 0x06, 0x50, 0x4c, 0x54, 0x45, 0xff, 0x35, 0x2f, 0x00, 0x00, 0x00, 0xd0, 0x33, 0x9a, 0x9d, 0x00, 0x00, 0x00, 0x0a, 0x49, 0x44, 0x41, 0x54, 0x78, 0xda, 0x63, 0x60, 0x00, 0x00, 0x00, 0x02, 0x00, 0x01, 0xe5, 0x27, 0xde, 0xfc, 0x00, 0x00, 0x00, 0x00, 0x49, 0x45, 0x4e, 0x44, 0xae, 0x42, 0x60, 0x82]; $data{"amber.png"} = [0x89, 0x50, 0x4e, 0x47, 0x0d, 0x0a, 0x1a, 0x0a, 0x00, 0x00, 0x00, 0x0d, 0x49, 0x48, 0x44, 0x52, 0x00, 0x00, 0x00, 0x01, 0x00, 0x00, 0x00, 0x01, 0x01, 0x03, 0x00, 0x00, 0x00, 0x25, 0xdb, 0x56, 0xca, 0x00, 0x00, 0x00, 0x07, 0x74, 0x49, 0x4d, 0x45, 0x07, 0xd2, 0x07, 0x11, 0x0f, 0x28, 0x04, 0x98, 0xcb, 0xd6, 0xe0, 0x00, 0x00, 0x00, 0x09, 0x70, 0x48, 0x59, 0x73, 0x00, 0x00, 0x0b, 0x12, 0x00, 0x00, 0x0b, 0x12, 0x01, 0xd2, 0xdd, 0x7e, 0xfc, 0x00, 0x00, 0x00, 0x04, 0x67, 0x41, 0x4d, 0x41, 0x00, 0x00, 0xb1, 0x8f, 0x0b, 0xfc, 0x61, 0x05, 0x00, 0x00, 0x00, 0x06, 0x50, 0x4c, 0x54, 0x45, 0xff, 0xe0, 0x50, 0x00, 0x00, 0x00, 0xa2, 0x7a, 0xda, 0x7e, 0x00, 0x00, 0x00, 0x0a, 0x49, 0x44, 0x41, 0x54, 0x78, 0xda, 0x63, 0x60, 0x00, 0x00, 0x00, 0x02, 0x00, 0x01, 0xe5, 0x27, 0xde, 0xfc, 0x00, 0x00, 0x00, 0x00, 0x49, 0x45, 0x4e, 0x44, 0xae, 0x42, 0x60, 0x82]; $data{"emerald.png"} = [0x89, 0x50, 0x4e, 0x47, 0x0d, 0x0a, 0x1a, 0x0a, 0x00, 0x00, 0x00, 0x0d, 0x49, 0x48, 0x44, 0x52, 0x00, 0x00, 0x00, 0x01, 0x00, 0x00, 0x00, 0x01, 0x01, 0x03, 0x00, 0x00, 0x00, 0x25, 0xdb, 0x56, 0xca, 0x00, 0x00, 0x00, 0x07, 0x74, 0x49, 0x4d, 0x45, 0x07, 0xd2, 0x07, 0x11, 0x0f, 0x22, 0x2b, 0xc9, 0xf5, 0x03, 0x33, 0x00, 0x00, 0x00, 0x09, 0x70, 0x48, 0x59, 0x73, 0x00, 0x00, 0x0b, 0x12, 0x00, 0x00, 0x0b, 0x12, 0x01, 0xd2, 0xdd, 0x7e, 0xfc, 0x00, 0x00, 0x00, 0x04, 0x67, 0x41, 0x4d, 0x41, 0x00, 0x00, 0xb1, 0x8f, 0x0b, 0xfc, 0x61, 0x05, 0x00, 0x00, 0x00, 0x06, 0x50, 0x4c, 0x54, 0x45, 0x1b, 0xea, 0x59, 0x0a, 0x0a, 0x0a, 0x0f, 0xba, 0x50, 0x83, 0x00, 0x00, 0x00, 0x0a, 0x49, 0x44, 0x41, 0x54, 0x78, 0xda, 0x63, 0x60, 0x00, 0x00, 0x00, 0x02, 0x00, 0x01, 0xe5, 0x27, 0xde, 0xfc, 0x00, 0x00, 0x00, 0x00, 0x49, 0x45, 0x4e, 0x44, 0xae, 0x42, 0x60, 0x82]; $data{"snow.png"} = [0x89, 0x50, 0x4e, 0x47, 0x0d, 0x0a, 0x1a, 0x0a, 0x00, 0x00, 0x00, 0x0d, 0x49, 0x48, 0x44, 0x52, 0x00, 0x00, 0x00, 0x01, 0x00, 0x00, 0x00, 0x01, 0x01, 0x03, 0x00, 0x00, 0x00, 0x25, 0xdb, 0x56, 0xca, 0x00, 0x00, 0x00, 0x07, 0x74, 0x49, 0x4d, 0x45, 0x07, 0xd2, 0x07, 0x11, 0x0f, 0x1e, 0x1d, 0x75, 0xbc, 0xef, 0x55, 0x00, 0x00, 0x00, 0x09, 0x70, 0x48, 0x59, 0x73, 0x00, 0x00, 0x0b, 0x12, 0x00, 0x00, 0x0b, 0x12, 0x01, 0xd2, 0xdd, 0x7e, 0xfc, 0x00, 0x00, 0x00, 0x04, 0x67, 0x41, 0x4d, 0x41, 0x00, 0x00, 0xb1, 0x8f, 0x0b, 0xfc, 0x61, 0x05, 0x00, 0x00, 0x00, 0x06, 0x50, 0x4c, 0x54, 0x45, 0xff, 0xff, 0xff, 0x00, 0x00, 0x00, 0x55, 0xc2, 0xd3, 0x7e, 0x00, 0x00, 0x00, 0x0a, 0x49, 0x44, 0x41, 0x54, 0x78, 0xda, 0x63, 0x60, 0x00, 0x00, 0x00, 0x02, 0x00, 0x01, 0xe5, 0x27, 0xde, 0xfc, 0x00, 0x00, 0x00, 0x00, 0x49, 0x45, 0x4e, 0x44, 0xae, 0x42, 0x60, 0x82]; $data{"glass.png"} = [0x89, 0x50, 0x4e, 0x47, 0x0d, 0x0a, 0x1a, 0x0a, 0x00, 0x00, 0x00, 0x0d, 0x49, 0x48, 0x44, 0x52, 0x00, 0x00, 0x00, 0x01, 0x00, 0x00, 0x00, 0x01, 0x01, 0x03, 0x00, 0x00, 0x00, 0x25, 0xdb, 0x56, 0xca, 0x00, 0x00, 0x00, 0x04, 0x67, 0x41, 0x4d, 0x41, 0x00, 0x00, 0xb1, 0x8f, 0x0b, 0xfc, 0x61, 0x05, 0x00, 0x00, 0x00, 0x06, 0x50, 0x4c, 0x54, 0x45, 0xff, 0xff, 0xff, 0x00, 0x00, 0x00, 0x55, 0xc2, 0xd3, 0x7e, 0x00, 0x00, 0x00, 0x01, 0x74, 0x52, 0x4e, 0x53, 0x00, 0x40, 0xe6, 0xd8, 0x66, 0x00, 0x00, 0x00, 0x01, 0x62, 0x4b, 0x47, 0x44, 0x00, 0x88, 0x05, 0x1d, 0x48, 0x00, 0x00, 0x00, 0x09, 0x70, 0x48, 0x59, 0x73, 0x00, 0x00, 0x0b, 0x12, 0x00, 0x00, 0x0b, 0x12, 0x01, 0xd2, 0xdd, 0x7e, 0xfc, 0x00, 0x00, 0x00, 0x07, 0x74, 0x49, 0x4d, 0x45, 0x07, 0xd2, 0x07, 0x13, 0x0f, 0x08, 0x19, 0xc4, 0x40, 0x56, 0x10, 0x00, 0x00, 0x00, 0x0a, 0x49, 0x44, 0x41, 0x54, 0x78, 0x9c, 0x63, 0x60, 0x00, 0x00, 0x00, 0x02, 0x00, 0x01, 0x48, 0xaf, 0xa4, 0x71, 0x00, 0x00, 0x00, 0x00, 0x49, 0x45, 0x4e, 0x44, 0xae, 0x42, 0x60, 0x82]; $data{"updown.png"} = [0x89, 0x50, 0x4e, 0x47, 0x0d, 0x0a, 0x1a, 0x0a, 0x00, 0x00, 0x00, 0x0d, 0x49, 0x48, 0x44, 0x52, 0x00, 0x00, 0x00, 0x0a, 0x00, 0x00, 0x00, 0x0e, 0x08, 0x06, 0x00, 0x00, 0x00, 0x16, 0xa3, 0x8d, 0xab, 0x00, 0x00, 0x00, 0x3c, 0x49, 0x44, 0x41, 0x54, 0x28, 0xcf, 0x63, 0x60, 0x40, 0x03, 0xff, 0xa1, 0x00, 0x5d, 0x9c, 0x11, 0x5d, 0x11, 0x8a, 0x24, 0x23, 0x23, 0x23, 0x86, 0x42, 0x6c, 0xa6, 0x20, 0x2b, 0x66, 0xc4, 0xa7, 0x08, 0x59, 0x31, 0x23, 0x21, 0x45, 0x30, 0xc0, 0xc4, 0x30, 0x60, 0x80, 0xfa, 0x6e, 0x24, 0x3e, 0x78, 0x48, 0x0a, 0x70, 0x62, 0xa2, 0x90, 0x81, 0xd8, 0x44, 0x01, 0x00, 0xe9, 0x5c, 0x2f, 0xf5, 0xe2, 0x9d, 0x0f, 0xf9, 0x00, 0x00, 0x00, 0x00, 0x49, 0x45, 0x4e, 0x44, 0xae, 0x42, 0x60, 0x82] if ($sort); foreach (keys(%data)) { open(PNG_HANDLE, ">", $_) or die("ERROR: cannot create $_!\n"); binmode(PNG_HANDLE); print(PNG_HANDLE map(chr,@{$data{$_}})); close(PNG_HANDLE); } } # # write_htaccess_file() # sub write_htaccess_file() { local *HTACCESS_HANDLE; my $htaccess_data; open(*HTACCESS_HANDLE, ">", ".htaccess") or die("ERROR: cannot open .htaccess for writing!\n"); $htaccess_data = (<<"END_OF_HTACCESS") AddEncoding x-gzip .html END_OF_HTACCESS ; print(HTACCESS_HANDLE $htaccess_data); close(*HTACCESS_HANDLE); } # # write_css_file() # # Write the cascading style sheet file gcov.css to the current directory. # This file defines basic layout attributes of all generated HTML pages. # sub write_css_file() { local *CSS_HANDLE; # Check for a specified external style sheet file if ($css_filename) { # Simply copy that file system("cp", $css_filename, "gcov.css") and die("ERROR: cannot copy file $css_filename!\n"); return; } open(CSS_HANDLE, ">", "gcov.css") or die ("ERROR: cannot open gcov.css for writing!\n"); # ************************************************************* my $css_data = ($_=<<"END_OF_CSS") /* All views: initial background and text color */ body { color: #000000; background-color: #FFFFFF; } /* All views: standard link format*/ a:link { color: #284FA8; text-decoration: underline; } /* All views: standard link - visited format */ a:visited { color: #00CB40; text-decoration: underline; } /* All views: standard link - activated format */ a:active { color: #FF0040; text-decoration: underline; } /* All views: main title format */ td.title { text-align: center; padding-bottom: 10px; font-family: sans-serif; font-size: 20pt; font-style: italic; font-weight: bold; } /* All views: header item format */ td.headerItem { text-align: right; padding-right: 6px; font-family: sans-serif; font-weight: bold; vertical-align: top; white-space: nowrap; } /* All views: header item value format */ td.headerValue { text-align: left; color: #284FA8; font-family: sans-serif; font-weight: bold; white-space: nowrap; } /* All views: header item coverage table heading */ td.headerCovTableHead { text-align: center; padding-right: 6px; padding-left: 6px; padding-bottom: 0px; font-family: sans-serif; font-size: 80%; white-space: nowrap; } /* All views: header item coverage table entry */ td.headerCovTableEntry { text-align: right; color: #284FA8; font-family: sans-serif; font-weight: bold; white-space: nowrap; padding-left: 12px; padding-right: 4px; background-color: #DAE7FE; } /* All views: header item coverage table entry for high coverage rate */ td.headerCovTableEntryHi { text-align: right; color: #000000; font-family: sans-serif; font-weight: bold; white-space: nowrap; padding-left: 12px; padding-right: 4px; background-color: #A7FC9D; } /* All views: header item coverage table entry for medium coverage rate */ td.headerCovTableEntryMed { text-align: right; color: #000000; font-family: sans-serif; font-weight: bold; white-space: nowrap; padding-left: 12px; padding-right: 4px; background-color: #FFEA20; } /* All views: header item coverage table entry for ow coverage rate */ td.headerCovTableEntryLo { text-align: right; color: #000000; font-family: sans-serif; font-weight: bold; white-space: nowrap; padding-left: 12px; padding-right: 4px; background-color: #FF0000; } /* All views: header legend value for legend entry */ td.headerValueLeg { text-align: left; color: #000000; font-family: sans-serif; font-size: 80%; white-space: nowrap; padding-top: 4px; } /* All views: color of horizontal ruler */ td.ruler { background-color: #6688D4; } /* All views: version string format */ td.versionInfo { text-align: center; padding-top: 2px; font-family: sans-serif; font-style: italic; } /* Directory view/File view (all)/Test case descriptions: table headline format */ td.tableHead { text-align: center; color: #FFFFFF; background-color: #6688D4; font-family: sans-serif; font-size: 120%; font-weight: bold; white-space: nowrap; padding-left: 4px; padding-right: 4px; } span.tableHeadSort { padding-right: 4px; } /* Directory view/File view (all): filename entry format */ td.coverFile { text-align: left; padding-left: 10px; padding-right: 20px; color: #284FA8; background-color: #DAE7FE; font-family: monospace; } /* Directory view/File view (all): bar-graph entry format*/ td.coverBar { padding-left: 10px; padding-right: 10px; background-color: #DAE7FE; } /* Directory view/File view (all): bar-graph outline color */ td.coverBarOutline { background-color: #000000; } /* Directory view/File view (all): percentage entry for files with high coverage rate */ td.coverPerHi { text-align: right; padding-left: 10px; padding-right: 10px; background-color: #A7FC9D; font-weight: bold; font-family: sans-serif; } /* Directory view/File view (all): line count entry for files with high coverage rate */ td.coverNumHi { text-align: right; padding-left: 10px; padding-right: 10px; background-color: #A7FC9D; white-space: nowrap; font-family: sans-serif; } /* Directory view/File view (all): percentage entry for files with medium coverage rate */ td.coverPerMed { text-align: right; padding-left: 10px; padding-right: 10px; background-color: #FFEA20; font-weight: bold; font-family: sans-serif; } /* Directory view/File view (all): line count entry for files with medium coverage rate */ td.coverNumMed { text-align: right; padding-left: 10px; padding-right: 10px; background-color: #FFEA20; white-space: nowrap; font-family: sans-serif; } /* Directory view/File view (all): percentage entry for files with low coverage rate */ td.coverPerLo { text-align: right; padding-left: 10px; padding-right: 10px; background-color: #FF0000; font-weight: bold; font-family: sans-serif; } /* Directory view/File view (all): line count entry for files with low coverage rate */ td.coverNumLo { text-align: right; padding-left: 10px; padding-right: 10px; background-color: #FF0000; white-space: nowrap; font-family: sans-serif; } /* File view (all): "show/hide details" link format */ a.detail:link { color: #B8D0FF; font-size:80%; } /* File view (all): "show/hide details" link - visited format */ a.detail:visited { color: #B8D0FF; font-size:80%; } /* File view (all): "show/hide details" link - activated format */ a.detail:active { color: #FFFFFF; font-size:80%; } /* File view (detail): test name entry */ td.testName { text-align: right; padding-right: 10px; background-color: #DAE7FE; font-family: sans-serif; } /* File view (detail): test percentage entry */ td.testPer { text-align: right; padding-left: 10px; padding-right: 10px; background-color: #DAE7FE; font-family: sans-serif; } /* File view (detail): test lines count entry */ td.testNum { text-align: right; padding-left: 10px; padding-right: 10px; background-color: #DAE7FE; font-family: sans-serif; } /* Test case descriptions: test name format*/ dt { font-family: sans-serif; font-weight: bold; } /* Test case descriptions: description table body */ td.testDescription { padding-top: 10px; padding-left: 30px; padding-bottom: 10px; padding-right: 30px; background-color: #DAE7FE; } /* Source code view: function entry */ td.coverFn { text-align: left; padding-left: 10px; padding-right: 20px; color: #284FA8; background-color: #DAE7FE; font-family: monospace; } /* Source code view: function entry zero count*/ td.coverFnLo { text-align: right; padding-left: 10px; padding-right: 10px; background-color: #FF0000; font-weight: bold; font-family: sans-serif; } /* Source code view: function entry nonzero count*/ td.coverFnHi { text-align: right; padding-left: 10px; padding-right: 10px; background-color: #DAE7FE; font-weight: bold; font-family: sans-serif; } /* Source code view: source code format */ pre.source { font-family: monospace; white-space: pre; margin-top: 2px; } /* Source code view: line number format */ span.lineNum { background-color: #EFE383; } /* Source code view: format for lines which were executed */ td.lineCov, span.lineCov { background-color: #CAD7FE; } /* Source code view: format for Cov legend */ span.coverLegendCov { padding-left: 10px; padding-right: 10px; padding-bottom: 2px; background-color: #CAD7FE; } /* Source code view: format for lines which were not executed */ td.lineNoCov, span.lineNoCov { background-color: #FF6230; } /* Source code view: format for NoCov legend */ span.coverLegendNoCov { padding-left: 10px; padding-right: 10px; padding-bottom: 2px; background-color: #FF6230; } /* Source code view (function table): standard link - visited format */ td.lineNoCov > a:visited, td.lineCov > a:visited { color: black; text-decoration: underline; } /* Source code view: format for lines which were executed only in a previous version */ span.lineDiffCov { background-color: #B5F7AF; } /* Source code view: format for branches which were executed * and taken */ span.branchCov { background-color: #CAD7FE; } /* Source code view: format for branches which were executed * but not taken */ span.branchNoCov { background-color: #FF6230; } /* Source code view: format for branches which were not executed */ span.branchNoExec { background-color: #FF6230; } /* Source code view: format for the source code heading line */ pre.sourceHeading { white-space: pre; font-family: monospace; font-weight: bold; margin: 0px; } /* All views: header legend value for low rate */ td.headerValueLegL { font-family: sans-serif; text-align: center; white-space: nowrap; padding-left: 4px; padding-right: 2px; background-color: #FF0000; font-size: 80%; } /* All views: header legend value for med rate */ td.headerValueLegM { font-family: sans-serif; text-align: center; white-space: nowrap; padding-left: 2px; padding-right: 2px; background-color: #FFEA20; font-size: 80%; } /* All views: header legend value for hi rate */ td.headerValueLegH { font-family: sans-serif; text-align: center; white-space: nowrap; padding-left: 2px; padding-right: 4px; background-color: #A7FC9D; font-size: 80%; } /* All views except source code view: legend format for low coverage */ span.coverLegendCovLo { padding-left: 10px; padding-right: 10px; padding-top: 2px; background-color: #FF0000; } /* All views except source code view: legend format for med coverage */ span.coverLegendCovMed { padding-left: 10px; padding-right: 10px; padding-top: 2px; background-color: #FFEA20; } /* All views except source code view: legend format for hi coverage */ span.coverLegendCovHi { padding-left: 10px; padding-right: 10px; padding-top: 2px; background-color: #A7FC9D; } END_OF_CSS ; # ************************************************************* # Remove leading tab from all lines $css_data =~ s/^\t//gm; print(CSS_HANDLE $css_data); close(CSS_HANDLE); } # # get_bar_graph_code(base_dir, cover_found, cover_hit) # # Return a string containing HTML code which implements a bar graph display # for a coverage rate of cover_hit * 100 / cover_found. # sub get_bar_graph_code($$$) { my ($base_dir, $found, $hit) = @_; my $rate; my $alt; my $width; my $remainder; my $png_name; my $graph_code; # Check number of instrumented lines if ($_[1] == 0) { return ""; } $alt = rate($hit, $found, "%"); $width = rate($hit, $found, undef, 0); $remainder = 100 - $width; # Decide which .png file to use $png_name = $rate_png[classify_rate($found, $hit, $med_limit, $hi_limit)]; if ($width == 0) { # Zero coverage $graph_code = (<$alt END_OF_HTML ; } elsif ($width == 100) { # Full coverage $graph_code = (<$alt END_OF_HTML ; } else { # Positive coverage $graph_code = (<$alt$alt END_OF_HTML ; } # Remove leading tabs from all lines $graph_code =~ s/^\t+//gm; chomp($graph_code); return($graph_code); } # # sub classify_rate(found, hit, med_limit, high_limit) # # Return 0 for low rate, 1 for medium rate and 2 for hi rate. # sub classify_rate($$$$) { my ($found, $hit, $med, $hi) = @_; my $rate; if ($found == 0) { return 2; } $rate = rate($hit, $found); if ($rate < $med) { return 0; } elsif ($rate < $hi) { return 1; } return 2; } # # write_html(filehandle, html_code) # # Write out HTML_CODE to FILEHANDLE while removing a leading tabulator mark # in each line of HTML_CODE. # sub write_html(*$) { local *HTML_HANDLE = $_[0]; my $html_code = $_[1]; # Remove leading tab from all lines $html_code =~ s/^\t//gm; print(HTML_HANDLE $html_code) or die("ERROR: cannot write HTML data ($!)\n"); } # # write_html_prolog(filehandle, base_dir, pagetitle) # # Write an HTML prolog common to all HTML files to FILEHANDLE. PAGETITLE will # be used as HTML page title. BASE_DIR contains a relative path which points # to the base directory. # sub write_html_prolog(*$$) { my $basedir = $_[1]; my $pagetitle = $_[2]; my $prolog; $prolog = $html_prolog; $prolog =~ s/\@pagetitle\@/$pagetitle/g; $prolog =~ s/\@basedir\@/$basedir/g; write_html($_[0], $prolog); } # # write_header_prolog(filehandle, base_dir) # # Write beginning of page header HTML code. # sub write_header_prolog(*$) { # ************************************************************* write_html($_[0], < $title END_OF_HTML ; # ************************************************************* } # # write_header_line(handle, content) # # Write a header line with the specified table contents. # sub write_header_line(*@) { my ($handle, @content) = @_; my $entry; write_html($handle, " \n"); foreach $entry (@content) { my ($width, $class, $text, $colspan) = @{$entry}; if (defined($width)) { $width = " width=\"$width\""; } else { $width = ""; } if (defined($class)) { $class = " class=\"$class\""; } else { $class = ""; } if (defined($colspan)) { $colspan = " colspan=\"$colspan\""; } else { $colspan = ""; } $text = "" if (!defined($text)); write_html($handle, " $text\n"); } write_html($handle, " \n"); } # # write_header_epilog(filehandle, base_dir) # # Write end of page header HTML code. # sub write_header_epilog(*$) { # ************************************************************* write_html($_[0], <
END_OF_HTML ; # ************************************************************* } # # write_file_table_prolog(handle, file_heading, ([heading, num_cols], ...)) # # Write heading for file table. # sub write_file_table_prolog(*$@) { my ($handle, $file_heading, @columns) = @_; my $num_columns = 0; my $file_width; my $col; my $width; $width = 20 if (scalar(@columns) == 1); $width = 10 if (scalar(@columns) == 2); $width = 8 if (scalar(@columns) > 2); foreach $col (@columns) { my ($heading, $cols) = @{$col}; $num_columns += $cols; } $file_width = 100 - $num_columns * $width; # Table definition write_html($handle, < END_OF_HTML # Empty first row foreach $col (@columns) { my ($heading, $cols) = @{$col}; while ($cols-- > 0) { write_html($handle, < END_OF_HTML } } # Next row write_html($handle, < END_OF_HTML # Heading row foreach $col (@columns) { my ($heading, $cols) = @{$col}; my $colspan = ""; $colspan = " colspan=$cols" if ($cols > 1); write_html($handle, <$heading END_OF_HTML } write_html($handle, < END_OF_HTML } # write_file_table_entry(handle, base_dir, filename, page_link, # ([ found, hit, med_limit, hi_limit, graph ], ..) # # Write an entry of the file table. # sub write_file_table_entry(*$$$@) { my ($handle, $base_dir, $filename, $page_link, @entries) = @_; my $file_code; my $entry; my $esc_filename = escape_html($filename); # Add link to source if provided if (defined($page_link) && $page_link ne "") { $file_code = "$esc_filename"; } else { $file_code = $esc_filename; } # First column: filename write_html($handle, < END_OF_HTML # Columns as defined foreach $entry (@entries) { my ($found, $hit, $med, $hi, $graph) = @{$entry}; my $bar_graph; my $class; my $rate; # Generate bar graph if requested if ($graph) { $bar_graph = get_bar_graph_code($base_dir, $found, $hit); write_html($handle, < $bar_graph END_OF_HTML } # Get rate color and text if ($found == 0) { $rate = "-"; $class = "Hi"; } else { $rate = rate($hit, $found, " %"); $class = $rate_name[classify_rate($found, $hit, $med, $hi)]; } if ($opt_missed) { # Show negative number of items without coverage $hit = -($found - $hit); } write_html($handle, <$rate END_OF_HTML } # End of row write_html($handle, < END_OF_HTML } # # write_file_table_detail_entry(filehandle, test_name, ([found, hit], ...)) # # Write entry for detail section in file table. # sub write_file_table_detail_entry(*$@) { my ($handle, $test, @entries) = @_; my $entry; if ($test eq "") { $test = "<unnamed>"; } elsif ($test =~ /^(.*),diff$/) { $test = $1." (converted)"; } # Testname write_html($handle, < END_OF_HTML # Test data foreach $entry (@entries) { my ($found, $hit) = @{$entry}; my $rate = rate($hit, $found, " %"); write_html($handle, <$rate END_OF_HTML } write_html($handle, < END_OF_HTML # ************************************************************* } # # write_file_table_epilog(filehandle) # # Write end of file table HTML code. # sub write_file_table_epilog(*) { # ************************************************************* write_html($_[0], <
END_OF_HTML ; # ************************************************************* } # # write_test_table_prolog(filehandle, table_heading) # # Write heading for test case description table. # sub write_test_table_prolog(*$) { # ************************************************************* write_html($_[0], <

$file_heading$file_code$hit / $found$test$hit / $found
$_[1]
END_OF_HTML ; # ************************************************************* } # # write_test_table_entry(filehandle, test_name, test_description) # # Write entry for the test table. # sub write_test_table_entry(*$$) { # ************************************************************* write_html($_[0], <$_[1] 
$_[2]

END_OF_HTML ; # ************************************************************* } # # write_test_table_epilog(filehandle) # # Write end of test description table HTML code. # sub write_test_table_epilog(*) { # ************************************************************* write_html($_[0], <

END_OF_HTML ; # ************************************************************* } sub fmt_centered($$) { my ($width, $text) = @_; my $w0 = length($text); my $w1 = $width > $w0 ? int(($width - $w0) / 2) : 0; my $w2 = $width > $w0 ? $width - $w0 - $w1 : 0; return (" "x$w1).$text.(" "x$w2); } # # write_source_prolog(filehandle) # # Write start of source code table. # sub write_source_prolog(*) { my $lineno_heading = " "; my $branch_heading = ""; my $line_heading = fmt_centered($line_field_width, "Line data"); my $source_heading = " Source code"; if ($br_coverage) { $branch_heading = fmt_centered($br_field_width, "Branch data"). " "; } # ************************************************************* write_html($_[0], < 
${lineno_heading}${branch_heading}${line_heading} ${source_heading}
END_OF_HTML
	;

	# *************************************************************
}

sub cmp_blocks($$)
{
	my ($a, $b) = @_;
	my ($fa, $fb) = ($a->[0], $b->[0]);

	return $fa->[0] <=> $fb->[0] if ($fa->[0] != $fb->[0]);
	return $fa->[1] <=> $fb->[1];
}

#
# get_branch_blocks(brdata)
#
# Group branches that belong to the same basic block.
#
# Returns: [block1, block2, ...]
# block:   [branch1, branch2, ...]
# branch:  [block_num, branch_num, taken_count, text_length, open, close]
#

sub get_branch_blocks($)
{
	my ($brdata) = @_;
	my $last_block_num;
	my $block = [];
	my @blocks;

	return () if (!defined($brdata));

	# Group branches
	foreach my $entry (split(/:/, $brdata)) {
		my ($block_num, $branch, $taken) = split(/,/, $entry);
		my $br;

		if (defined($last_block_num) && $block_num != $last_block_num) {
			push(@blocks, $block);
			$block = [];
		}
		$br = [$block_num, $branch, $taken, 3, 0, 0];
		push(@{$block}, $br);
		$last_block_num = $block_num;
	}
	push(@blocks, $block) if (scalar(@{$block}) > 0);

	# Add braces to first and last branch in group
	foreach $block (@blocks) {
		$block->[0]->[$BR_OPEN] = 1;
		$block->[0]->[$BR_LEN]++;
		$block->[scalar(@{$block}) - 1]->[$BR_CLOSE] = 1;
		$block->[scalar(@{$block}) - 1]->[$BR_LEN]++;
	}

	return sort(cmp_blocks @blocks);
}

#
# get_block_len(block)
#
# Calculate total text length of all branches in a block of branches.
#

sub get_block_len($)
{
	my ($block) = @_;
	my $len = 0;
	my $branch;

	foreach $branch (@{$block}) {
		$len += $branch->[$BR_LEN];
	}

	return $len;
}


#
# get_branch_html(brdata)
#
# Return a list of HTML lines which represent the specified branch coverage
# data in source code view.
#

sub get_branch_html($)
{
	my ($brdata) = @_;
	my @blocks = get_branch_blocks($brdata);
	my $block;
	my $branch;
	my $line_len = 0;
	my $line = [];	# [branch2|" ", branch|" ", ...]
	my @lines;	# [line1, line2, ...]
	my @result;

	# Distribute blocks to lines
	foreach $block (@blocks) {
		my $block_len = get_block_len($block);

		# Does this block fit into the current line?
		if ($line_len + $block_len <= $br_field_width) {
			# Add it
			$line_len += $block_len;
			push(@{$line}, @{$block});
			next;
		} elsif ($block_len <= $br_field_width) {
			# It would fit if the line was empty - add it to new
			# line
			push(@lines, $line);
			$line_len = $block_len;
			$line = [ @{$block} ];
			next;
		}
		# Split the block into several lines
		foreach $branch (@{$block}) {
			if ($line_len + $branch->[$BR_LEN] >= $br_field_width) {
				# Start a new line
				if (($line_len + 1 <= $br_field_width) &&
				    scalar(@{$line}) > 0 &&
				    !$line->[scalar(@$line) - 1]->[$BR_CLOSE]) {
					# Try to align branch symbols to be in
					# one # row
					push(@{$line}, " ");
				}
				push(@lines, $line);
				$line_len = 0;
				$line = [];
			}
			push(@{$line}, $branch);
			$line_len += $branch->[$BR_LEN];
		}
	}
	push(@lines, $line);

	# Convert to HTML
	foreach $line (@lines) {
		my $current = "";
		my $current_len = 0;

		foreach $branch (@$line) {
			# Skip alignment space
			if ($branch eq " ") {
				$current .= " ";
				$current_len++;
				next;
			}

			my ($block_num, $br_num, $taken, $len, $open, $close) =
			   @{$branch};
			my $class;
			my $title;
			my $text;

			if ($taken eq '-') {
				$class	= "branchNoExec";
				$text	= " # ";
				$title	= "Branch $br_num was not executed";
			} elsif ($taken == 0) {
				$class	= "branchNoCov";
				$text	= " - ";
				$title	= "Branch $br_num was not taken";
			} else {
				$class	= "branchCov";
				$text	= " + ";
				$title	= "Branch $br_num was taken $taken ".
					  "time";
				$title .= "s" if ($taken > 1);
			}
			$current .= "[" if ($open);
			$current .= "";
			$current .= $text."";
			$current .= "]" if ($close);
			$current_len += $len;
		}

		# Right-align result text
		if ($current_len < $br_field_width) {
			$current = (" "x($br_field_width - $current_len)).
				   $current;
		}
		push(@result, $current);
	}

	return @result;
}


#
# format_count(count, width)
#
# Return a right-aligned representation of count that fits in width characters.
#

sub format_count($$)
{
	my ($count, $width) = @_;
	my $result;
	my $exp;

	$result = sprintf("%*.0f", $width, $count);
	while (length($result) > $width) {
		last if ($count < 10);
		$exp++;
		$count = int($count/10);
		$result = sprintf("%*s", $width, ">$count*10^$exp");
	}
	return $result;
}

#
# write_source_line(filehandle, line_num, source, hit_count, converted,
#                   brdata)
#
# Write formatted source code line. Return a line in a format as needed
# by gen_png()
#

sub write_source_line(*$$$$$)
{
	my ($handle, $line, $source, $count, $converted, $brdata) = @_;
	my $source_format;
	my $count_format;
	my $result;
	my $anchor_start = "";
	my $anchor_end = "";
	my $count_field_width = $line_field_width - 1;
	my @br_html;
	my $html;

	# Get branch HTML data for this line
	@br_html = get_branch_html($brdata) if ($br_coverage);

	if (!defined($count)) {
		$result		= "";
		$source_format	= "";
		$count_format	= " "x$count_field_width;
	}
	elsif ($count == 0) {
		$result		= $count;
		$source_format	= '';
		$count_format	= format_count($count, $count_field_width);
	}
	elsif ($converted && defined($highlight)) {
		$result		= "*".$count;
		$source_format	= '';
		$count_format	= format_count($count, $count_field_width);
	}
	else {
		$result		= $count;
		$source_format	= '';
		$count_format	= format_count($count, $count_field_width);
	}
	$result .= ":".$source;

	# Write out a line number navigation anchor every $nav_resolution
	# lines if necessary
	$anchor_start	= "";
	$anchor_end	= "";


	# *************************************************************

	$html = $anchor_start;
	$html .= "".sprintf("%8d", $line)." ";
	$html .= shift(@br_html).":" if ($br_coverage);
	$html .= "$source_format$count_format : ";
	$html .= escape_html($source);
	$html .= "" if ($source_format);
	$html .= $anchor_end."\n";

	write_html($handle, $html);

	if ($br_coverage) {
		# Add lines for overlong branch information
		foreach (@br_html) {
			write_html($handle, "".
				   "         $_\n");
		}
	}
	# *************************************************************

	return($result);
}


#
# write_source_epilog(filehandle)
#
# Write end of source code table.
#

sub write_source_epilog(*)
{
	# *************************************************************

	write_html($_[0], <
	      
	    
	  
	  


END_OF_HTML
	;

	# *************************************************************
}


#
# write_html_epilog(filehandle, base_dir[, break_frames])
#
# Write HTML page footer to FILEHANDLE. BREAK_FRAMES should be set when
# this page is embedded in a frameset, clicking the URL link will then
# break this frameset.
#

sub write_html_epilog(*$;$)
{
	my $basedir = $_[1];
	my $break_code = "";
	my $epilog;

	if (defined($_[2]))
	{
		$break_code = " target=\"_parent\"";
	}

	# *************************************************************

	write_html($_[0], <
	    
	    Generated by: $lcov_version
	  
	  

END_OF_HTML
	;

	$epilog = $html_epilog;
	$epilog =~ s/\@basedir\@/$basedir/g;

	write_html($_[0], $epilog);
}


#
# write_frameset(filehandle, basedir, basename, pagetitle)
#
#

sub write_frameset(*$$$)
{
	my $frame_width = $overview_width + 40;

	# *************************************************************

	write_html($_[0], <

	

	
	  
	  $_[3]
	  
	

	
	  
	  
	  
	    <center>Frames not supported by your browser!<br></center>
	  
	

	
END_OF_HTML
	;

	# *************************************************************
}


#
# sub write_overview_line(filehandle, basename, line, link)
#
#

sub write_overview_line(*$$$)
{
	my $y1 = $_[2] - 1;
	my $y2 = $y1 + $nav_resolution - 1;
	my $x2 = $overview_width - 1;

	# *************************************************************

	write_html($_[0], <
END_OF_HTML
	;

	# *************************************************************
}


#
# write_overview(filehandle, basedir, basename, pagetitle, lines)
#
#

sub write_overview(*$$$$)
{
	my $index;
	my $max_line = $_[4] - 1;
	my $offset;

	# *************************************************************

	write_html($_[0], <

	

	
	  $_[3]
	  
	  
	

	
	  
END_OF_HTML
	;

	# *************************************************************

	# Make $offset the next higher multiple of $nav_resolution
	$offset = ($nav_offset + $nav_resolution - 1) / $nav_resolution;
	$offset = sprintf("%d", $offset ) * $nav_resolution;

	# Create image map for overview image
	for ($index = 1; $index <= $_[4]; $index += $nav_resolution)
	{
		# Enforce nav_offset
		if ($index < $offset + 1)
		{
			write_overview_line($_[0], $_[2], $index, 1);
		}
		else
		{
			write_overview_line($_[0], $_[2], $index, $index - $offset);
		}
	}

	# *************************************************************

	write_html($_[0], <

	  

	  Top


	  Overview
	  

	
	
END_OF_HTML
	;

	# *************************************************************
}


sub max($$)
{
	my ($a, $b) = @_;

	return $a if ($a > $b);
	return $b;
}


#
# write_header(filehandle, type, trunc_file_name, rel_file_name, lines_found,
# lines_hit, funcs_found, funcs_hit, sort_type)
#
# Write a complete standard page header. TYPE may be (0, 1, 2, 3, 4)
# corresponding to (directory view header, file view header, source view
# header, test case description header, function view header)
#

sub write_header(*$$$$$$$$$$)
{
	local *HTML_HANDLE = $_[0];
	my $type = $_[1];
	my $trunc_name = $_[2];
	my $rel_filename = $_[3];
	my $lines_found = $_[4];
	my $lines_hit = $_[5];
	my $fn_found = $_[6];
	my $fn_hit = $_[7];
	my $br_found = $_[8];
	my $br_hit = $_[9];
	my $sort_type = $_[10];
	my $base_dir;
	my $view;
	my $test;
	my $base_name;
	my $style;
	my $rate;
	my @row_left;
	my @row_right;
	my $num_rows;
	my $i;
	my $esc_trunc_name = escape_html($trunc_name);

	$base_name = basename($rel_filename);

	# Prepare text for "current view" field
	if ($type == $HDR_DIR)
	{
		# Main overview
		$base_dir = "";
		$view = $overview_title;
	}
	elsif ($type == $HDR_FILE)
	{
		# Directory overview
		$base_dir = get_relative_base_path($rel_filename);
		$view = "".
			"$overview_title - $esc_trunc_name";
	}
	elsif ($type == $HDR_SOURCE || $type == $HDR_FUNC)
	{
		# File view
		my $dir_name = dirname($rel_filename);
		my $esc_base_name = escape_html($base_name);
		my $esc_dir_name = escape_html($dir_name);

		$base_dir = get_relative_base_path($dir_name);
		if ($frames)
		{
			# Need to break frameset when clicking any of these
			# links
			$view = "$overview_title - ".
				"".
				"$esc_dir_name - $esc_base_name";
		}
		else
		{
			$view = "".
				"$overview_title - ".
				"".
				"$esc_dir_name - $esc_base_name";
		}

		# Add function suffix
		if ($func_coverage) {
			$view .= "";
			if ($type == $HDR_SOURCE) {
				if ($sort) {
					$view .= " (source / functions)";
				} else {
					$view .= " (source / functions)";
				}
			} elsif ($type == $HDR_FUNC) {
				$view .= " (source / functions)";
			}
			$view .= "";
		}
	}
	elsif ($type == $HDR_TESTDESC)
	{
		# Test description header
		$base_dir = "";
		$view = "".
			"$overview_title - test case descriptions";
	}

	# Prepare text for "test" field
	$test = escape_html($test_title);

	# Append link to test description page if available
	if (%test_description && ($type != $HDR_TESTDESC))
	{
		if ($frames && ($type == $HDR_SOURCE || $type == $HDR_FUNC))
		{
			# Need to break frameset when clicking this link
			$test .= " ( ".
				 "".
				 "view descriptions )";
		}
		else
		{
			$test .= " ( ".
				 "".
				 "view descriptions )";
		}
	}

	# Write header
	write_header_prolog(*HTML_HANDLE, $base_dir);

	# Left row
	push(@row_left, [[ "10%", "headerItem", "Current view:" ],
			 [ "35%", "headerValue", $view ]]);
	push(@row_left, [[undef, "headerItem", "Test:"],
			 [undef, "headerValue", $test]]);
	push(@row_left, [[undef, "headerItem", "Date:"],
			 [undef, "headerValue", $date]]);

	# Right row
	if ($legend && ($type == $HDR_SOURCE || $type == $HDR_FUNC)) {
		my $text = <hit
            not hit
END_OF_HTML
		if ($br_coverage) {
			$text .= <+ taken
            - not taken
            # not executed
END_OF_HTML
		}
		push(@row_left, [[undef, "headerItem", "Legend:"],
				 [undef, "headerValueLeg", $text]]);
	} elsif ($legend && ($type != $HDR_TESTDESC)) {
		my $text = <low: < $med_limit %
            medium: >= $med_limit %
            high: >= $hi_limit %
END_OF_HTML
		push(@row_left, [[undef, "headerItem", "Legend:"],
				 [undef, "headerValueLeg", $text]]);
	}
	if ($type == $HDR_TESTDESC) {
		push(@row_right, [[ "55%" ]]);
	} else {
		push(@row_right, [["15%", undef, undef ],
				  ["10%", "headerCovTableHead", "Hit" ],
				  ["10%", "headerCovTableHead", "Total" ],
				  ["15%", "headerCovTableHead", "Coverage"]]);
	}
	# Line coverage
	$style = $rate_name[classify_rate($lines_found, $lines_hit,
					  $med_limit, $hi_limit)];
	$rate = rate($lines_hit, $lines_found, " %");
	push(@row_right, [[undef, "headerItem", "Lines:"],
			  [undef, "headerCovTableEntry", $lines_hit],
			  [undef, "headerCovTableEntry", $lines_found],
			  [undef, "headerCovTableEntry$style", $rate]])
			if ($type != $HDR_TESTDESC);
	# Function coverage
	if ($func_coverage) {
		$style = $rate_name[classify_rate($fn_found, $fn_hit,
						  $fn_med_limit, $fn_hi_limit)];
		$rate = rate($fn_hit, $fn_found, " %");
		push(@row_right, [[undef, "headerItem", "Functions:"],
				  [undef, "headerCovTableEntry", $fn_hit],
				  [undef, "headerCovTableEntry", $fn_found],
				  [undef, "headerCovTableEntry$style", $rate]])
			if ($type != $HDR_TESTDESC);
	}
	# Branch coverage
	if ($br_coverage) {
		$style = $rate_name[classify_rate($br_found, $br_hit,
						  $br_med_limit, $br_hi_limit)];
		$rate = rate($br_hit, $br_found, " %");
		push(@row_right, [[undef, "headerItem", "Branches:"],
				  [undef, "headerCovTableEntry", $br_hit],
				  [undef, "headerCovTableEntry", $br_found],
				  [undef, "headerCovTableEntry$style", $rate]])
			if ($type != $HDR_TESTDESC);
	}

	# Print rows
	$num_rows = max(scalar(@row_left), scalar(@row_right));
	for ($i = 0; $i < $num_rows; $i++) {
		my $left = $row_left[$i];
		my $right = $row_right[$i];

		if (!defined($left)) {
			$left = [[undef, undef, undef], [undef, undef, undef]];
		}
		if (!defined($right)) {
			$right = [];
		}
		write_header_line(*HTML_HANDLE, @{$left},
				  [ $i == 0 ? "5%" : undef, undef, undef],
				  @{$right});
	}

	# Fourth line
	write_header_epilog(*HTML_HANDLE, $base_dir);
}

sub get_sorted_by_rate($$)
{
	my ($hash, $type) = @_;

	if ($type == $SORT_LINE) {
		# Sort by line coverage
		return sort({$hash->{$a}[7] <=> $hash->{$b}[7]} keys(%{$hash}));
	} elsif ($type == $SORT_FUNC) {
		# Sort by function coverage;
		return sort({$hash->{$a}[8] <=> $hash->{$b}[8]}	keys(%{$hash}));
	} elsif ($type == $SORT_BRANCH) {
		# Sort by br coverage;
		return sort({$hash->{$a}[9] <=> $hash->{$b}[9]}	keys(%{$hash}));
	}
}

sub get_sorted_by_missed($$)
{
	my ($hash, $type) = @_;

	if ($type == $SORT_LINE) {
		# Sort by number of instrumented lines without coverage
		return sort(
			{
				($hash->{$b}[0] - $hash->{$b}[1]) <=>
				($hash->{$a}[0] - $hash->{$a}[1])
			} keys(%{$hash}));
	} elsif ($type == $SORT_FUNC) {
		# Sort by number of instrumented functions without coverage
		return sort(
			{
				($hash->{$b}[2] - $hash->{$b}[3]) <=>
				($hash->{$a}[2] - $hash->{$a}[3])
			} keys(%{$hash}));
	} elsif ($type == $SORT_BRANCH) {
		# Sort by number of instrumented branches without coverage
		return sort(
			{
				($hash->{$b}[4] - $hash->{$b}[5]) <=>
				($hash->{$a}[4] - $hash->{$a}[5])
			} keys(%{$hash}));
	}
}

#
# get_sorted_keys(hash_ref, sort_type)
#
# hash_ref: filename -> stats
# stats: [ lines_found, lines_hit, fn_found, fn_hit, br_found, br_hit,
#          link_name, line_rate, fn_rate, br_rate ]
#

sub get_sorted_keys($$)
{
	my ($hash, $type) = @_;

	if ($type == $SORT_FILE) {
		# Sort by name
		return sort(keys(%{$hash}));
	} elsif ($opt_missed) {
		return get_sorted_by_missed($hash, $type);
	} else {
		return get_sorted_by_rate($hash, $type);
	}
}

sub get_sort_code($$$)
{
	my ($link, $alt, $base) = @_;
	my $png;
	my $link_start;
	my $link_end;

	if (!defined($link)) {
		$png = "glass.png";
		$link_start = "";
		$link_end = "";
	} else {
		$png = "updown.png";
		$link_start = '';
		$link_end = "";
	}

	return ' '.$link_start.
	       ''.$link_end.'';
}

sub get_file_code($$$$)
{
	my ($type, $text, $sort_button, $base) = @_;
	my $result = $text;
	my $link;

	if ($sort_button) {
		if ($type == $HEAD_NO_DETAIL) {
			$link = "index.$html_ext";
		} else {
			$link = "index-detail.$html_ext";
		}
	}
	$result .= get_sort_code($link, "Sort by name", $base);

	return $result;
}

sub get_line_code($$$$$)
{
	my ($type, $sort_type, $text, $sort_button, $base) = @_;
	my $result = $text;
	my $sort_link;

	if ($type == $HEAD_NO_DETAIL) {
		# Just text
		if ($sort_button) {
			$sort_link = "index-sort-l.$html_ext";
		}
	} elsif ($type == $HEAD_DETAIL_HIDDEN) {
		# Text + link to detail view
		$result .= ' ( show details )';
		if ($sort_button) {
			$sort_link = "index-sort-l.$html_ext";
		}
	} else {
		# Text + link to standard view
		$result .= ' ( hide details )';
		if ($sort_button) {
			$sort_link = "index-detail-sort-l.$html_ext";
		}
	}
	# Add sort button
	$result .= get_sort_code($sort_link, "Sort by line coverage", $base);

	return $result;
}

sub get_func_code($$$$)
{
	my ($type, $text, $sort_button, $base) = @_;
	my $result = $text;
	my $link;

	if ($sort_button) {
		if ($type == $HEAD_NO_DETAIL) {
			$link = "index-sort-f.$html_ext";
		} else {
			$link = "index-detail-sort-f.$html_ext";
		}
	}
	$result .= get_sort_code($link, "Sort by function coverage", $base);
	return $result;
}

sub get_br_code($$$$)
{
	my ($type, $text, $sort_button, $base) = @_;
	my $result = $text;
	my $link;

	if ($sort_button) {
		if ($type == $HEAD_NO_DETAIL) {
			$link = "index-sort-b.$html_ext";
		} else {
			$link = "index-detail-sort-b.$html_ext";
		}
	}
	$result .= get_sort_code($link, "Sort by branch coverage", $base);
	return $result;
}

#
# write_file_table(filehandle, base_dir, overview, testhash, testfnchash,
#                  testbrhash, fileview, sort_type)
#
# Write a complete file table. OVERVIEW is a reference to a hash containing
# the following mapping:
#
#   filename -> "lines_found,lines_hit,funcs_found,funcs_hit,page_link,
#		 func_link"
#
# TESTHASH is a reference to the following hash:
#
#   filename -> \%testdata
#   %testdata: name of test affecting this file -> \%testcount
#   %testcount: line number -> execution count for a single test
#
# Heading of first column is "Filename" if FILEVIEW is true, "Directory name"
# otherwise.
#

sub write_file_table(*$$$$$$$)
{
	local *HTML_HANDLE = $_[0];
	my $base_dir = $_[1];
	my $overview = $_[2];
	my $testhash = $_[3];
	my $testfnchash = $_[4];
	my $testbrhash = $_[5];
	my $fileview = $_[6];
	my $sort_type = $_[7];
	my $filename;
	my $bar_graph;
	my $hit;
	my $found;
	my $fn_found;
	my $fn_hit;
	my $br_found;
	my $br_hit;
	my $page_link;
	my $testname;
	my $testdata;
	my $testfncdata;
	my $testbrdata;
	my %affecting_tests;
	my $line_code = "";
	my $func_code;
	my $br_code;
	my $file_code;
	my @head_columns;

	# Determine HTML code for column headings
	if (($base_dir ne "") && $show_details)
	{
		my $detailed = keys(%{$testhash});

		$file_code = get_file_code($detailed ? $HEAD_DETAIL_HIDDEN :
					$HEAD_NO_DETAIL,
					$fileview ? "Filename" : "Directory",
					$sort && $sort_type != $SORT_FILE,
					$base_dir);
		$line_code = get_line_code($detailed ? $HEAD_DETAIL_SHOWN :
					$HEAD_DETAIL_HIDDEN,
					$sort_type,
					"Line Coverage",
					$sort && $sort_type != $SORT_LINE,
					$base_dir);
		$func_code = get_func_code($detailed ? $HEAD_DETAIL_HIDDEN :
					$HEAD_NO_DETAIL,
					"Functions",
					$sort && $sort_type != $SORT_FUNC,
					$base_dir);
		$br_code = get_br_code($detailed ? $HEAD_DETAIL_HIDDEN :
					$HEAD_NO_DETAIL,
					"Branches",
					$sort && $sort_type != $SORT_BRANCH,
					$base_dir);
	} else {
		$file_code = get_file_code($HEAD_NO_DETAIL,
					$fileview ? "Filename" : "Directory",
					$sort && $sort_type != $SORT_FILE,
					$base_dir);
		$line_code = get_line_code($HEAD_NO_DETAIL, $sort_type, "Line Coverage",
					$sort && $sort_type != $SORT_LINE,
					$base_dir);
		$func_code = get_func_code($HEAD_NO_DETAIL, "Functions",
					$sort && $sort_type != $SORT_FUNC,
					$base_dir);
		$br_code = get_br_code($HEAD_NO_DETAIL, "Branches",
					$sort && $sort_type != $SORT_BRANCH,
					$base_dir);
	}
	push(@head_columns, [ $line_code, 3 ]);
	push(@head_columns, [ $func_code, 2]) if ($func_coverage);
	push(@head_columns, [ $br_code, 2]) if ($br_coverage);

	write_file_table_prolog(*HTML_HANDLE, $file_code, @head_columns);

	foreach $filename (get_sorted_keys($overview, $sort_type))
	{
		my @columns;
		($found, $hit, $fn_found, $fn_hit, $br_found, $br_hit,
		 $page_link) = @{$overview->{$filename}};

		# Line coverage
		push(@columns, [$found, $hit, $med_limit, $hi_limit, 1]);
		# Function coverage
		if ($func_coverage) {
			push(@columns, [$fn_found, $fn_hit, $fn_med_limit,
					$fn_hi_limit, 0]);
		}
		# Branch coverage
		if ($br_coverage) {
			push(@columns, [$br_found, $br_hit, $br_med_limit,
					$br_hi_limit, 0]);
		}
		write_file_table_entry(*HTML_HANDLE, $base_dir, $filename,
				       $page_link, @columns);

		$testdata = $testhash->{$filename};
		$testfncdata = $testfnchash->{$filename};
		$testbrdata = $testbrhash->{$filename};

		# Check whether we should write test specific coverage
		# as well
		if (!($show_details && $testdata)) { next; }

		# Filter out those tests that actually affect this file
		%affecting_tests = %{ get_affecting_tests($testdata,
					$testfncdata, $testbrdata) };

		# Does any of the tests affect this file at all?
		if (!%affecting_tests) { next; }

		foreach $testname (keys(%affecting_tests))
		{
			my @results;
			($found, $hit, $fn_found, $fn_hit, $br_found, $br_hit) =
				split(",", $affecting_tests{$testname});

			# Insert link to description of available
			if ($test_description{$testname})
			{
				$testname = "".
					    "$testname";
			}

			push(@results, [$found, $hit]);
			push(@results, [$fn_found, $fn_hit]) if ($func_coverage);
			push(@results, [$br_found, $br_hit]) if ($br_coverage);
			write_file_table_detail_entry(*HTML_HANDLE, $testname,
				@results);
		}
	}

	write_file_table_epilog(*HTML_HANDLE);
}


#
# get_found_and_hit(hash)
#
# Return the count for entries (found) and entries with an execution count
# greater than zero (hit) in a hash (linenumber -> execution count) as
# a list (found, hit)
#

sub get_found_and_hit($)
{
	my %hash = %{$_[0]};
	my $found = 0;
	my $hit = 0;

	# Calculate sum
	$found = 0;
	$hit = 0;
			
	foreach (keys(%hash))
	{
		$found++;
		if ($hash{$_}>0) { $hit++; }
	}

	return ($found, $hit);
}


#
# get_func_found_and_hit(sumfnccount)
#
# Return (f_found, f_hit) for sumfnccount
#

sub get_func_found_and_hit($)
{
	my ($sumfnccount) = @_;
	my $function;
	my $fn_found;
	my $fn_hit;

	$fn_found = scalar(keys(%{$sumfnccount}));
	$fn_hit = 0;
	foreach $function (keys(%{$sumfnccount})) {
		if ($sumfnccount->{$function} > 0) {
			$fn_hit++;
		}
	}
	return ($fn_found, $fn_hit);
}


sub get_br_found_and_hit($)
{
	my ($brcount) = @_;
	my $db;

	$db = brcount_to_db($brcount);

	return brcount_db_get_found_and_hit($db);
}


#
# get_affecting_tests(testdata, testfncdata, testbrdata)
#
# HASHREF contains a mapping filename -> (linenumber -> exec count). Return
# a hash containing mapping filename -> "lines found, lines hit" for each
# filename which has a nonzero hit count.
#

sub get_affecting_tests($$$)
{
	my ($testdata, $testfncdata, $testbrdata) = @_;
	my $testname;
	my $testcount;
	my $testfnccount;
	my $testbrcount;
	my %result;
	my $found;
	my $hit;
	my $fn_found;
	my $fn_hit;
	my $br_found;
	my $br_hit;

	foreach $testname (keys(%{$testdata}))
	{
		# Get (line number -> count) hash for this test case
		$testcount = $testdata->{$testname};
		$testfnccount = $testfncdata->{$testname};
		$testbrcount = $testbrdata->{$testname};

		# Calculate sum
		($found, $hit) = get_found_and_hit($testcount);
		($fn_found, $fn_hit) = get_func_found_and_hit($testfnccount);
		($br_found, $br_hit) = get_br_found_and_hit($testbrcount);

		if ($hit>0)
		{
			$result{$testname} = "$found,$hit,$fn_found,$fn_hit,".
					     "$br_found,$br_hit";
		}
	}

	return(\%result);
}


sub get_hash_reverse($)
{
	my ($hash) = @_;
	my %result;

	foreach (keys(%{$hash})) {
		$result{$hash->{$_}} = $_;
	}

	return \%result;
}

#
# write_source(filehandle, source_filename, count_data, checksum_data,
#              converted_data, func_data, sumbrcount)
#
# Write an HTML view of a source code file. Returns a list containing
# data as needed by gen_png().
#
# Die on error.
#

sub write_source($$$$$$$)
{
	local *HTML_HANDLE = $_[0];
	local *SOURCE_HANDLE;
	my $source_filename = $_[1];
	my %count_data;
	my $line_number;
	my @result;
	my $checkdata = $_[3];
	my $converted = $_[4];
	my $funcdata  = $_[5];
	my $sumbrcount = $_[6];
	my $datafunc = get_hash_reverse($funcdata);
	my @file;

	if ($_[2])
	{
		%count_data = %{$_[2]};
	}

	if (!open(SOURCE_HANDLE, "<", $source_filename)) {
		my @lines;
		my $last_line = 0;

		if (!$ignore[$ERROR_SOURCE]) {
			die("ERROR: cannot read $source_filename\n");
		}

		# Continue without source file
		warn("WARNING: cannot read $source_filename!\n");

		@lines = sort( { $a <=> $b }  keys(%count_data));
		if (@lines) {
			$last_line = $lines[scalar(@lines) - 1];
		}
		return ( ":" ) if ($last_line < 1);

		# Simulate gcov behavior
		for ($line_number = 1; $line_number <= $last_line;
		     $line_number++) {
			push(@file, "/* EOF */");
		}
	} else {
		@file = ;
	}
	
	write_source_prolog(*HTML_HANDLE);
	$line_number = 0;
	foreach (@file) {
		$line_number++;
		chomp($_);

		# Also remove CR from line-end
		s/\015$//;

		# Source code matches coverage data?
		if (defined($checkdata->{$line_number}) &&
		    ($checkdata->{$line_number} ne md5_base64($_)))
		{
			die("ERROR: checksum mismatch  at $source_filename:".
			    "$line_number\n");
		}

		push (@result,
		      write_source_line(HTML_HANDLE, $line_number,
					$_, $count_data{$line_number},
					$converted->{$line_number},
					$sumbrcount->{$line_number}));
	}

	close(SOURCE_HANDLE);
	write_source_epilog(*HTML_HANDLE);
	return(@result);
}


sub funcview_get_func_code($$$)
{
	my ($name, $base, $type) = @_;
	my $result;
	my $link;

	if ($sort && $type == 1) {
		$link = "$name.func.$html_ext";
	}
	$result = "Function Name";
	$result .= get_sort_code($link, "Sort by function name", $base);

	return $result;
}

sub funcview_get_count_code($$$)
{
	my ($name, $base, $type) = @_;
	my $result;
	my $link;

	if ($sort && $type == 0) {
		$link = "$name.func-sort-c.$html_ext";
	}
	$result = "Hit count";
	$result .= get_sort_code($link, "Sort by hit count", $base);

	return $result;
}

#
# funcview_get_sorted(funcdata, sumfncdata, sort_type)
#
# Depending on the value of sort_type, return a list of functions sorted
# by name (type 0) or by the associated call count (type 1).
#

sub funcview_get_sorted($$$)
{
	my ($funcdata, $sumfncdata, $type) = @_;

	if ($type == 0) {
		return sort(keys(%{$funcdata}));
	}
	return sort({
		$sumfncdata->{$b} == $sumfncdata->{$a} ?
			$a cmp $b : $sumfncdata->{$a} <=> $sumfncdata->{$b}
		} keys(%{$sumfncdata}));
}

sub demangle_list($)
{
	my ($list) = @_;
	my $tmpfile;
	my $handle;
	my %demangle;
	my $demangle_arg = "";
	my %versions;

	# Write function names to file
	($handle, $tmpfile) = tempfile();
	die("ERROR: could not create temporary file") if (!defined($tmpfile));
	print($handle join("\n", @$list));
	close($handle);

	# Extra flag necessary on OS X so that symbols listed by gcov get demangled
	# properly.
	if ($^O eq "darwin") {
		$demangle_arg = "--no-strip-underscores";
	}

	# Build translation hash from c++filt output
	open($handle, "-|", "c++filt $demangle_arg < $tmpfile") or
		die("ERROR: could not run c++filt: $!\n");
	foreach my $func (@$list) {
		my $translated = <$handle>;
		my $version;

		last if (!defined($translated));
		chomp($translated);

		$version = ++$versions{$translated};
		$translated .= ".$version" if ($version > 1);
		$demangle{$func} = $translated;
	}
	close($handle);

	if (scalar(keys(%demangle)) != scalar(@$list)) {
		die("ERROR: c++filt output not as expected (".
		    scalar(keys(%demangle))." vs ".scalar(@$list).") lines\n");
	}

	unlink($tmpfile) or
		warn("WARNING: could not remove temporary file $tmpfile: $!\n");

	return \%demangle;
}

#
# write_function_table(filehandle, source_file, sumcount, funcdata,
#		       sumfnccount, testfncdata, sumbrcount, testbrdata,
#		       base_name, base_dir, sort_type)
#
# Write an HTML table listing all functions in a source file, including
# also function call counts and line coverages inside of each function.
#
# Die on error.
#

sub write_function_table(*$$$$$$$$$$)
{
	local *HTML_HANDLE = $_[0];
	my $source = $_[1];
	my $sumcount = $_[2];
	my $funcdata = $_[3];
	my $sumfncdata = $_[4];
	my $testfncdata = $_[5];
	my $sumbrcount = $_[6];
	my $testbrdata = $_[7];
	my $name = $_[8];
	my $base = $_[9];
	my $type = $_[10];
	my $func;
	my $func_code;
	my $count_code;
	my $demangle;

	# Get HTML code for headings
	$func_code = funcview_get_func_code($name, $base, $type);
	$count_code = funcview_get_count_code($name, $base, $type);
	write_html(*HTML_HANDLE, <
	  
END_OF_HTML
	;

	# Get demangle translation hash
	if ($demangle_cpp) {
		$demangle = demangle_list([ sort(keys(%{$funcdata})) ]);
	}

	# Get a sorted table
	foreach $func (funcview_get_sorted($funcdata, $sumfncdata, $type)) {
		if (!defined($funcdata->{$func}))
		{
			next;
		}

		my $startline = $funcdata->{$func} - $func_offset;
		my $name = $func;
		my $count = $sumfncdata->{$name};
		my $countstyle;

		# Replace function name with demangled version if available
		$name = $demangle->{$name} if (exists($demangle->{$name}));

		# Escape special characters
		$name = escape_html($name);
		if ($startline < 1) {
			$startline = 1;
		}
		if ($count == 0) {
			$countstyle = "coverFnLo";
		} else {
			$countstyle = "coverFnHi";
		}

		write_html(*HTML_HANDLE, <
              
END_OF_HTML
                ;
	}
	write_html(*HTML_HANDLE, <
	  

	  
END_OF_HTML
	;
}


#
# info(printf_parameter)
#
# Use printf to write PRINTF_PARAMETER to stdout only when the $quiet flag
# is not set.
#

sub info(@)
{
	if (!$quiet)
	{
		# Print info string
		printf(@_);
	}
}


#
# subtract_counts(data_ref, base_ref)
#

sub subtract_counts($$)
{
	my %data = %{$_[0]};
	my %base = %{$_[1]};
	my $line;
	my $data_count;
	my $base_count;
	my $hit = 0;
	my $found = 0;

	foreach $line (keys(%data))
	{
		$found++;
		$data_count = $data{$line};
		$base_count = $base{$line};

		if (defined($base_count))
		{
			$data_count -= $base_count;

			# Make sure we don't get negative numbers
			if ($data_count<0) { $data_count = 0; }
		}

		$data{$line} = $data_count;
		if ($data_count > 0) { $hit++; }
	}

	return (\%data, $found, $hit);
}


#
# subtract_fnccounts(data, base)
#
# Subtract function call counts found in base from those in data.
# Return (data, f_found, f_hit).
#

sub subtract_fnccounts($$)
{
	my %data;
	my %base;
	my $func;
	my $data_count;
	my $base_count;
	my $fn_hit = 0;
	my $fn_found = 0;

	%data = %{$_[0]} if (defined($_[0]));
	%base = %{$_[1]} if (defined($_[1]));
	foreach $func (keys(%data)) {
		$fn_found++;
		$data_count = $data{$func};
		$base_count = $base{$func};

		if (defined($base_count)) {
			$data_count -= $base_count;

			# Make sure we don't get negative numbers
			if ($data_count < 0) {
				$data_count = 0;
			}
		}

		$data{$func} = $data_count;
		if ($data_count > 0) {
			$fn_hit++;
		}
	}

	return (\%data, $fn_found, $fn_hit);
}


#
# apply_baseline(data_ref, baseline_ref)
#
# Subtract the execution counts found in the baseline hash referenced by
# BASELINE_REF from actual data in DATA_REF.
#

sub apply_baseline($$)
{
	my %data_hash = %{$_[0]};
	my %base_hash = %{$_[1]};
	my $filename;
	my $testname;
	my $data;
	my $data_testdata;
	my $data_funcdata;
	my $data_checkdata;
	my $data_testfncdata;
	my $data_testbrdata;
	my $data_count;
	my $data_testfnccount;
	my $data_testbrcount;
	my $base;
	my $base_checkdata;
	my $base_sumfnccount;
	my $base_sumbrcount;
	my $base_count;
	my $sumcount;
	my $sumfnccount;
	my $sumbrcount;
	my $found;
	my $hit;
	my $fn_found;
	my $fn_hit;
	my $br_found;
	my $br_hit;

	foreach $filename (keys(%data_hash))
	{
		# Get data set for data and baseline
		$data = $data_hash{$filename};
		$base = $base_hash{$filename};

		# Skip data entries for which no base entry exists
		if (!defined($base))
		{
			next;
		}

		# Get set entries for data and baseline
		($data_testdata, undef, $data_funcdata, $data_checkdata,
		 $data_testfncdata, undef, $data_testbrdata) =
			get_info_entry($data);
		(undef, $base_count, undef, $base_checkdata, undef,
		 $base_sumfnccount, undef, $base_sumbrcount) =
			get_info_entry($base);

		# Check for compatible checksums
		merge_checksums($data_checkdata, $base_checkdata, $filename);

		# sumcount has to be calculated anew
		$sumcount = {};
		$sumfnccount = {};
		$sumbrcount = {};

		# For each test case, subtract test specific counts
		foreach $testname (keys(%{$data_testdata}))
		{
			# Get counts of both data and baseline
			$data_count = $data_testdata->{$testname};
			$data_testfnccount = $data_testfncdata->{$testname};
			$data_testbrcount = $data_testbrdata->{$testname};

			($data_count, undef, $hit) =
				subtract_counts($data_count, $base_count);
			($data_testfnccount) =
				subtract_fnccounts($data_testfnccount,
						   $base_sumfnccount);
			($data_testbrcount) =
				combine_brcount($data_testbrcount,
						 $base_sumbrcount, $BR_SUB);


			# Check whether this test case did hit any line at all
			if ($hit > 0)
			{
				# Write back resulting hash
				$data_testdata->{$testname} = $data_count;
				$data_testfncdata->{$testname} =
					$data_testfnccount;
				$data_testbrdata->{$testname} =
					$data_testbrcount;
			}
			else
			{
				# Delete test case which did not impact this
				# file
				delete($data_testdata->{$testname});
				delete($data_testfncdata->{$testname});
				delete($data_testbrdata->{$testname});
			}

			# Add counts to sum of counts
			($sumcount, $found, $hit) =
				add_counts($sumcount, $data_count);
			($sumfnccount, $fn_found, $fn_hit) =
				add_fnccount($sumfnccount, $data_testfnccount);
			($sumbrcount, $br_found, $br_hit) =
				combine_brcount($sumbrcount, $data_testbrcount,
						$BR_ADD);
		}

		# Write back resulting entry
		set_info_entry($data, $data_testdata, $sumcount, $data_funcdata,
			       $data_checkdata, $data_testfncdata, $sumfnccount,
			       $data_testbrdata, $sumbrcount, $found, $hit,
			       $fn_found, $fn_hit, $br_found, $br_hit);

		$data_hash{$filename} = $data;
	}

	return (\%data_hash);
}


#
# remove_unused_descriptions()
#
# Removes all test descriptions from the global hash %test_description which
# are not present in %info_data.
#

sub remove_unused_descriptions()
{
	my $filename;		# The current filename
	my %test_list;		# Hash containing found test names
	my $test_data;		# Reference to hash test_name -> count_data
	my $before;		# Initial number of descriptions
	my $after;		# Remaining number of descriptions
	
	$before = scalar(keys(%test_description));

	foreach $filename (keys(%info_data))
	{
		($test_data) = get_info_entry($info_data{$filename});
		foreach (keys(%{$test_data}))
		{
			$test_list{$_} = "";
		}
	}

	# Remove descriptions for tests which are not in our list
	foreach (keys(%test_description))
	{
		if (!defined($test_list{$_}))
		{
			delete($test_description{$_});
		}
	}

	$after = scalar(keys(%test_description));
	if ($after < $before)
	{
		info("Removed ".($before - $after).
		     " unused descriptions, $after remaining.\n");
	}
}


#
# apply_prefix(filename, PREFIXES)
#
# If FILENAME begins with PREFIX from PREFIXES, remove PREFIX from FILENAME
# and return resulting string, otherwise return FILENAME.
#

sub apply_prefix($@)
{
	my $filename = shift;
	my @dir_prefix = @_;

	if (@dir_prefix)
	{
		foreach my $prefix (@dir_prefix)
		{
			if ($prefix ne "" && $filename =~ /^\Q$prefix\E\/(.*)$/)
			{
				return substr($filename, length($prefix) + 1);
			}
		}
	}

	return $filename;
}


#
# system_no_output(mode, parameters)
#
# Call an external program using PARAMETERS while suppressing depending on
# the value of MODE:
#
#   MODE & 1: suppress STDOUT
#   MODE & 2: suppress STDERR
#
# Return 0 on success, non-zero otherwise.
#

sub system_no_output($@)
{
	my $mode = shift;
	my $result;
	local *OLD_STDERR;
	local *OLD_STDOUT;

	# Save old stdout and stderr handles
	($mode & 1) && open(OLD_STDOUT, ">>&", "STDOUT");
	($mode & 2) && open(OLD_STDERR, ">>&", "STDERR");

	# Redirect to /dev/null
	($mode & 1) && open(STDOUT, ">", "/dev/null");
	($mode & 2) && open(STDERR, ">", "/dev/null");

	system(@_);
	$result = $?;

	# Close redirected handles
	($mode & 1) && close(STDOUT);
	($mode & 2) && close(STDERR);

	# Restore old handles
	($mode & 1) && open(STDOUT, ">>&", "OLD_STDOUT");
	($mode & 2) && open(STDERR, ">>&", "OLD_STDERR");

	return $result;
}


#
# read_config(filename)
#
# Read configuration file FILENAME and return a reference to a hash containing
# all valid key=value pairs found.
#

sub read_config($)
{
	my $filename = $_[0];
	my %result;
	my $key;
	my $value;
	local *HANDLE;

	if (!open(HANDLE, "<", $filename))
	{
		warn("WARNING: cannot read configuration file $filename\n");
		return undef;
	}
	while ()
	{
		chomp;
		# Skip comments
		s/#.*//;
		# Remove leading blanks
		s/^\s+//;
		# Remove trailing blanks
		s/\s+$//;
		next unless length;
		($key, $value) = split(/\s*=\s*/, $_, 2);
		if (defined($key) && defined($value))
		{
			$result{$key} = $value;
		}
		else
		{
			warn("WARNING: malformed statement in line $. ".
			     "of configuration file $filename\n");
		}
	}
	close(HANDLE);
	return \%result;
}


#
# apply_config(REF)
#
# REF is a reference to a hash containing the following mapping:
#
#   key_string => var_ref
#
# where KEY_STRING is a keyword and VAR_REF is a reference to an associated
# variable. If the global configuration hashes CONFIG or OPT_RC contain a value
# for keyword KEY_STRING, VAR_REF will be assigned the value for that keyword. 
#

sub apply_config($)
{
	my $ref = $_[0];

	foreach (keys(%{$ref}))
	{
		if (defined($opt_rc{$_})) {
			${$ref->{$_}} = $opt_rc{$_};
		} elsif (defined($config->{$_})) {
			${$ref->{$_}} = $config->{$_};
		}
	}
}


#
# get_html_prolog(FILENAME)
#
# If FILENAME is defined, return contents of file. Otherwise return default
# HTML prolog. Die on error.
#

sub get_html_prolog($)
{
	my $filename = $_[0];
	my $result = "";

	if (defined($filename))
	{
		local *HANDLE;

		open(HANDLE, "<", $filename)
			or die("ERROR: cannot open html prolog $filename!\n");
		while ()
		{
			$result .= $_;
		}
		close(HANDLE);
	}
	else
	{
		$result = <




  
  \@pagetitle\@
  




END_OF_HTML
		;
	}

	return $result;
}


#
# get_html_epilog(FILENAME)
#
# If FILENAME is defined, return contents of file. Otherwise return default
# HTML epilog. Die on error.
#
sub get_html_epilog($)
{
	my $filename = $_[0];
	my $result = "";

	if (defined($filename))
	{
		local *HANDLE;

		open(HANDLE, "<", $filename)
			or die("ERROR: cannot open html epilog $filename!\n");
		while ()
		{
			$result .= $_;
		}
		close(HANDLE);
	}
	else
	{
		$result = <

END_OF_HTML
		;
	}

	return $result;

}

sub warn_handler($)
{
	my ($msg) = @_;

	warn("$tool_name: $msg");
}

sub die_handler($)
{
	my ($msg) = @_;

	die("$tool_name: $msg");
}

#
# parse_ignore_errors(@ignore_errors)
#
# Parse user input about which errors to ignore.
#

sub parse_ignore_errors(@)
{
	my (@ignore_errors) = @_;
	my @items;
	my $item;

	return if (!@ignore_errors);

	foreach $item (@ignore_errors) {
		$item =~ s/\s//g;
		if ($item =~ /,/) {
			# Split and add comma-separated parameters
			push(@items, split(/,/, $item));
		} else {
			# Add single parameter
			push(@items, $item);
		}
	}
	foreach $item (@items) {
		my $item_id = $ERROR_ID{lc($item)};

		if (!defined($item_id)) {
			die("ERROR: unknown argument for --ignore-errors: ".
			    "$item\n");
		}
		$ignore[$item_id] = 1;
	}
}

#
# parse_dir_prefix(@dir_prefix)
#
# Parse user input about the prefix list
#

sub parse_dir_prefix(@)
{
	my (@opt_dir_prefix) = @_;
	my $item;

	return if (!@opt_dir_prefix);

	foreach $item (@opt_dir_prefix) {
		if ($item =~ /,/) {
			# Split and add comma-separated parameters
			push(@dir_prefix, split(/,/, $item));
		} else {
			# Add single parameter
			push(@dir_prefix, $item);
		}
	}
}

#
# rate(hit, found[, suffix, precision, width])
#
# Return the coverage rate [0..100] for HIT and FOUND values. 0 is only
# returned when HIT is 0. 100 is only returned when HIT equals FOUND.
# PRECISION specifies the precision of the result. SUFFIX defines a
# string that is appended to the result if FOUND is non-zero. Spaces
# are added to the start of the resulting string until it is at least WIDTH
# characters wide.
#

sub rate($$;$$$)
{
        my ($hit, $found, $suffix, $precision, $width) = @_;
        my $rate; 

	# Assign defaults if necessary
	$precision	= $default_precision if (!defined($precision));
	$suffix		= ""	if (!defined($suffix));
	$width		= 0	if (!defined($width));
        
        return sprintf("%*s", $width, "-") if (!defined($found) || $found == 0);
        $rate = sprintf("%.*f", $precision, $hit * 100 / $found);

	# Adjust rates if necessary
        if ($rate == 0 && $hit > 0) {
		$rate = sprintf("%.*f", $precision, 1 / 10 ** $precision);
        } elsif ($rate == 100 && $hit != $found) {
		$rate = sprintf("%.*f", $precision, 100 - 1 / 10 ** $precision);
	}

	return sprintf("%*s", $width, $rate.$suffix);
}
lcov-1.14/bin/gendesc0000755000000000000000000001103513435751076013221 0ustar  rootroot#!/usr/bin/env perl
#
#   Copyright (c) International Business Machines  Corp., 2002
#
#   This program is free software;  you can redistribute it and/or modify
#   it under the terms of the GNU General Public License as published by
#   the Free Software Foundation; either version 2 of the License, or (at
#   your option) any later version.
#
#   This program is distributed in the hope that it will be useful, but
#   WITHOUT ANY WARRANTY;  without even the implied warranty of
#   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
#   General Public License for more details.                 
#
#   You should have received a copy of the GNU General Public License
#   along with this program;  if not, write to the Free Software
#   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
#
#
# gendesc
#
#   This script creates a description file as understood by genhtml.
#   Input file format:
#
#   For each test case:
#     
#     
#   
#   Actual description may consist of several lines. By default, output is
#   written to stdout. Test names consist of alphanumeric characters
#   including _ and -.
#
#
# History:
#   2002-09-02: created by Peter Oberparleiter 
#

use strict;
use warnings;
use File::Basename; 
use Getopt::Long;
use Cwd qw/abs_path/;


# Constants
our $tool_dir		= abs_path(dirname($0));
our $lcov_version	= "LCOV version 1.14";
our $lcov_url		= "http://ltp.sourceforge.net/coverage/lcov.php";
our $tool_name		= basename($0);


# Prototypes
sub print_usage(*);
sub gen_desc();
sub warn_handler($);
sub die_handler($);


# Global variables
our $help;
our $version;
our $output_filename;
our $input_filename;


#
# Code entry point
#

$SIG{__WARN__} = \&warn_handler;
$SIG{__DIE__} = \&die_handler;

# Parse command line options
if (!GetOptions("output-filename=s" => \$output_filename,
		"version" =>\$version,
		"help|?" => \$help
		))
{
	print(STDERR "Use $tool_name --help to get usage information\n");
	exit(1);
}

$input_filename = $ARGV[0];

# Check for help option
if ($help)
{
	print_usage(*STDOUT);
	exit(0);
}

# Check for version option
if ($version)
{
	print("$tool_name: $lcov_version\n");
	exit(0);
}


# Check for input filename
if (!$input_filename)
{
	die("No input filename specified\n".
	    "Use $tool_name --help to get usage information\n");
}

# Do something
gen_desc();


#
# print_usage(handle)
#
# Write out command line usage information to given filehandle.
#

sub print_usage(*)
{
	local *HANDLE = $_[0];

	print(HANDLE <
#    TD:
#
# If defined, write output to OUTPUT_FILENAME, otherwise to stdout.
#
# Die on error.
#

sub gen_desc()
{
	local *INPUT_HANDLE;
	local *OUTPUT_HANDLE;
	my $empty_line = "ignore";

	open(INPUT_HANDLE, "<", $input_filename)
		or die("ERROR: cannot open $input_filename!\n");

	# Open output file for writing
	if ($output_filename)
	{
		open(OUTPUT_HANDLE, ">", $output_filename)
			or die("ERROR: cannot create $output_filename!\n");
	}
	else
	{
		*OUTPUT_HANDLE = *STDOUT;
	}

	# Process all lines in input file
	while ()
	{
		chomp($_);

		if (/^(\w[\w-]*)(\s*)$/)
		{
			# Matched test name
			# Name starts with alphanum or _, continues with
			# alphanum, _ or -
			print(OUTPUT_HANDLE "TN: $1\n");
			$empty_line = "ignore";
		}
		elsif (/^(\s+)(\S.*?)\s*$/)
		{
			# Matched test description
			if ($empty_line eq "insert")
			{
				# Write preserved empty line
				print(OUTPUT_HANDLE "TD: \n");
			}
			print(OUTPUT_HANDLE "TD: $2\n");
			$empty_line = "observe";
		}
		elsif (/^\s*$/)
		{
			# Matched empty line to preserve paragraph separation
			# inside description text
			if ($empty_line eq "observe")
			{
				$empty_line = "insert";
			}
		}
	}

	# Close output file if defined
	if ($output_filename)
	{
		close(OUTPUT_HANDLE);
	}

	close(INPUT_HANDLE);
}

sub warn_handler($)
{
	my ($msg) = @_;

	warn("$tool_name: $msg");
}

sub die_handler($)
{
	my ($msg) = @_;

	die("$tool_name: $msg");
}
lcov-1.14/bin/get_changes.sh0000755000000000000000000000044413435760116014466 0ustar  rootroot#!/usr/bin/env bash
#
# Usage: get_changes.sh
#
# Print lcov change log information as provided by Git

TOOLDIR=$(cd $(dirname $0) >/dev/null ; pwd)

cd $TOOLDIR

if ! git --no-pager log --no-merges --decorate=short --color=never 2>/dev/null ; then
	cat "$TOOLDIR/../CHANGES" 2>/dev/null
fi 
lcov-1.14/bin/get_version.sh0000755000000000000000000000154013435760116014541 0ustar  rootroot#!/usr/bin/env bash
#
# Usage: get_version.sh --version|--release|--full
#
# Print lcov version or release information as provided by Git, .version
# or a fallback.

TOOLDIR=$(cd $(dirname $0) >/dev/null ; pwd)
GITVER=$(cd $TOOLDIR ; git describe --tags 2>/dev/null)

if [ -z "$GITVER" ] ; then
	# Get version information from file
	if [ -e "$TOOLDIR/../.version" ] ; then
		source "$TOOLDIR/../.version"
	fi
else
	# Get version information from git
	FULL=${GITVER:1}
	VERSION=${GITVER%%-*}
	VERSION=${VERSION:1}
	if [ "${GITVER#*-}" != "$GITVER" ] ; then
		RELEASE=${GITVER#*-}
		RELEASE=${RELEASE/-/.}
	fi
fi

# Fallback
[ -z "$VERSION" ] && VERSION="1.0"
[ -z "$RELEASE" ] && RELEASE="1"
[ -z "$FULL" ]    && FULL="$VERSION"

[ "$1" == "--version" ] && echo -n "$VERSION"
[ "$1" == "--release" ] && echo -n "$RELEASE"
[ "$1" == "--full"    ] && echo -n "$FULL"
lcov-1.14/bin/install.sh0000755000000000000000000000253513435775412013675 0ustar  rootroot#!/usr/bin/env bash
#
# install.sh [--uninstall] sourcefile targetfile [install options]
#


# Check for uninstall option
if test "x$1" == "x--uninstall" ; then
  UNINSTALL=true
  SOURCE=$2
  TARGET=$3
  shift 3
else
  UNINSTALL=false
  SOURCE=$1
  TARGET=$2
  shift 2
fi

# Check usage
if test -z "$SOURCE" || test -z "$TARGET" ; then
  echo Usage: install.sh [--uninstall] source target [install options] >&2
  exit 1
fi


#
# do_install(SOURCE_FILE, TARGET_FILE)
#

do_install()
{
  local SOURCE=$1
  local TARGET=$2
  local PARAMS=$3

  install -d $(dirname $TARGET)
  install -p $PARAMS $SOURCE $TARGET
  if [ -n "$LCOV_PERL_PATH" ] ; then
    # Replace Perl interpreter specification
    sed -e "1 s%^#\!.*perl.*$%#\!$LCOV_PERL_PATH%" -i $TARGET
  fi
}


#
# do_uninstall(SOURCE_FILE, TARGET_FILE)
#

do_uninstall()
{
  local SOURCE=$1
  local TARGET=$2

  # Does target exist?
  if test -r $TARGET ; then
    # Is target of the same version as this package?
    if diff -I '^our \$lcov_version' -I '^\.TH ' -I '^#!' $SOURCE $TARGET >/dev/null; then
      rm -f $TARGET
    else
      echo WARNING: Skipping uninstall for $TARGET - versions differ! >&2
    fi
  else
    echo WARNING: Skipping uninstall for $TARGET - not installed! >&2
  fi
}


# Call sub routine
if $UNINSTALL ; then
  do_uninstall $SOURCE $TARGET
else
  do_install $SOURCE $TARGET "$*"
fi

exit 0
lcov-1.14/bin/copy_dates.sh0000755000000000000000000000210113435760116014341 0ustar  rootroot#!/usr/bin/env bash
#
# Usage: copy_dates.sh SOURCE TARGET
#
# For each file found in SOURCE, set the modification time of the copy of that
# file in TARGET to either the time of the latest Git commit (if SOURCE contains
# a Git repository and the file was not modified after the last commit), or the
# modification time of the original file.

SOURCE="$1"
TARGET="$2"

if [ -z "$SOURCE" -o -z "$TARGET" ] ; then
	echo "Usage: $0 SOURCE TARGET" >&2
	exit 1
fi

[ -d "$SOURCE/.git" ] ; NOGIT=$?

echo "Copying modification/commit times from $SOURCE to $TARGET"

cd "$SOURCE" || exit 1
find * -type f | while read FILENAME ; do
	[ ! -e "$TARGET/$FILENAME" ] && continue

	# Copy modification time
	touch -m "$TARGET/$FILENAME" -r "$FILENAME"

	[ $NOGIT -eq 1 ] && continue				# No Git
	git diff --quiet -- "$FILENAME" || continue		# Modified
	git diff --quiet --cached -- "$FILENAME" || continue	# Modified

	# Apply modification time from Git commit time
	TIME=$(git log --pretty=format:%cd -n 1 --date=iso -- "$FILENAME")
	[ -n "$TIME" ] && touch -m "$TARGET/$FILENAME" --date "$TIME"
done
lcov-1.14/bin/updateversion.pl0000755000000000000000000001143713435760116015114 0ustar  rootroot#!/usr/bin/env perl

use strict;
use warnings;

use File::Basename;

sub update_man_page($);
sub update_bin_tool($);
sub update_txt_file($);
sub update_spec_file($);
sub write_version_file($);
sub get_file_info($);

our $directory = $ARGV[0];
our $version = $ARGV[1];
our $release = $ARGV[2];
our $full = $ARGV[3];

our @man_pages = ("man/gendesc.1",  "man/genhtml.1",  "man/geninfo.1",
		  "man/genpng.1", "man/lcov.1", "man/lcovrc.5");
our @bin_tools = ("bin/gendesc", "bin/genhtml", "bin/geninfo",
		  "bin/genpng", "bin/lcov");
our @txt_files = ("README");
our @spec_files = ("rpm/lcov.spec");

if (!defined($directory) || !defined($version) || !defined($release)) {
	die("Usage: $0 DIRECTORY|FILE VERSION RELEASE FULL_VERSION\n");
}

# Determine mode of operation
if (-f $directory) {
	my $file = $directory;
	my $base = basename($file);

	if (grep(/^$base$/, map({ basename($_) } @man_pages))) {
		print("Updating man page $file\n");
		update_man_page($file);
	} elsif (grep(/^$base$/, map({ basename($_) } @bin_tools))) {
		print("Updating bin tool $file\n");
		update_bin_tool($file);
	} elsif (grep(/^$base$/, map({ basename($_) } @txt_files))) {
		print("Updating text file $file\n");
		update_txt_file($file);
	} elsif (grep(/^$base$/, map({ basename($_) } @spec_files))) {
		print("Updating spec file $file\n");
		update_spec_file($file);
	} elsif ($base eq ".version") {
		print("Updating version file $file\n");
		write_version_file($file);
	} else {
		print("WARNING: Skipping unknown file $file\n");
	}
	print("Done.\n");
	exit(0);
}

foreach (@man_pages) {
	print("Updating man page $_\n");
	update_man_page($directory."/".$_);
}
foreach (@bin_tools) {
	print("Updating bin tool $_\n");
	update_bin_tool($directory."/".$_);
}
foreach (@txt_files) {
	print("Updating text file $_\n");
	update_txt_file($directory."/".$_);
}
foreach (@spec_files) {
	print("Updating spec file $_\n");
	update_spec_file($directory."/".$_);
}
print("Updating version file $directory/.version\n");
write_version_file("$directory/.version");
print("Done.\n");

sub get_file_info($)
{
	my ($filename) = @_;
	my ($sec, $min, $hour, $year, $month, $day);
	my @stat;
	my $gittime;

	return (0, 0, 0) if (!-e $filename);
	@stat = stat($filename);
	($sec, $min, $hour, $day, $month, $year) = gmtime($stat[9]);
	$year += 1900;
	$month += 1;

	return (sprintf("%04d-%02d-%02d", $year, $month, $day),
		sprintf("%04d%02d%02d%02d%02d.%02d", $year, $month, $day,
			$hour, $min, $sec),
		sprintf("%o", $stat[2] & 07777));
}

sub update_man_page($)
{
	my ($filename) = @_;
	my @date = get_file_info($filename);
	my $date_string = $date[0];
	local *IN;
	local *OUT;

	$date_string =~ s/-/\\-/g;
	open(IN, "<$filename") || die ("Error: cannot open $filename\n");
	open(OUT, ">$filename.new") ||
		die("Error: cannot create $filename.new\n");
	while () {
		s/\"LCOV\s+\d+\.\d+\"/\"LCOV $version\"/g;
		s/\d\d\d\d\\\-\d\d\\\-\d\d/$date_string/g;
		print(OUT $_);
	}
	close(OUT);
	close(IN);
	chmod(oct($date[2]), "$filename.new");
	system("mv", "-f", "$filename.new", "$filename");
	system("touch", "$filename", "-t", $date[1]);
}

sub update_bin_tool($)
{
	my ($filename) = @_;
	my @date = get_file_info($filename);
	local *IN;
	local *OUT;

	open(IN, "<$filename") || die ("Error: cannot open $filename\n");
	open(OUT, ">$filename.new") ||
		die("Error: cannot create $filename.new\n");
	while () {
		s/^(our\s+\$lcov_version\s*=).*$/$1 "LCOV version $full";/g;
		print(OUT $_);
	}
	close(OUT);
	close(IN);
	chmod(oct($date[2]), "$filename.new");
	system("mv", "-f", "$filename.new", "$filename");
	system("touch", "$filename", "-t", $date[1]);
}

sub update_txt_file($)
{
	my ($filename) = @_;
	my @date = get_file_info($filename);
	local *IN;
	local *OUT;

	open(IN, "<$filename") || die ("Error: cannot open $filename\n");
	open(OUT, ">$filename.new") ||
		die("Error: cannot create $filename.new\n");
	while () {
		s/(Last\s+changes:\s+)\d\d\d\d-\d\d-\d\d/$1$date[0]/g;
		print(OUT $_);
	}
	close(OUT);
	close(IN);
	chmod(oct($date[2]), "$filename.new");
	system("mv", "-f", "$filename.new", "$filename");
	system("touch", "$filename", "-t", $date[1]);
}

sub update_spec_file($)
{
	my ($filename) = @_;
	my @date = get_file_info($filename);
	local *IN;
	local *OUT;

	open(IN, "<$filename") || die ("Error: cannot open $filename\n");
	open(OUT, ">$filename.new") ||
		die("Error: cannot create $filename.new\n");
	while () {
		s/^(Version:\s*)\d+\.\d+.*$/$1$version/;
		s/^(Release:\s*).*$/$1$release/;
		print(OUT $_);
	}
	close(OUT);
	close(IN);
	system("mv", "-f", "$filename.new", "$filename");
	system("touch", "$filename", "-t", $date[1]);
}

sub write_version_file($)
{
	my ($filename) = @_;
	my $fd;

	open($fd, ">", $filename) or die("Error: cannot write $filename: $!\n");
	print($fd "VERSION=$version\n");
	print($fd "RELEASE=$release\n");
	print($fd "FULL=$full\n");
	close($fd);
}


	    


	    

	      $func_code
	      $count_code
	    
$name
              $count