css-parser-1.0.4/ 0000755 0001750 0001750 00000000000 13413156014 014046 5 ustar kovid kovid 0000000 0000000 css-parser-1.0.4/PKG-INFO 0000644 0001750 0001750 00000003477 13413156014 015156 0 ustar kovid kovid 0000000 0000000 Metadata-Version: 1.1
Name: css-parser
Version: 1.0.4
Summary: A CSS Cascading Style Sheets library for Python
Home-page: https://github.com/ebook-utils/css-parser
Author: Various People
Author-email: redacted@anonymous.net
License: LGPL 2.1 or later
Description:
css-parser
================
[](https://travis-ci.org/ebook-utils/css-parser)
[](https://ci.appveyor.com/project/kovidgoyal/css-parser)
A fork of the cssutils project based on version 1.0.2.
This fork includes general bug fixes and extensions
specific to editing and working with ebooks.
The main python source code has been modified so
that it will run without further conversion on both
Python >= 2.7 and Python 3.X without any further
modules required. All required modifications
are handled local to each file
For more information on usage, please see the cssutils
documentation
Keywords: CSS,Cascading Style Sheets,CSSParser,DOM Level 2 Stylesheets,DOM Level 2 CSS
Platform: UNKNOWN
Classifier: Development Status :: 5 - Production/Stable
Classifier: Environment :: Web Environment
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: GNU Library or Lesser General Public License (LGPL)
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python
Classifier: Programming Language :: Python :: 2
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Internet
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Classifier: Topic :: Text Processing :: Markup :: HTML
css-parser-1.0.4/README.md 0000644 0001750 0001750 00000001351 13413152352 015326 0 ustar kovid kovid 0000000 0000000 css-parser
================
[](https://travis-ci.org/ebook-utils/css-parser)
[](https://ci.appveyor.com/project/kovidgoyal/css-parser)
A fork of the cssutils project based on version 1.0.2.
This fork includes general bug fixes and extensions
specific to editing and working with ebooks.
The main python source code has been modified so
that it will run without further conversion on both
Python >= 2.7 and Python 3.X without any further
modules required. All required modifications
are handled local to each file
For more information on usage, please see the cssutils
documentation
css-parser-1.0.4/setup.cfg 0000644 0001750 0001750 00000000144 13413156014 015666 0 ustar kovid kovid 0000000 0000000 [flake8]
max-line-length = 120
builtins = unicode,basestring
[egg_info]
tag_build =
tag_date = 0
css-parser-1.0.4/setup.py 0000755 0001750 0001750 00000005021 13413155230 015560 0 ustar kovid kovid 0000000 0000000 #!/usr/bin/env python
# -*- coding: utf-8 -*-
"""
css_parser setup
use EasyInstall or install with
>python setup.py install
"""
__docformat__ = 'restructuredtext'
__author__ = 'Christof Hoeke with contributions by Walter Doerwald and lots of other people'
__date__ = '$LastChangedDate:: $:'
import ast
import re
import sys
import os
from setuptools import find_packages, setup
from setuptools.command.test import test
# extract the version without importing the module
VERSION = open('src/css_parser/version.py', 'rb').read().decode('utf-8')
VERSION = '.'.join(map(str, ast.literal_eval(re.search(r'^version\s+=\s+(.+)', VERSION, flags=re.M).group(1))))
long_description = '\n' + open('README.md', 'rb').read().decode('utf-8') + '\n' # + read('CHANGELOG.txt')
class Test(test):
user_options = [
('which-test=', 'w', "Specify which test to run as either"
" the test method name (without the leading test_)"
" or a module name with a trailing period"),
]
def initialize_options(self):
self.which_test = None
def finalize_options(self):
pass
def run(self):
import importlib
orig = sys.path[:]
try:
sys.path.insert(0, os.getcwd())
m = importlib.import_module('run_tests')
which_test = (self.which_test,) if self.which_test else ()
m.run_tests(which_test)
finally:
sys.path = orig
setup(
name='css-parser',
version=VERSION,
package_dir={'': 'src'},
packages=find_packages('src'),
description='A CSS Cascading Style Sheets library for Python',
long_description=long_description,
cmdclass={'test': Test},
author='Various People',
author_email='redacted@anonymous.net',
url='https://github.com/ebook-utils/css-parser',
license='LGPL 2.1 or later',
keywords='CSS, Cascading Style Sheets, CSSParser, DOM Level 2 Stylesheets, DOM Level 2 CSS',
classifiers=[
'Development Status :: 5 - Production/Stable',
'Environment :: Web Environment',
'Intended Audience :: Developers',
'License :: OSI Approved :: GNU Library or Lesser General Public License (LGPL)',
'Operating System :: OS Independent',
'Programming Language :: Python',
'Programming Language :: Python :: 2',
'Programming Language :: Python :: 3',
'Topic :: Internet',
'Topic :: Software Development :: Libraries :: Python Modules',
'Topic :: Text Processing :: Markup :: HTML'
]
)
css-parser-1.0.4/src/ 0000755 0001750 0001750 00000000000 13413156014 014635 5 ustar kovid kovid 0000000 0000000 css-parser-1.0.4/src/css_parser/ 0000755 0001750 0001750 00000000000 13413156014 017001 5 ustar kovid kovid 0000000 0000000 css-parser-1.0.4/src/css_parser/__init__.py 0000644 0001750 0001750 00000037337 13407702010 021123 0 ustar kovid kovid 0000000 0000000 #!/usr/bin/env python
from __future__ import unicode_literals, division, absolute_import, print_function
from .profiles import Profiles
from .serialize import CSSSerializer
from .parse import CSSParser
from . import css
from . import errorhandler
from .version import VERSION
import xml.dom
import os.path
import sys
"""css_parser - CSS Cascading Style Sheets library for Python
Copyright (C) 2004-2013 Christof Hoeke
css_parser is free software: you can redistribute it and/or modify
it under the terms of the GNU Lesser General Public License as published by
the Free Software Foundation, either version 3 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 Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License
along with this program. If not, see .
A Python package to parse and build CSS Cascading Style Sheets. DOM only, not
any rendering facilities!
Based upon and partly implementing the following specifications :
`CSS 2.1 `__
General CSS rules and properties are defined here
`CSS 2.1 Errata `__
A few errata, mainly the definition of CHARSET_SYM tokens
`CSS3 Module: Syntax `__
Used in parts since css_parser 0.9.4. css_parser tries to use the features from
CSS 2.1 and CSS 3 with preference to CSS3 but as this is not final yet some
parts are from CSS 2.1
`MediaQueries `__
MediaQueries are part of ``stylesheets.MediaList`` since v0.9.4, used in
@import and @media rules.
`Namespaces `__
Added in v0.9.1, updated to definition in CSSOM in v0.9.4, updated in 0.9.5
for dev version
`CSS3 Module: Pages Media `__
Most properties of this spec are implemented including MarginRules
`Selectors `__
The selector syntax defined here (and not in CSS 2.1) should be parsable
with css_parser (*should* mind though ;) )
`DOM Level 2 Style CSS `__
DOM for package css. 0.9.8 removes support for CSSValue and related API,
see PropertyValue and Value API for now
`DOM Level 2 Style Stylesheets `__
DOM for package stylesheets
`CSSOM `__
A few details (mainly the NamespaceRule DOM) is taken from here. Plan is
to move implementation to the stuff defined here which is newer but still
no REC so might change anytime...
The css_parser tokenizer is a customized implementation of `CSS3 Module: Syntax
(W3C Working Draft 13 August 2003) `__ which
itself is based on the CSS 2.1 tokenizer. It tries to be as compliant as
possible but uses some (helpful) parts of the CSS 2.1 tokenizer.
I guess css_parser is neither CSS 2.1 nor CSS 3 compliant but tries to at least
be able to parse both grammars including some more real world cases (some CSS
hacks are actually parsed and serialized). Both official grammars are not final
nor bugfree but still feasible. css_parser aim is not to be fully compliant to
any CSS specification (the specifications seem to be in a constant flow anyway)
but css_parser *should* be able to read and write as many as possible CSS
stylesheets "in the wild" while at the same time implement the official APIs
which are well documented. Some minor extensions are provided as well.
Please visit http://cthedot.de/css_parser/ for more details.
Tested with Python 2.7.6 and 3.3.3 on Windows 8.1 64bit.
This library may be used ``from css_parser import *`` which
import subpackages ``css`` and ``stylesheets``, CSSParser and
CSSSerializer classes only.
Usage may be::
>>> from css_parser import *
>>> parser = CSSParser()
>>> sheet = parser.parseString(u'a { color: red}')
>>> print sheet.cssText
a {
color: red
}
"""
__all__ = list(map(str, ('css', 'stylesheets', 'CSSParser', 'CSSSerializer')))
__docformat__ = 'restructuredtext'
__author__ = 'Christof Hoeke with contributions by Walter Doerwald'
__date__ = '$LastChangedDate:: $:'
if sys.version_info[0] >= 3:
text_type = str
from urllib.parse import urlsplit as urllib_urlsplit
from urllib.request import pathname2url as urllib_pathname2url
else:
text_type = unicode
from urlparse import urlsplit as urllib_urlsplit
from urllib import pathname2url as urllib_pathname2url
if sys.version_info < (2, 6):
bytes = str
# order of imports is important (partly circular)
log = errorhandler.ErrorHandler()
VERSION
ser = CSSSerializer()
profile = Profiles(log=log)
# used by Selector defining namespace prefix '*'
_ANYNS = -1
class DOMImplementationCSS(object):
"""This interface allows the DOM user to create a CSSStyleSheet
outside the context of a document. There is no way to associate
the new CSSStyleSheet with a document in DOM Level 2.
This class is its *own factory*, as it is given to
xml.dom.registerDOMImplementation which simply calls it and receives
an instance of this class then.
"""
_features = [
('css', '1.0'),
('css', '2.0'),
('stylesheets', '1.0'),
('stylesheets', '2.0')
]
def createCSSStyleSheet(self, title, media):
"""
Creates a new CSSStyleSheet.
title of type DOMString
The advisory title. See also the Style Sheet Interfaces
section.
media of type DOMString
The comma-separated list of media associated with the new style
sheet. See also the Style Sheet Interfaces section.
returns
CSSStyleSheet: A new CSS style sheet.
TODO: DOMException
SYNTAX_ERR: Raised if the specified media string value has a
syntax error and is unparsable.
"""
import warnings
warning = ("Deprecated, see "
"https://bitbucket.org/cthedot/css_parser/issues/69#comment-30669799")
warnings.warn(warning, DeprecationWarning)
return css.CSSStyleSheet(title=title, media=media)
def createDocument(self, *args, **kwargs):
# sometimes css_parser is picked automatically for
# xml.dom.getDOMImplementation, so we should provide an implementation
# see https://bitbucket.org/cthedot/css_parser/issues/69
import xml.dom.minidom as minidom
return minidom.DOMImplementation().createDocument(*args, **kwargs)
def createDocumentType(self, *args, **kwargs):
# sometimes css_parser is picked automatically for
# xml.dom.getDOMImplementation, so we should provide an implementation
# see https://bitbucket.org/cthedot/css_parser/issues/69
import xml.dom.minidom as minidom
return minidom.DOMImplementation().createDocumentType(*args, **kwargs)
def hasFeature(self, feature, version):
return (feature.lower(), text_type(version)) in self._features
xml.dom.registerDOMImplementation('css_parser', DOMImplementationCSS)
def parseString(*a, **k):
return CSSParser().parseString(*a, **k)
parseString.__doc__ = CSSParser.parseString.__doc__
def parseFile(*a, **k):
return CSSParser().parseFile(*a, **k)
parseFile.__doc__ = CSSParser.parseFile.__doc__
def parseUrl(*a, **k):
return CSSParser().parseUrl(*a, **k)
parseUrl.__doc__ = CSSParser.parseUrl.__doc__
def parseStyle(*a, **k):
return CSSParser().parseStyle(*a, **k)
parseStyle.__doc__ = CSSParser.parseStyle.__doc__
# set "ser", default serializer
def setSerializer(serializer):
"""Set the global serializer used by all class in css_parser."""
global ser
ser = serializer
def getUrls(sheet):
"""Retrieve all ``url(urlstring)`` values (in e.g.
:class:`css_parser.css.CSSImportRule` or :class:`css_parser.css.CSSValue`
objects of given `sheet`.
:param sheet:
:class:`css_parser.css.CSSStyleSheet` object whose URLs are yielded
This function is a generator. The generated URL values exclude ``url(`` and
``)`` and surrounding single or double quotes.
"""
for importrule in (r for r in sheet if r.type == r.IMPORT_RULE):
yield importrule.href
def styleDeclarations(base):
"recursive generator to find all CSSStyleDeclarations"
if hasattr(base, 'cssRules'):
for rule in base.cssRules:
for s in styleDeclarations(rule):
yield s
elif hasattr(base, 'style'):
yield base.style
for style in styleDeclarations(sheet):
for p in style.getProperties(all=True):
for v in p.propertyValue:
if v.type == 'URI':
yield v.uri
def replaceUrls(sheetOrStyle, replacer, ignoreImportRules=False):
"""Replace all URLs in :class:`css_parser.css.CSSImportRule` or
:class:`css_parser.css.CSSValue` objects of given `sheetOrStyle`.
:param sheetOrStyle:
a :class:`css_parser.css.CSSStyleSheet` or a
:class:`css_parser.css.CSSStyleDeclaration` which is changed in place
:param replacer:
a function which is called with a single argument `url` which
is the current value of each url() excluding ``url(``, ``)`` and
surrounding (single or double) quotes.
:param ignoreImportRules:
if ``True`` does not call `replacer` with URLs from @import rules.
"""
if not ignoreImportRules and not isinstance(sheetOrStyle,
css.CSSStyleDeclaration):
for importrule in (r for r in sheetOrStyle if r.type == r.IMPORT_RULE):
importrule.href = replacer(importrule.href)
def styleDeclarations(base):
"recursive generator to find all CSSStyleDeclarations"
if hasattr(base, 'cssRules'):
for rule in base.cssRules:
for s in styleDeclarations(rule):
yield s
elif hasattr(base, 'style'):
yield base.style
elif isinstance(sheetOrStyle, css.CSSStyleDeclaration):
# base is a style already
yield base
for style in styleDeclarations(sheetOrStyle):
for p in style.getProperties(all=True):
for v in p.propertyValue:
if v.type == v.URI:
v.uri = replacer(v.uri)
def resolveImports(sheet, target=None):
"""Recurcively combine all rules in given `sheet` into a `target` sheet.
@import rules which use media information are tried to be wrapped into
@media rules so keeping the media information. This may not work in
all instances (if e.g. an @import rule itself contains an @import rule
with different media infos or if it contains rules which may not be
used inside an @media block like @namespace rules.). In these cases
the @import rule is kept as in the original sheet and a WARNING is issued.
:param sheet:
in this given :class:`css_parser.css.CSSStyleSheet` all import rules are
resolved and added to a resulting *flat* sheet.
:param target:
A :class:`css_parser.css.CSSStyleSheet` object which will be the
resulting *flat* sheet if given
:returns: given `target` or a new :class:`css_parser.css.CSSStyleSheet`
object
"""
if not target:
target = css.CSSStyleSheet(href=sheet.href,
media=sheet.media,
title=sheet.title)
def getReplacer(targetbase):
"Return a replacer which uses base to return adjusted URLs"
basesch, baseloc, basepath, basequery, basefrag = urllib_urlsplit(targetbase)
basepath, basepathfilename = os.path.split(basepath)
def replacer(uri):
scheme, location, path, query, fragment = urllib_urlsplit(uri)
if not scheme and not location and not path.startswith('/'):
# relative
path, filename = os.path.split(path)
combined = os.path.normpath(os.path.join(basepath, path, filename))
return urllib_pathname2url(combined)
else:
# keep anything absolute
return uri
return replacer
for rule in sheet.cssRules:
if rule.type == rule.CHARSET_RULE:
pass
elif rule.type == rule.IMPORT_RULE:
log.info('Processing @import %r' % rule.href, neverraise=True)
if rule.hrefFound:
# add all rules of @import to current sheet
target.add(css.CSSComment(cssText='/* START @import "%s" */'
% rule.href))
try:
# nested imports
importedSheet = resolveImports(rule.styleSheet)
except xml.dom.HierarchyRequestErr as e:
log.warn('@import: Cannot resolve target, keeping rule: %s'
% e, neverraise=True)
target.add(rule)
else:
# adjust relative URI references
log.info('@import: Adjusting paths for %r' % rule.href,
neverraise=True)
replaceUrls(importedSheet,
getReplacer(rule.href),
ignoreImportRules=True)
# might have to wrap rules in @media if media given
if rule.media.mediaText == 'all':
mediaproxy = None
else:
keepimport = False
for r in importedSheet:
# check if rules present which may not be
# combined with media
if r.type not in (r.COMMENT,
r.STYLE_RULE,
r.IMPORT_RULE):
keepimport = True
break
if keepimport:
log.warn('Cannot combine imported sheet with'
' given media as other rules then'
' comments or stylerules found %r,'
' keeping %r' % (r,
rule.cssText),
neverraise=True)
target.add(rule)
continue
# wrap in @media if media is not `all`
log.info('@import: Wrapping some rules in @media '
' to keep media: %s'
% rule.media.mediaText, neverraise=True)
mediaproxy = css.CSSMediaRule(rule.media.mediaText)
for r in importedSheet:
if mediaproxy:
mediaproxy.add(r)
else:
# add to top sheet directly but are difficult anyway
target.add(r)
if mediaproxy:
target.add(mediaproxy)
else:
# keep @import as it is
log.error('Cannot get referenced stylesheet %r, keeping rule'
% rule.href, neverraise=True)
target.add(rule)
else:
target.add(rule)
return target
if __name__ == '__main__':
print(__doc__)
css-parser-1.0.4/src/css_parser/_codec2.py 0000644 0001750 0001750 00000056347 13407702010 020664 0 ustar kovid kovid 0000000 0000000 #!/usr/bin/env python
"""Python codec for CSS."""
__docformat__ = 'restructuredtext'
__author__ = 'Walter Doerwald'
__version__ = '$Id: util.py 1114 2008-03-05 13:22:59Z cthedot $'
import codecs
import marshal
# We're using bits to store all possible candidate encodings (or variants, i.e.
# we have two bits for the variants of UTF-16 and two for the
# variants of UTF-32).
#
# Prefixes for various CSS encodings
# UTF-8-SIG xEF xBB xBF
# UTF-16 (LE) xFF xFE ~x00|~x00
# UTF-16 (BE) xFE xFF
# UTF-16-LE @ x00 @ x00
# UTF-16-BE x00 @
# UTF-32 (LE) xFF xFE x00 x00
# UTF-32 (BE) x00 x00 xFE xFF
# UTF-32-LE @ x00 x00 x00
# UTF-32-BE x00 x00 x00 @
# CHARSET @ c h a ...
def detectencoding_str(input, final=False):
"""
Detect the encoding of the byte string ``input``, which contains the
beginning of a CSS file. This function returns the detected encoding (or
``None`` if it hasn't got enough data), and a flag that indicates whether
that encoding has been detected explicitely or implicitely. To detect the
encoding the first few bytes are used (or if ``input`` is ASCII compatible
and starts with a charset rule the encoding name from the rule). "Explicit"
detection means that the bytes start with a BOM or a charset rule.
If the encoding can't be detected yet, ``None`` is returned as the encoding.
``final`` specifies whether more data will be available in later calls or
not. If ``final`` is true, ``detectencoding_str()`` will never return
``None`` as the encoding.
"""
# A bit for every candidate
CANDIDATE_UTF_8_SIG = 1
CANDIDATE_UTF_16_AS_LE = 2
CANDIDATE_UTF_16_AS_BE = 4
CANDIDATE_UTF_16_LE = 8
CANDIDATE_UTF_16_BE = 16
CANDIDATE_UTF_32_AS_LE = 32
CANDIDATE_UTF_32_AS_BE = 64
CANDIDATE_UTF_32_LE = 128
CANDIDATE_UTF_32_BE = 256
CANDIDATE_CHARSET = 512
candidates = 1023 # all candidates
li = len(input)
if li >= 1:
# Check first byte
c = input[0]
if c != "\xef":
candidates &= ~CANDIDATE_UTF_8_SIG
if c != "\xff":
candidates &= ~(CANDIDATE_UTF_32_AS_LE | CANDIDATE_UTF_16_AS_LE)
if c != "\xfe":
candidates &= ~CANDIDATE_UTF_16_AS_BE
if c != "@":
candidates &= ~(CANDIDATE_UTF_32_LE | CANDIDATE_UTF_16_LE | CANDIDATE_CHARSET)
if c != "\x00":
candidates &= ~(CANDIDATE_UTF_32_AS_BE | CANDIDATE_UTF_32_BE | CANDIDATE_UTF_16_BE)
if li >= 2:
# Check second byte
c = input[1]
if c != "\xbb":
candidates &= ~CANDIDATE_UTF_8_SIG
if c != "\xfe":
candidates &= ~(CANDIDATE_UTF_16_AS_LE | CANDIDATE_UTF_32_AS_LE)
if c != "\xff":
candidates &= ~CANDIDATE_UTF_16_AS_BE
if c != "\x00":
candidates &= ~(CANDIDATE_UTF_16_LE | CANDIDATE_UTF_32_AS_BE |
CANDIDATE_UTF_32_LE | CANDIDATE_UTF_32_BE)
if c != "@":
candidates &= ~CANDIDATE_UTF_16_BE
if c != "c":
candidates &= ~CANDIDATE_CHARSET
if li >= 3:
# Check third byte
c = input[2]
if c != "\xbf":
candidates &= ~CANDIDATE_UTF_8_SIG
if c != "c":
candidates &= ~CANDIDATE_UTF_16_LE
if c != "\x00":
candidates &= ~(CANDIDATE_UTF_32_AS_LE | CANDIDATE_UTF_32_LE | CANDIDATE_UTF_32_BE)
if c != "\xfe":
candidates &= ~CANDIDATE_UTF_32_AS_BE
if c != "h":
candidates &= ~CANDIDATE_CHARSET
if li >= 4:
# Check fourth byte
c = input[3]
if input[2:4] == "\x00\x00":
candidates &= ~CANDIDATE_UTF_16_AS_LE
if c != "\x00":
candidates &= ~(CANDIDATE_UTF_16_LE | CANDIDATE_UTF_32_AS_LE | CANDIDATE_UTF_32_LE)
if c != "\xff":
candidates &= ~CANDIDATE_UTF_32_AS_BE
if c != "@":
candidates &= ~CANDIDATE_UTF_32_BE
if c != "a":
candidates &= ~CANDIDATE_CHARSET
if candidates == 0:
return ("utf-8", False)
if not (candidates & (candidates-1)): # only one candidate remaining
if candidates == CANDIDATE_UTF_8_SIG and li >= 3:
return ("utf-8-sig", True)
elif candidates == CANDIDATE_UTF_16_AS_LE and li >= 2:
return ("utf-16", True)
elif candidates == CANDIDATE_UTF_16_AS_BE and li >= 2:
return ("utf-16", True)
elif candidates == CANDIDATE_UTF_16_LE and li >= 4:
return ("utf-16-le", False)
elif candidates == CANDIDATE_UTF_16_BE and li >= 2:
return ("utf-16-be", False)
elif candidates == CANDIDATE_UTF_32_AS_LE and li >= 4:
return ("utf-32", True)
elif candidates == CANDIDATE_UTF_32_AS_BE and li >= 4:
return ("utf-32", True)
elif candidates == CANDIDATE_UTF_32_LE and li >= 4:
return ("utf-32-le", False)
elif candidates == CANDIDATE_UTF_32_BE and li >= 4:
return ("utf-32-be", False)
elif candidates == CANDIDATE_CHARSET and li >= 4:
prefix = '@charset "'
if input[:len(prefix)] == prefix:
pos = input.find('"', len(prefix))
if pos >= 0:
return (input[len(prefix):pos], True)
# if this is the last call, and we haven't determined an encoding yet,
# we default to UTF-8
if final:
return ("utf-8", False)
return (None, False) # dont' know yet
def detectencoding_unicode(input, final=False):
"""
Detect the encoding of the unicode string ``input``, which contains the
beginning of a CSS file. The encoding is detected from the charset rule
at the beginning of ``input``. If there is no charset rule, ``"utf-8"``
will be returned.
If the encoding can't be detected yet, ``None`` is returned. ``final``
specifies whether more data will be available in later calls or not. If
``final`` is true, ``detectencoding_unicode()`` will never return ``None``.
"""
prefix = u'@charset "'
if input.startswith(prefix):
pos = input.find(u'"', len(prefix))
if pos >= 0:
return (input[len(prefix):pos], True)
elif final or not prefix.startswith(input):
# if this is the last call, and we haven't determined an encoding yet,
# (or the string definitely doesn't start with prefix) we default to UTF-8
return ("utf-8", False)
return (None, False) # don't know yet
def _fixencoding(input, encoding, final=False):
"""
Replace the name of the encoding in the charset rule at the beginning of
``input`` with ``encoding``. If ``input`` doesn't starts with a charset
rule, ``input`` will be returned unmodified.
If the encoding can't be found yet, ``None`` is returned. ``final``
specifies whether more data will be available in later calls or not.
If ``final`` is true, ``_fixencoding()`` will never return ``None``.
"""
prefix = u'@charset "'
if len(input) > len(prefix):
if input.startswith(prefix):
pos = input.find(u'"', len(prefix))
if pos >= 0:
if encoding.replace("_", "-").lower() == "utf-8-sig":
encoding = u"utf-8"
return prefix + encoding + input[pos:]
# we haven't seen the end of the encoding name yet => fall through
else:
return input # doesn't start with prefix, so nothing to fix
elif not prefix.startswith(input) or final:
# can't turn out to be a @charset rule later (or there is no "later")
return input
if final:
return input
return None # don't know yet
def decode(input, errors="strict", encoding=None, force=True):
if encoding is None or not force:
(_encoding, explicit) = detectencoding_str(input, True)
if _encoding == "css":
raise ValueError("css not allowed as encoding name")
if (explicit and not force) or encoding is None: # Take the encoding from the input
encoding = _encoding
(input, consumed) = codecs.getdecoder(encoding)(input, errors)
return (_fixencoding(input, unicode(encoding), True), consumed)
def encode(input, errors="strict", encoding=None):
consumed = len(input)
if encoding is None:
encoding = detectencoding_unicode(input, True)[0]
if encoding.replace("_", "-").lower() == "utf-8-sig":
input = _fixencoding(input, u"utf-8", True)
else:
input = _fixencoding(input, unicode(encoding), True)
if encoding == "css":
raise ValueError("css not allowed as encoding name")
encoder = codecs.getencoder(encoding)
return (encoder(input, errors)[0], consumed)
def _bytes2int(bytes):
# Helper: convert an 8 bit string into an ``int``.
i = 0
for byte in bytes:
i = (i << 8) + ord(byte)
return i
def _int2bytes(i):
# Helper: convert an ``int`` into an 8-bit string.
v = []
while i:
v.insert(0, chr(i & 0xff))
i >>= 8
return "".join(v)
if hasattr(codecs, "IncrementalDecoder"):
class IncrementalDecoder(codecs.IncrementalDecoder):
def __init__(self, errors="strict", encoding=None, force=True):
self.decoder = None
self.encoding = encoding
self.force = force
codecs.IncrementalDecoder.__init__(self, errors)
# Store ``errors`` somewhere else,
# because we have to hide it in a property
self._errors = errors
self.buffer = u"".encode()
self.headerfixed = False
def iterdecode(self, input):
for part in input:
result = self.decode(part, False)
if result:
yield result
result = self.decode("", True)
if result:
yield result
def decode(self, input, final=False):
# We're doing basically the same as a ``BufferedIncrementalDecoder``,
# but since the buffer is only relevant until the encoding has been
# detected (in which case the buffer of the underlying codec might
# kick in), we're implementing buffering ourselves to avoid some
# overhead.
if self.decoder is None:
input = self.buffer + input
# Do we have to detect the encoding from the input?
if self.encoding is None or not self.force:
(encoding, explicit) = detectencoding_str(input, final)
if encoding is None: # no encoding determined yet
self.buffer = input # retry the complete input on the next call
return u"" # no encoding determined yet, so no output
elif encoding == "css":
raise ValueError("css not allowed as encoding name")
if (explicit and not self.force) or self.encoding is None: # Take the encoding from the input
self.encoding = encoding
self.buffer = "" # drop buffer, as the decoder might keep its own
decoder = codecs.getincrementaldecoder(self.encoding)
self.decoder = decoder(self._errors)
if self.headerfixed:
return self.decoder.decode(input, final)
# If we haven't fixed the header yet,
# the content of ``self.buffer`` is a ``unicode`` object
output = self.buffer + self.decoder.decode(input, final)
encoding = self.encoding
if encoding.replace("_", "-").lower() == "utf-8-sig":
encoding = "utf-8"
newoutput = _fixencoding(output, unicode(encoding), final)
if newoutput is None:
# retry fixing the @charset rule (but keep the decoded stuff)
self.buffer = output
return u""
self.headerfixed = True
return newoutput
def reset(self):
codecs.IncrementalDecoder.reset(self)
self.decoder = None
self.buffer = u"".encode()
self.headerfixed = False
def _geterrors(self):
return self._errors
def _seterrors(self, errors):
# Setting ``errors`` must be done on the real decoder too
if self.decoder is not None:
self.decoder.errors = errors
self._errors = errors
errors = property(_geterrors, _seterrors)
def getstate(self):
if self.decoder is not None:
state = (self.encoding, self.buffer, self.headerfixed, True, self.decoder.getstate())
else:
state = (self.encoding, self.buffer, self.headerfixed, False, None)
return ("", _bytes2int(marshal.dumps(state)))
def setstate(self, state):
state = _int2bytes(marshal.loads(state[1])) # ignore buffered input
self.encoding = state[0]
self.buffer = state[1]
self.headerfixed = state[2]
if state[3] is not None:
self.decoder = codecs.getincrementaldecoder(self.encoding)(self._errors)
self.decoder.setstate(state[4])
else:
self.decoder = None
if hasattr(codecs, "IncrementalEncoder"):
class IncrementalEncoder(codecs.IncrementalEncoder):
def __init__(self, errors="strict", encoding=None):
self.encoder = None
self.encoding = encoding
codecs.IncrementalEncoder.__init__(self, errors)
# Store ``errors`` somewhere else,
# because we have to hide it in a property
self._errors = errors
self.buffer = u""
def iterencode(self, input):
for part in input:
result = self.encode(part, False)
if result:
yield result
result = self.encode(u"", True)
if result:
yield result
def encode(self, input, final=False):
if self.encoder is None:
input = self.buffer + input
if self.encoding is not None:
# Replace encoding in the @charset rule with the specified one
encoding = self.encoding
if encoding.replace("_", "-").lower() == "utf-8-sig":
encoding = "utf-8"
newinput = _fixencoding(input, unicode(encoding), final)
if newinput is None: # @charset rule incomplete => Retry next time
self.buffer = input
return ""
input = newinput
else:
# Use encoding from the @charset declaration
self.encoding = detectencoding_unicode(input, final)[0]
if self.encoding is not None:
if self.encoding == "css":
raise ValueError("css not allowed as encoding name")
info = codecs.lookup(self.encoding)
encoding = self.encoding
if self.encoding.replace("_", "-").lower() == "utf-8-sig":
input = _fixencoding(input, u"utf-8", True)
self.encoder = info.incrementalencoder(self._errors)
self.buffer = u""
else:
self.buffer = input
return ""
return self.encoder.encode(input, final)
def reset(self):
codecs.IncrementalEncoder.reset(self)
self.encoder = None
self.buffer = u""
def _geterrors(self):
return self._errors
def _seterrors(self, errors):
# Setting ``errors ``must be done on the real encoder too
if self.encoder is not None:
self.encoder.errors = errors
self._errors = errors
errors = property(_geterrors, _seterrors)
def getstate(self):
if self.encoder is not None:
state = (self.encoding, self.buffer, True, self.encoder.getstate())
else:
state = (self.encoding, self.buffer, False, None)
return _bytes2int(marshal.dumps(state))
def setstate(self, state):
state = _int2bytes(marshal.loads(state))
self.encoding = state[0]
self.buffer = state[1]
if state[2] is not None:
self.encoder = codecs.getincrementalencoder(self.encoding)(self._errors)
self.encoder.setstate(state[4])
else:
self.encoder = None
class StreamWriter(codecs.StreamWriter):
def __init__(self, stream, errors="strict", encoding=None, header=False):
codecs.StreamWriter.__init__(self, stream, errors)
self.streamwriter = None
self.encoding = encoding
self._errors = errors
self.buffer = u""
def encode(self, input, errors='strict'):
li = len(input)
if self.streamwriter is None:
input = self.buffer + input
li = len(input)
if self.encoding is not None:
# Replace encoding in the @charset rule with the specified one
encoding = self.encoding
if encoding.replace("_", "-").lower() == "utf-8-sig":
encoding = "utf-8"
newinput = _fixencoding(input, unicode(encoding), False)
if newinput is None: # @charset rule incomplete => Retry next time
self.buffer = input
return ("", 0)
input = newinput
else:
# Use encoding from the @charset declaration
self.encoding = detectencoding_unicode(input, False)[0]
if self.encoding is not None:
if self.encoding == "css":
raise ValueError("css not allowed as encoding name")
self.streamwriter = codecs.getwriter(self.encoding)(self.stream, self._errors)
encoding = self.encoding
if self.encoding.replace("_", "-").lower() == "utf-8-sig":
input = _fixencoding(input, u"utf-8", True)
self.buffer = u""
else:
self.buffer = input
return ("", 0)
return (self.streamwriter.encode(input, errors)[0], li)
def _geterrors(self):
return self._errors
def _seterrors(self, errors):
# Setting ``errors`` must be done on the streamwriter too
if self.streamwriter is not None:
self.streamwriter.errors = errors
self._errors = errors
errors = property(_geterrors, _seterrors)
class StreamReader(codecs.StreamReader):
def __init__(self, stream, errors="strict", encoding=None, force=True):
codecs.StreamReader.__init__(self, stream, errors)
self.streamreader = None
self.encoding = encoding
self.force = force
self._errors = errors
def decode(self, input, errors='strict'):
if self.streamreader is None:
if self.encoding is None or not self.force:
(encoding, explicit) = detectencoding_str(input, False)
if encoding is None: # no encoding determined yet
return (u"", 0) # no encoding determined yet, so no output
elif encoding == "css":
raise ValueError("css not allowed as encoding name")
if (explicit and not self.force) or self.encoding is None: # Take the encoding from the input
self.encoding = encoding
streamreader = codecs.getreader(self.encoding)
streamreader = streamreader(self.stream, self._errors)
(output, consumed) = streamreader.decode(input, errors)
encoding = self.encoding
if encoding.replace("_", "-").lower() == "utf-8-sig":
encoding = "utf-8"
newoutput = _fixencoding(output, unicode(encoding), False)
if newoutput is not None:
self.streamreader = streamreader
return (newoutput, consumed)
return (u"", 0) # we will create a new streamreader on the next call
return self.streamreader.decode(input, errors)
def _geterrors(self):
return self._errors
def _seterrors(self, errors):
# Setting ``errors`` must be done on the streamreader too
if self.streamreader is not None:
self.streamreader.errors = errors
self._errors = errors
errors = property(_geterrors, _seterrors)
if hasattr(codecs, "CodecInfo"):
# We're running on Python 2.5 or better
def search_function(name):
if name == "css":
return codecs.CodecInfo(
name="css",
encode=encode,
decode=decode,
incrementalencoder=IncrementalEncoder,
incrementaldecoder=IncrementalDecoder,
streamwriter=StreamWriter,
streamreader=StreamReader,
)
else:
# If we're running on Python 2.4, define the utf-8-sig codec here
def utf8sig_encode(input, errors='strict'):
return (codecs.BOM_UTF8 + codecs.utf_8_encode(input, errors)[0], len(input))
def utf8sig_decode(input, errors='strict'):
prefix = 0
if input[:3] == codecs.BOM_UTF8:
input = input[3:]
prefix = 3
(output, consumed) = codecs.utf_8_decode(input, errors, True)
return (output, consumed+prefix)
class UTF8SigStreamWriter(codecs.StreamWriter):
def reset(self):
codecs.StreamWriter.reset(self)
try:
del self.encode
except AttributeError:
pass
def encode(self, input, errors='strict'):
self.encode = codecs.utf_8_encode
return utf8sig_encode(input, errors)
class UTF8SigStreamReader(codecs.StreamReader):
def reset(self):
codecs.StreamReader.reset(self)
try:
del self.decode
except AttributeError:
pass
def decode(self, input, errors='strict'):
if len(input) < 3 and codecs.BOM_UTF8.startswith(input):
# not enough data to decide if this is a BOM
# => try again on the next call
return (u"", 0)
self.decode = codecs.utf_8_decode
return utf8sig_decode(input, errors)
def search_function(name):
import encodings
name = encodings.normalize_encoding(name)
if name == "css":
return (encode, decode, StreamReader, StreamWriter)
elif name == "utf_8_sig":
return (utf8sig_encode, utf8sig_decode, UTF8SigStreamReader, UTF8SigStreamWriter)
codecs.register(search_function)
# Error handler for CSS escaping
def cssescape(exc):
if not isinstance(exc, UnicodeEncodeError):
raise TypeError("don't know how to handle %r" % exc)
return (u"".join(u"\\%06x" % ord(c) for c in exc.object[exc.start:exc.end]), exc.end)
codecs.register_error("cssescape", cssescape)
css-parser-1.0.4/src/css_parser/_codec3.py 0000644 0001750 0001750 00000057535 13407702010 020665 0 ustar kovid kovid 0000000 0000000 #!/usr/bin/env python
"""Python codec for CSS."""
__docformat__ = 'restructuredtext'
__author__ = 'Walter Doerwald'
__version__ = '$Id: util.py 1114 2008-03-05 13:22:59Z cthedot $'
import codecs
import marshal
# We're using bits to store all possible candidate encodings (or variants, i.e.
# we have two bits for the variants of UTF-16 and two for the
# variants of UTF-32).
#
# Prefixes for various CSS encodings
# UTF-8-SIG xEF xBB xBF
# UTF-16 (LE) xFF xFE ~x00|~x00
# UTF-16 (BE) xFE xFF
# UTF-16-LE @ x00 @ x00
# UTF-16-BE x00 @
# UTF-32 (LE) xFF xFE x00 x00
# UTF-32 (BE) x00 x00 xFE xFF
# UTF-32-LE @ x00 x00 x00
# UTF-32-BE x00 x00 x00 @
# CHARSET @ c h a ...
def chars(bytestring):
return ''.join(chr(byte) for byte in bytestring)
def detectencoding_str(input, final=False):
"""
Detect the encoding of the byte string ``input``, which contains the
beginning of a CSS file. This function returns the detected encoding (or
``None`` if it hasn't got enough data), and a flag that indicates whether
that encoding has been detected explicitely or implicitely. To detect the
encoding the first few bytes are used (or if ``input`` is ASCII compatible
and starts with a charset rule the encoding name from the rule). "Explicit"
detection means that the bytes start with a BOM or a charset rule.
If the encoding can't be detected yet, ``None`` is returned as the encoding.
``final`` specifies whether more data will be available in later calls or
not. If ``final`` is true, ``detectencoding_str()`` will never return
``None`` as the encoding.
"""
# A bit for every candidate
CANDIDATE_UTF_8_SIG = 1
CANDIDATE_UTF_16_AS_LE = 2
CANDIDATE_UTF_16_AS_BE = 4
CANDIDATE_UTF_16_LE = 8
CANDIDATE_UTF_16_BE = 16
CANDIDATE_UTF_32_AS_LE = 32
CANDIDATE_UTF_32_AS_BE = 64
CANDIDATE_UTF_32_LE = 128
CANDIDATE_UTF_32_BE = 256
CANDIDATE_CHARSET = 512
candidates = 1023 # all candidates
# input = chars(input)
li = len(input)
if li >= 1:
# Check first byte
c = input[0]
if c != b"\xef"[0]:
candidates &= ~CANDIDATE_UTF_8_SIG
if c != b"\xff"[0]:
candidates &= ~(CANDIDATE_UTF_32_AS_LE | CANDIDATE_UTF_16_AS_LE)
if c != b"\xfe"[0]:
candidates &= ~CANDIDATE_UTF_16_AS_BE
if c != b"@"[0]:
candidates &= ~(CANDIDATE_UTF_32_LE | CANDIDATE_UTF_16_LE | CANDIDATE_CHARSET)
if c != b"\x00"[0]:
candidates &= ~(CANDIDATE_UTF_32_AS_BE | CANDIDATE_UTF_32_BE | CANDIDATE_UTF_16_BE)
if li >= 2:
# Check second byte
c = input[1]
if c != b"\xbb"[0]:
candidates &= ~CANDIDATE_UTF_8_SIG
if c != b"\xfe"[0]:
candidates &= ~(CANDIDATE_UTF_16_AS_LE | CANDIDATE_UTF_32_AS_LE)
if c != b"\xff"[0]:
candidates &= ~CANDIDATE_UTF_16_AS_BE
if c != b"\x00"[0]:
candidates &= ~(CANDIDATE_UTF_16_LE | CANDIDATE_UTF_32_AS_BE |
CANDIDATE_UTF_32_LE | CANDIDATE_UTF_32_BE)
if c != b"@"[0]:
candidates &= ~CANDIDATE_UTF_16_BE
if c != b"c"[0]:
candidates &= ~CANDIDATE_CHARSET
if li >= 3:
# Check third byte
c = input[2]
if c != b"\xbf"[0]:
candidates &= ~CANDIDATE_UTF_8_SIG
if c != b"c"[0]:
candidates &= ~CANDIDATE_UTF_16_LE
if c != b"\x00"[0]:
candidates &= ~(CANDIDATE_UTF_32_AS_LE | CANDIDATE_UTF_32_LE | CANDIDATE_UTF_32_BE)
if c != b"\xfe"[0]:
candidates &= ~CANDIDATE_UTF_32_AS_BE
if c != b"h"[0]:
candidates &= ~CANDIDATE_CHARSET
if li >= 4:
# Check fourth byte
c = input[3]
if input[2:4] == b"\x00\x00"[0:2]:
candidates &= ~CANDIDATE_UTF_16_AS_LE
if c != b"\x00"[0]:
candidates &= ~(CANDIDATE_UTF_16_LE | CANDIDATE_UTF_32_AS_LE | CANDIDATE_UTF_32_LE)
if c != b"\xff"[0]:
candidates &= ~CANDIDATE_UTF_32_AS_BE
if c != b"@"[0]:
candidates &= ~CANDIDATE_UTF_32_BE
if c != b"a"[0]:
candidates &= ~CANDIDATE_CHARSET
if candidates == 0:
return ("utf-8", False)
if not (candidates & (candidates-1)): # only one candidate remaining
if candidates == CANDIDATE_UTF_8_SIG and li >= 3:
return ("utf-8-sig", True)
elif candidates == CANDIDATE_UTF_16_AS_LE and li >= 2:
return ("utf-16", True)
elif candidates == CANDIDATE_UTF_16_AS_BE and li >= 2:
return ("utf-16", True)
elif candidates == CANDIDATE_UTF_16_LE and li >= 4:
return ("utf-16-le", False)
elif candidates == CANDIDATE_UTF_16_BE and li >= 2:
return ("utf-16-be", False)
elif candidates == CANDIDATE_UTF_32_AS_LE and li >= 4:
return ("utf-32", True)
elif candidates == CANDIDATE_UTF_32_AS_BE and li >= 4:
return ("utf-32", True)
elif candidates == CANDIDATE_UTF_32_LE and li >= 4:
return ("utf-32-le", False)
elif candidates == CANDIDATE_UTF_32_BE and li >= 4:
return ("utf-32-be", False)
elif candidates == CANDIDATE_CHARSET and li >= 4:
prefix = '@charset "'
charsinput = chars(input)
if charsinput[:len(prefix)] == prefix:
pos = charsinput.find('"', len(prefix))
if pos >= 0:
# TODO: return str and not bytes!
return (charsinput[len(prefix):pos], True)
# if this is the last call, and we haven't determined an encoding yet,
# we default to UTF-8
if final:
return ("utf-8", False)
return (None, False) # dont' know yet
def detectencoding_unicode(input, final=False):
"""
Detect the encoding of the unicode string ``input``, which contains the
beginning of a CSS file. The encoding is detected from the charset rule
at the beginning of ``input``. If there is no charset rule, ``"utf-8"``
will be returned.
If the encoding can't be detected yet, ``None`` is returned. ``final``
specifies whether more data will be available in later calls or not. If
``final`` is true, ``detectencoding_unicode()`` will never return ``None``.
"""
prefix = '@charset "'
if input.startswith(prefix):
pos = input.find('"', len(prefix))
if pos >= 0:
return (input[len(prefix):pos], True)
elif final or not prefix.startswith(input):
# if this is the last call, and we haven't determined an encoding yet,
# (or the string definitely doesn't start with prefix) we default to UTF-8
return ("utf-8", False)
return (None, False) # don't know yet
def _fixencoding(input, encoding, final=False):
"""
Replace the name of the encoding in the charset rule at the beginning of
``input`` with ``encoding``. If ``input`` doesn't starts with a charset
rule, ``input`` will be returned unmodified.
If the encoding can't be found yet, ``None`` is returned. ``final``
specifies whether more data will be available in later calls or not.
If ``final`` is true, ``_fixencoding()`` will never return ``None``.
"""
prefix = '@charset "'
if len(input) > len(prefix):
if input.startswith(prefix):
pos = input.find('"', len(prefix))
if pos >= 0:
if encoding.replace("_", "-").lower() == "utf-8-sig":
encoding = "utf-8"
return prefix + encoding + input[pos:]
# we haven't seen the end of the encoding name yet => fall through
else:
return input # doesn't start with prefix, so nothing to fix
elif not prefix.startswith(input) or final:
# can't turn out to be a @charset rule later (or there is no "later")
return input
if final:
return input
return None # don't know yet
def decode(input, errors="strict", encoding=None, force=True):
try:
# py 3 only, memory?! object to bytes
input = input.tobytes()
except AttributeError:
pass
if encoding is None or not force:
(_encoding, explicit) = detectencoding_str(input, True)
if _encoding == "css":
raise ValueError("css not allowed as encoding name")
if (explicit and not force) or encoding is None: # Take the encoding from the input
encoding = _encoding
# NEEDS: change in parse.py (str to bytes!)
(input, consumed) = codecs.getdecoder(encoding)(input, errors)
return (_fixencoding(input, str(encoding), True), consumed)
def encode(input, errors="strict", encoding=None):
consumed = len(input)
if encoding is None:
encoding = detectencoding_unicode(input, True)[0]
if encoding.replace("_", "-").lower() == "utf-8-sig":
input = _fixencoding(input, "utf-8", True)
else:
input = _fixencoding(input, str(encoding), True)
if encoding == "css":
raise ValueError("css not allowed as encoding name")
encoder = codecs.getencoder(encoding)
return (encoder(input, errors)[0], consumed)
def _bytes2int(bytes):
# Helper: convert an 8 bit string into an ``int``.
i = 0
for byte in bytes:
i = (i << 8) + ord(byte)
return i
def _int2bytes(i):
# Helper: convert an ``int`` into an 8-bit string.
v = []
while i:
v.insert(0, chr(i & 0xff))
i >>= 8
return "".join(v)
if hasattr(codecs, "IncrementalDecoder"):
class IncrementalDecoder(codecs.IncrementalDecoder):
def __init__(self, errors="strict", encoding=None, force=True):
self.decoder = None
self.encoding = encoding
self.force = force
codecs.IncrementalDecoder.__init__(self, errors)
# Store ``errors`` somewhere else,
# because we have to hide it in a property
self._errors = errors
self.buffer = b""
self.headerfixed = False
def iterdecode(self, input):
for part in input:
result = self.decode(part, False)
if result:
yield result
result = self.decode("", True)
if result:
yield result
def decode(self, input, final=False):
# We're doing basically the same as a ``BufferedIncrementalDecoder``,
# but since the buffer is only relevant until the encoding has been
# detected (in which case the buffer of the underlying codec might
# kick in), we're implementing buffering ourselves to avoid some
# overhead.
if self.decoder is None:
input = self.buffer + input
# Do we have to detect the encoding from the input?
if self.encoding is None or not self.force:
(encoding, explicit) = detectencoding_str(input, final)
if encoding is None: # no encoding determined yet
self.buffer = input # retry the complete input on the next call
return "" # no encoding determined yet, so no output
elif encoding == "css":
raise ValueError("css not allowed as encoding name")
if (explicit and not self.force) or self.encoding is None: # Take the encoding from the input
self.encoding = encoding
self.buffer = "" # drop buffer, as the decoder might keep its own
decoder = codecs.getincrementaldecoder(self.encoding)
self.decoder = decoder(self._errors)
if self.headerfixed:
return self.decoder.decode(input, final)
# If we haven't fixed the header yet,
# the content of ``self.buffer`` is a ``unicode`` object
output = self.buffer + self.decoder.decode(input, final)
encoding = self.encoding
if encoding.replace("_", "-").lower() == "utf-8-sig":
encoding = "utf-8"
newoutput = _fixencoding(output, str(encoding), final)
if newoutput is None:
# retry fixing the @charset rule (but keep the decoded stuff)
self.buffer = output
return ""
self.headerfixed = True
return newoutput
def reset(self):
codecs.IncrementalDecoder.reset(self)
self.decoder = None
self.buffer = b""
self.headerfixed = False
def _geterrors(self):
return self._errors
def _seterrors(self, errors):
# Setting ``errors`` must be done on the real decoder too
if self.decoder is not None:
self.decoder.errors = errors
self._errors = errors
errors = property(_geterrors, _seterrors)
def getstate(self):
if self.decoder is not None:
state = (self.encoding, self.buffer, self.headerfixed, True, self.decoder.getstate())
else:
state = (self.encoding, self.buffer, self.headerfixed, False, None)
return ("", _bytes2int(marshal.dumps(state)))
def setstate(self, state):
state = _int2bytes(marshal.loads(state[1])) # ignore buffered input
self.encoding = state[0]
self.buffer = state[1]
self.headerfixed = state[2]
if state[3] is not None:
self.decoder = codecs.getincrementaldecoder(self.encoding)(self._errors)
self.decoder.setstate(state[4])
else:
self.decoder = None
if hasattr(codecs, "IncrementalEncoder"):
class IncrementalEncoder(codecs.IncrementalEncoder):
def __init__(self, errors="strict", encoding=None):
self.encoder = None
self.encoding = encoding
codecs.IncrementalEncoder.__init__(self, errors)
# Store ``errors`` somewhere else,
# because we have to hide it in a property
self._errors = errors
self.buffer = ""
def iterencode(self, input):
for part in input:
result = self.encode(part, False)
if result:
yield result
result = self.encode("", True)
if result:
yield result
def encode(self, input, final=False):
if self.encoder is None:
input = self.buffer + input
if self.encoding is not None:
# Replace encoding in the @charset rule with the specified one
encoding = self.encoding
if encoding.replace("_", "-").lower() == "utf-8-sig":
encoding = "utf-8"
newinput = _fixencoding(input, str(encoding), final)
if newinput is None: # @charset rule incomplete => Retry next time
self.buffer = input
return ""
input = newinput
else:
# Use encoding from the @charset declaration
self.encoding = detectencoding_unicode(input, final)[0]
if self.encoding is not None:
if self.encoding == "css":
raise ValueError("css not allowed as encoding name")
info = codecs.lookup(self.encoding)
encoding = self.encoding
if self.encoding.replace("_", "-").lower() == "utf-8-sig":
input = _fixencoding(input, "utf-8", True)
self.encoder = info.incrementalencoder(self._errors)
self.buffer = ""
else:
self.buffer = input
return ""
return self.encoder.encode(input, final)
def reset(self):
codecs.IncrementalEncoder.reset(self)
self.encoder = None
self.buffer = ""
def _geterrors(self):
return self._errors
def _seterrors(self, errors):
# Setting ``errors ``must be done on the real encoder too
if self.encoder is not None:
self.encoder.errors = errors
self._errors = errors
errors = property(_geterrors, _seterrors)
def getstate(self):
if self.encoder is not None:
state = (self.encoding, self.buffer, True, self.encoder.getstate())
else:
state = (self.encoding, self.buffer, False, None)
return _bytes2int(marshal.dumps(state))
def setstate(self, state):
state = _int2bytes(marshal.loads(state))
self.encoding = state[0]
self.buffer = state[1]
if state[2] is not None:
self.encoder = codecs.getincrementalencoder(self.encoding)(self._errors)
self.encoder.setstate(state[4])
else:
self.encoder = None
class StreamWriter(codecs.StreamWriter):
def __init__(self, stream, errors="strict", encoding=None, header=False):
codecs.StreamWriter.__init__(self, stream, errors)
self.streamwriter = None
self.encoding = encoding
self._errors = errors
self.buffer = ""
def encode(self, input, errors='strict'):
li = len(input)
if self.streamwriter is None:
input = self.buffer + input
li = len(input)
if self.encoding is not None:
# Replace encoding in the @charset rule with the specified one
encoding = self.encoding
if encoding.replace("_", "-").lower() == "utf-8-sig":
encoding = "utf-8"
newinput = _fixencoding(input, str(encoding), False)
if newinput is None: # @charset rule incomplete => Retry next time
self.buffer = input
return ("", 0)
input = newinput
else:
# Use encoding from the @charset declaration
self.encoding = detectencoding_unicode(input, False)[0]
if self.encoding is not None:
if self.encoding == "css":
raise ValueError("css not allowed as encoding name")
self.streamwriter = codecs.getwriter(self.encoding)(self.stream, self._errors)
encoding = self.encoding
if self.encoding.replace("_", "-").lower() == "utf-8-sig":
input = _fixencoding(input, "utf-8", True)
self.buffer = ""
else:
self.buffer = input
return ("", 0)
return (self.streamwriter.encode(input, errors)[0], li)
def _geterrors(self):
return self._errors
def _seterrors(self, errors):
# Setting ``errors`` must be done on the streamwriter too
try:
if self.streamwriter is not None:
self.streamwriter.errors = errors
except AttributeError:
# TODO: py3 only exception?
pass
self._errors = errors
errors = property(_geterrors, _seterrors)
class StreamReader(codecs.StreamReader):
def __init__(self, stream, errors="strict", encoding=None, force=True):
codecs.StreamReader.__init__(self, stream, errors)
self.streamreader = None
self.encoding = encoding
self.force = force
self._errors = errors
def decode(self, input, errors='strict'):
if self.streamreader is None:
if self.encoding is None or not self.force:
(encoding, explicit) = detectencoding_str(input, False)
if encoding is None: # no encoding determined yet
return ("", 0) # no encoding determined yet, so no output
elif encoding == "css":
raise ValueError("css not allowed as encoding name")
if (explicit and not self.force) or self.encoding is None: # Take the encoding from the input
self.encoding = encoding
streamreader = codecs.getreader(self.encoding)
streamreader = streamreader(self.stream, self._errors)
(output, consumed) = streamreader.decode(input, errors)
encoding = self.encoding
if encoding.replace("_", "-").lower() == "utf-8-sig":
encoding = "utf-8"
newoutput = _fixencoding(output, str(encoding), False)
if newoutput is not None:
self.streamreader = streamreader
return (newoutput, consumed)
return ("", 0) # we will create a new streamreader on the next call
return self.streamreader.decode(input, errors)
def _geterrors(self):
return self._errors
def _seterrors(self, errors):
# Setting ``errors`` must be done on the streamreader too
try:
if self.streamreader is not None:
self.streamreader.errors = errors
except AttributeError:
# TODO: py3 only exception?
pass
self._errors = errors
errors = property(_geterrors, _seterrors)
if hasattr(codecs, "CodecInfo"):
# We're running on Python 2.5 or better
def search_function(name):
if name == "css":
return codecs.CodecInfo(
name="css",
encode=encode,
decode=decode,
incrementalencoder=IncrementalEncoder,
incrementaldecoder=IncrementalDecoder,
streamwriter=StreamWriter,
streamreader=StreamReader,
)
else:
# If we're running on Python 2.4, define the utf-8-sig codec here
def utf8sig_encode(input, errors='strict'):
return (codecs.BOM_UTF8 + codecs.utf_8_encode(input, errors)[0], len(input))
def utf8sig_decode(input, errors='strict'):
prefix = 0
if input[:3] == codecs.BOM_UTF8:
input = input[3:]
prefix = 3
(output, consumed) = codecs.utf_8_decode(input, errors, True)
return (output, consumed+prefix)
class UTF8SigStreamWriter(codecs.StreamWriter):
def reset(self):
codecs.StreamWriter.reset(self)
try:
del self.encode
except AttributeError:
pass
def encode(self, input, errors='strict'):
self.encode = codecs.utf_8_encode
return utf8sig_encode(input, errors)
class UTF8SigStreamReader(codecs.StreamReader):
def reset(self):
codecs.StreamReader.reset(self)
try:
del self.decode
except AttributeError:
pass
def decode(self, input, errors='strict'):
if len(input) < 3 and codecs.BOM_UTF8.startswith(input):
# not enough data to decide if this is a BOM
# => try again on the next call
return ("", 0)
self.decode = codecs.utf_8_decode
return utf8sig_decode(input, errors)
def search_function(name):
import encodings
name = encodings.normalize_encoding(name)
if name == "css":
return (encode, decode, StreamReader, StreamWriter)
elif name == "utf_8_sig":
return (utf8sig_encode, utf8sig_decode, UTF8SigStreamReader, UTF8SigStreamWriter)
codecs.register(search_function)
# Error handler for CSS escaping
def cssescape(exc):
if not isinstance(exc, UnicodeEncodeError):
raise TypeError("don't know how to handle %r" % exc)
return ("".join("\\%06x" % ord(c) for c in exc.object[exc.start:exc.end]), exc.end)
codecs.register_error("cssescape", cssescape)
css-parser-1.0.4/src/css_parser/_fetch.py 0000644 0001750 0001750 00000004305 13407702566 020621 0 ustar kovid kovid 0000000 0000000 from __future__ import unicode_literals, division, absolute_import, print_function
from . import errorhandler
import css_parser.encutils as encutils
from css_parser.version import VERSION
"""Default URL reading functions"""
__all__ = ['_defaultFetcher']
__docformat__ = 'restructuredtext'
__version__ = '$Id: tokenize2.py 1547 2008-12-10 20:42:26Z cthedot $'
import sys
if sys.version_info[0] >= 3:
from urllib.request import urlopen as urllib_urlopen
from urllib.request import Request as urllib_Request
from urllib.error import HTTPError as urllib_HTTPError
from urllib.error import URLError as urllib_URLError
else:
from urllib2 import urlopen as urllib_urlopen
from urllib2 import Request as urllib_Request
from urllib2 import HTTPError as urllib_HTTPError
from urllib2 import URLError as urllib_URLError
log = errorhandler.ErrorHandler()
def _defaultFetcher(url):
"""Retrieve data from ``url``. css_parser default implementation of fetch
URL function.
Returns ``(encoding, string)`` or ``None``
"""
try:
request = urllib_Request(url)
request.add_header('User-agent',
'css_parser %s (http://www.cthedot.de/css_parser/)' % VERSION)
res = urllib_urlopen(request)
except urllib_HTTPError as e:
# http error, e.g. 404, e can be raised
log.warn('HTTPError opening url=%s: %s %s' %
(url, e.code, e.msg), error=e)
except urllib_URLError as e:
# URLError like mailto: or other IO errors, e can be raised
log.warn('URLError, %s' % e.reason, error=e)
except OSError as e:
# e.g if file URL and not found
log.warn(e, error=OSError)
except ValueError as e:
# invalid url, e.g. "1"
log.warn('ValueError, %s' % e.args[0], error=ValueError)
else:
if res:
mimeType, encoding = encutils.getHTTPInfo(res)
if mimeType != 'text/css':
log.error('Expected "text/css" mime type for url=%r but found: %r' %
(url, mimeType), error=ValueError)
content = res.read()
if hasattr(res, 'close'):
res.close()
return encoding, content
css-parser-1.0.4/src/css_parser/_fetchgae.py 0000644 0001750 0001750 00000005162 13407702010 021260 0 ustar kovid kovid 0000000 0000000 from __future__ import unicode_literals, division, absolute_import, print_function
from . import errorhandler
import cgi
from google.appengine.api import urlfetch
"""GAE specific URL reading functions"""
import sys
PY3 = sys.version_info[0] >= 3
__all__ = ['_defaultFetcher']
__docformat__ = 'restructuredtext'
__version__ = '$Id: tokenize2.py 1547 2008-12-10 20:42:26Z cthedot $'
# raises ImportError of not on GAE
log = errorhandler.ErrorHandler()
def _defaultFetcher(url):
"""
uses GoogleAppEngine (GAE)
fetch(url, payload=None, method=GET, headers={}, allow_truncated=False)
Response
content
The body content of the response.
content_was_truncated
True if the allow_truncated parameter to fetch() was True and
the response exceeded the maximum response size. In this case,
the content attribute contains the truncated response.
status_code
The HTTP status code.
headers
The HTTP response headers, as a mapping of names to values.
Exceptions
exception InvalidURLError()
The URL of the request was not a valid URL, or it used an
unsupported method. Only http and https URLs are supported.
exception DownloadError()
There was an error retrieving the data.
This exception is not raised if the server returns an HTTP
error code: In that case, the response data comes back intact,
including the error code.
exception ResponseTooLargeError()
The response data exceeded the maximum allowed size, and the
allow_truncated parameter passed to fetch() was False.
"""
# from google.appengine.api import urlfetch
try:
r = urlfetch.fetch(url, method=urlfetch.GET)
except urlfetch.Error as e:
log.warn('Error opening url=%r: %s' % (url, e),
error=IOError)
else:
if r.status_code == 200:
# find mimetype and encoding
mimetype = 'application/octet-stream'
try:
mimetype, params = cgi.parse_header(r.headers['content-type'])
encoding = params['charset']
except KeyError:
encoding = None
if mimetype != 'text/css':
log.error('Expected "text/css" mime type for url %r but found: %r' %
(url, mimetype), error=ValueError)
return encoding, r.content
else:
# TODO: 301 etc
log.warn('Error opening url=%r: HTTP status %s' %
(url, r.status_code), error=IOError)
css-parser-1.0.4/src/css_parser/codec.py 0000644 0001750 0001750 00000000717 13407702010 020431 0 ustar kovid kovid 0000000 0000000 #!/usr/bin/env python
"""Python codec for CSS."""
from __future__ import absolute_import
__docformat__ = 'restructuredtext'
__author__ = 'Walter Doerwald'
__version__ = '$Id: util.py 1114 2008-03-05 13:22:59Z cthedot $'
import sys
if sys.version_info < (3,):
from ._codec2 import * # noqa
# for tests
from ._codec2 import _fixencoding # noqa
else:
from ._codec3 import * # noqa
# for tests
from ._codec3 import _fixencoding # noqa
css-parser-1.0.4/src/css_parser/css/ 0000755 0001750 0001750 00000000000 13413156014 017571 5 ustar kovid kovid 0000000 0000000 css-parser-1.0.4/src/css_parser/css/__init__.py 0000644 0001750 0001750 00000004265 13407702010 021705 0 ustar kovid kovid 0000000 0000000 """Implements Document Object Model Level 2 CSS
http://www.w3.org/TR/2000/PR-DOM-Level-2-Style-20000927/css.html
currently implemented
- CSSStyleSheet
- CSSRuleList
- CSSRule
- CSSComment (css_parser addon)
- CSSCharsetRule
- CSSFontFaceRule
- CSSImportRule
- CSSMediaRule
- CSSNamespaceRule (WD)
- CSSPageRule
- CSSStyleRule
- CSSUnkownRule
- Selector and SelectorList
- CSSStyleDeclaration
- CSS2Properties
- CSSValue
- CSSPrimitiveValue
- CSSValueList
- CSSVariablesRule
- CSSVariablesDeclaration
todo
- RGBColor, Rect, Counter
"""
from __future__ import division, absolute_import, print_function
__all__ = [
'CSSStyleSheet',
'CSSRuleList',
'CSSRule',
'CSSComment',
'CSSCharsetRule',
'CSSFontFaceRule',
'CSSImportRule',
'CSSMediaRule',
'CSSNamespaceRule',
'CSSPageRule',
'MarginRule',
'CSSStyleRule',
'CSSUnknownRule',
'CSSVariablesRule',
'CSSVariablesDeclaration',
'Selector', 'SelectorList',
'CSSStyleDeclaration', 'Property',
# 'CSSValue', 'CSSPrimitiveValue', 'CSSValueList'
'PropertyValue',
'Value',
'ColorValue',
'DimensionValue',
'URIValue',
'CSSFunction',
'CSSVariable',
'MSValue',
]
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
from .cssstylesheet import CSSStyleSheet
from .cssrulelist import CSSRuleList
from .cssrule import CSSRule
from .csscomment import CSSComment
from .csscharsetrule import CSSCharsetRule
from .cssfontfacerule import CSSFontFaceRule
from .cssimportrule import CSSImportRule
from .cssmediarule import CSSMediaRule
from .cssnamespacerule import CSSNamespaceRule
from .csspagerule import CSSPageRule
from .marginrule import MarginRule
from .cssstylerule import CSSStyleRule
from .cssvariablesrule import CSSVariablesRule
from .cssunknownrule import CSSUnknownRule
from .selector import Selector
from .selectorlist import SelectorList
from .cssstyledeclaration import CSSStyleDeclaration
from .cssvariablesdeclaration import CSSVariablesDeclaration
from .property import Property
from .value import PropertyValue, Value, ColorValue, DimensionValue, URIValue, CSSFunction, CSSVariable, MSValue
css-parser-1.0.4/src/css_parser/css/colors.py 0000644 0001750 0001750 00000014650 13407702010 021446 0 ustar kovid kovid 0000000 0000000 # -*- coding: utf-8 -*-
"""
Built from something like this:
print [
(
row[2].text_content().strip(),
eval(row[4].text_content().strip())
)
for row in lxml.html.parse('http://www.w3.org/TR/css3-color/')
.xpath("//*[@class='colortable']//tr[position()>1]")
]
by Simon Sapin
"""
from __future__ import unicode_literals, division, absolute_import, print_function
COLORS = {
'transparent': (0, 0, 0, 0.0),
'black': (0, 0, 0, 1.0),
'silver': (192, 192, 192, 1.0),
'gray': (128, 128, 128, 1.0),
'white': (255, 255, 255, 1.0),
'maroon': (128, 0, 0, 1.0),
'red': (255, 0, 0, 1.0),
'purple': (128, 0, 128, 1.0),
'fuchsia': (255, 0, 255, 1.0),
'green': (0, 128, 0, 1.0),
'lime': (0, 255, 0, 1.0),
'olive': (128, 128, 0, 1.0),
'yellow': (255, 255, 0, 1.0),
'navy': (0, 0, 128, 1.0),
'blue': (0, 0, 255, 1.0),
'teal': (0, 128, 128, 1.0),
'aqua': (0, 255, 255, 1.0),
'aliceblue': (240, 248, 255, 1.0),
'antiquewhite': (250, 235, 215, 1.0),
'aqua': (0, 255, 255, 1.0),
'aquamarine': (127, 255, 212, 1.0),
'azure': (240, 255, 255, 1.0),
'beige': (245, 245, 220, 1.0),
'bisque': (255, 228, 196, 1.0),
'black': (0, 0, 0, 1.0),
'blanchedalmond': (255, 235, 205, 1.0),
'blue': (0, 0, 255, 1.0),
'blueviolet': (138, 43, 226, 1.0),
'brown': (165, 42, 42, 1.0),
'burlywood': (222, 184, 135, 1.0),
'cadetblue': (95, 158, 160, 1.0),
'chartreuse': (127, 255, 0, 1.0),
'chocolate': (210, 105, 30, 1.0),
'coral': (255, 127, 80, 1.0),
'cornflowerblue': (100, 149, 237, 1.0),
'cornsilk': (255, 248, 220, 1.0),
'crimson': (220, 20, 60, 1.0),
'cyan': (0, 255, 255, 1.0),
'darkblue': (0, 0, 139, 1.0),
'darkcyan': (0, 139, 139, 1.0),
'darkgoldenrod': (184, 134, 11, 1.0),
'darkgray': (169, 169, 169, 1.0),
'darkgreen': (0, 100, 0, 1.0),
'darkgrey': (169, 169, 169, 1.0),
'darkkhaki': (189, 183, 107, 1.0),
'darkmagenta': (139, 0, 139, 1.0),
'darkolivegreen': (85, 107, 47, 1.0),
'darkorange': (255, 140, 0, 1.0),
'darkorchid': (153, 50, 204, 1.0),
'darkred': (139, 0, 0, 1.0),
'darksalmon': (233, 150, 122, 1.0),
'darkseagreen': (143, 188, 143, 1.0),
'darkslateblue': (72, 61, 139, 1.0),
'darkslategray': (47, 79, 79, 1.0),
'darkslategrey': (47, 79, 79, 1.0),
'darkturquoise': (0, 206, 209, 1.0),
'darkviolet': (148, 0, 211, 1.0),
'deeppink': (255, 20, 147, 1.0),
'deepskyblue': (0, 191, 255, 1.0),
'dimgray': (105, 105, 105, 1.0),
'dimgrey': (105, 105, 105, 1.0),
'dodgerblue': (30, 144, 255, 1.0),
'firebrick': (178, 34, 34, 1.0),
'floralwhite': (255, 250, 240, 1.0),
'forestgreen': (34, 139, 34, 1.0),
'fuchsia': (255, 0, 255, 1.0),
'gainsboro': (220, 220, 220, 1.0),
'ghostwhite': (248, 248, 255, 1.0),
'gold': (255, 215, 0, 1.0),
'goldenrod': (218, 165, 32, 1.0),
'gray': (128, 128, 128, 1.0),
'green': (0, 128, 0, 1.0),
'greenyellow': (173, 255, 47, 1.0),
'grey': (128, 128, 128, 1.0),
'honeydew': (240, 255, 240, 1.0),
'hotpink': (255, 105, 180, 1.0),
'indianred': (205, 92, 92, 1.0),
'indigo': (75, 0, 130, 1.0),
'ivory': (255, 255, 240, 1.0),
'khaki': (240, 230, 140, 1.0),
'lavender': (230, 230, 250, 1.0),
'lavenderblush': (255, 240, 245, 1.0),
'lawngreen': (124, 252, 0, 1.0),
'lemonchiffon': (255, 250, 205, 1.0),
'lightblue': (173, 216, 230, 1.0),
'lightcoral': (240, 128, 128, 1.0),
'lightcyan': (224, 255, 255, 1.0),
'lightgoldenrodyellow': (250, 250, 210, 1.0),
'lightgray': (211, 211, 211, 1.0),
'lightgreen': (144, 238, 144, 1.0),
'lightgrey': (211, 211, 211, 1.0),
'lightpink': (255, 182, 193, 1.0),
'lightsalmon': (255, 160, 122, 1.0),
'lightseagreen': (32, 178, 170, 1.0),
'lightskyblue': (135, 206, 250, 1.0),
'lightslategray': (119, 136, 153, 1.0),
'lightslategrey': (119, 136, 153, 1.0),
'lightsteelblue': (176, 196, 222, 1.0),
'lightyellow': (255, 255, 224, 1.0),
'lime': (0, 255, 0, 1.0),
'limegreen': (50, 205, 50, 1.0),
'linen': (250, 240, 230, 1.0),
'magenta': (255, 0, 255, 1.0),
'maroon': (128, 0, 0, 1.0),
'mediumaquamarine': (102, 205, 170, 1.0),
'mediumblue': (0, 0, 205, 1.0),
'mediumorchid': (186, 85, 211, 1.0),
'mediumpurple': (147, 112, 219, 1.0),
'mediumseagreen': (60, 179, 113, 1.0),
'mediumslateblue': (123, 104, 238, 1.0),
'mediumspringgreen': (0, 250, 154, 1.0),
'mediumturquoise': (72, 209, 204, 1.0),
'mediumvioletred': (199, 21, 133, 1.0),
'midnightblue': (25, 25, 112, 1.0),
'mintcream': (245, 255, 250, 1.0),
'mistyrose': (255, 228, 225, 1.0),
'moccasin': (255, 228, 181, 1.0),
'navajowhite': (255, 222, 173, 1.0),
'navy': (0, 0, 128, 1.0),
'oldlace': (253, 245, 230, 1.0),
'olive': (128, 128, 0, 1.0),
'olivedrab': (107, 142, 35, 1.0),
'orange': (255, 165, 0, 1.0),
'orangered': (255, 69, 0, 1.0),
'orchid': (218, 112, 214, 1.0),
'palegoldenrod': (238, 232, 170, 1.0),
'palegreen': (152, 251, 152, 1.0),
'paleturquoise': (175, 238, 238, 1.0),
'palevioletred': (219, 112, 147, 1.0),
'papayawhip': (255, 239, 213, 1.0),
'peachpuff': (255, 218, 185, 1.0),
'peru': (205, 133, 63, 1.0),
'pink': (255, 192, 203, 1.0),
'plum': (221, 160, 221, 1.0),
'powderblue': (176, 224, 230, 1.0),
'purple': (128, 0, 128, 1.0),
'red': (255, 0, 0, 1.0),
'rosybrown': (188, 143, 143, 1.0),
'royalblue': (65, 105, 225, 1.0),
'saddlebrown': (139, 69, 19, 1.0),
'salmon': (250, 128, 114, 1.0),
'sandybrown': (244, 164, 96, 1.0),
'seagreen': (46, 139, 87, 1.0),
'seashell': (255, 245, 238, 1.0),
'sienna': (160, 82, 45, 1.0),
'silver': (192, 192, 192, 1.0),
'skyblue': (135, 206, 235, 1.0),
'slateblue': (106, 90, 205, 1.0),
'slategray': (112, 128, 144, 1.0),
'slategrey': (112, 128, 144, 1.0),
'snow': (255, 250, 250, 1.0),
'springgreen': (0, 255, 127, 1.0),
'steelblue': (70, 130, 180, 1.0),
'tan': (210, 180, 140, 1.0),
'teal': (0, 128, 128, 1.0),
'thistle': (216, 191, 216, 1.0),
'tomato': (255, 99, 71, 1.0),
'turquoise': (64, 224, 208, 1.0),
'violet': (238, 130, 238, 1.0),
'wheat': (245, 222, 179, 1.0),
'white': (255, 255, 255, 1.0),
'whitesmoke': (245, 245, 245, 1.0),
'yellow': (255, 255, 0, 1.0),
'yellowgreen': (154, 205, 50, 1.0),
}
css-parser-1.0.4/src/css_parser/css/csscharsetrule.py 0000644 0001750 0001750 00000013763 13407702010 023203 0 ustar kovid kovid 0000000 0000000 """CSSCharsetRule implements DOM Level 2 CSS CSSCharsetRule."""
from __future__ import unicode_literals, division, absolute_import, print_function
__all__ = ['CSSCharsetRule']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
import codecs
from . import cssrule
import css_parser
import xml.dom
class CSSCharsetRule(cssrule.CSSRule):
"""
The CSSCharsetRule interface represents an @charset rule in a CSS style
sheet. The value of the encoding attribute does not affect the encoding
of text data in the DOM objects; this encoding is always UTF-16
(also in Python?). After a stylesheet is loaded, the value of the
encoding attribute is the value found in the @charset rule. If there
was no @charset in the original document, then no CSSCharsetRule is
created. The value of the encoding attribute may also be used as a hint
for the encoding used on serialization of the style sheet.
The value of the @charset rule (and therefore of the CSSCharsetRule)
may not correspond to the encoding the document actually came in;
character encoding information e.g. in an HTTP header, has priority
(see CSS document representation) but this is not reflected in the
CSSCharsetRule.
This rule is not really needed anymore as setting
:attr:`CSSStyleSheet.encoding` is much easier.
Format::
charsetrule:
CHARSET_SYM S* STRING S* ';'
BUT: Only valid format is (single space, double quotes!)::
@charset "ENCODING";
"""
def __init__(self, encoding=None, parentRule=None,
parentStyleSheet=None, readonly=False):
"""
:param encoding:
a valid character encoding
:param readonly:
defaults to False, not used yet
"""
super(CSSCharsetRule, self).__init__(parentRule=parentRule,
parentStyleSheet=parentStyleSheet)
self._atkeyword = '@charset'
if encoding:
self.encoding = encoding
else:
self._encoding = None
self._readonly = readonly
def __repr__(self):
return "css_parser.css.%s(encoding=%r)" % (
self.__class__.__name__,
self.encoding)
def __str__(self):
return "" % (
self.__class__.__name__,
self.encoding,
id(self))
def _getCssText(self):
"""The parsable textual representation."""
return css_parser.ser.do_CSSCharsetRule(self)
def _setCssText(self, cssText):
"""
:param cssText:
A parsable DOMString.
:exceptions:
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error and
is unparsable.
- :exc:`~xml.dom.InvalidModificationErr`:
Raised if the specified CSS string value represents a different
type of rule than the current one.
- :exc:`~xml.dom.HierarchyRequestErr`:
Raised if the rule cannot be inserted at this point in the
style sheet.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if the rule is readonly.
"""
super(CSSCharsetRule, self)._setCssText(cssText)
wellformed = True
tokenizer = self._tokenize2(cssText)
if self._type(self._nexttoken(tokenizer)) != self._prods.CHARSET_SYM:
wellformed = False
self._log.error('CSSCharsetRule must start with "@charset "',
error=xml.dom.InvalidModificationErr)
encodingtoken = self._nexttoken(tokenizer)
encodingtype = self._type(encodingtoken)
encoding = self._stringtokenvalue(encodingtoken)
if self._prods.STRING != encodingtype or not encoding:
wellformed = False
self._log.error('CSSCharsetRule: no encoding found; %r.' %
self._valuestr(cssText))
semicolon = self._tokenvalue(self._nexttoken(tokenizer))
EOFtype = self._type(self._nexttoken(tokenizer))
if ';' != semicolon or EOFtype not in ('EOF', None):
wellformed = False
self._log.error('CSSCharsetRule: Syntax Error: %r.' %
self._valuestr(cssText))
if wellformed:
self.encoding = encoding
cssText = property(fget=_getCssText, fset=_setCssText,
doc="(DOM) The parsable textual representation.")
def _setEncoding(self, encoding):
"""
:param encoding:
a valid encoding to be used. Currently only valid Python encodings
are allowed.
:exceptions:
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this encoding rule is readonly.
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified encoding value has a syntax error and
is unparsable.
"""
self._checkReadonly()
tokenizer = self._tokenize2(encoding)
encodingtoken = self._nexttoken(tokenizer)
unexpected = self._nexttoken(tokenizer)
if not encodingtoken or unexpected or\
self._prods.IDENT != self._type(encodingtoken):
self._log.error('CSSCharsetRule: Syntax Error in encoding value '
'%r.' % encoding)
else:
try:
codecs.lookup(encoding)
except LookupError:
self._log.error('CSSCharsetRule: Unknown (Python) encoding %r.'
% encoding)
else:
self._encoding = encoding.lower()
encoding = property(lambda self: self._encoding, _setEncoding,
doc="(DOM)The encoding information used in this @charset rule.")
type = property(lambda self: self.CHARSET_RULE,
doc="The type of this rule, as defined by a CSSRule "
"type constant.")
wellformed = property(lambda self: bool(self.encoding))
css-parser-1.0.4/src/css_parser/css/csscomment.py 0000644 0001750 0001750 00000005576 13407702010 022327 0 ustar kovid kovid 0000000 0000000 """CSSComment is not defined in DOM Level 2 at all but a css_parser defined
class only.
Implements CSSRule which is also extended for a CSSComment rule type.
"""
from __future__ import unicode_literals, division, absolute_import, print_function
__all__ = ['CSSComment']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
from . import cssrule
import css_parser
import xml.dom
class CSSComment(cssrule.CSSRule):
"""
Represents a CSS comment (css_parser only).
Format::
/*...*/
"""
def __init__(self, cssText=None, parentRule=None,
parentStyleSheet=None, readonly=False):
super(CSSComment, self).__init__(parentRule=parentRule,
parentStyleSheet=parentStyleSheet)
self._cssText = None
if cssText:
self._setCssText(cssText)
self._readonly = readonly
def __repr__(self):
return "css_parser.css.%s(cssText=%r)" % (
self.__class__.__name__,
self.cssText)
def __str__(self):
return "" % (
self.__class__.__name__,
self.cssText,
id(self))
def _getCssText(self):
"""Return serialized property cssText."""
return css_parser.ser.do_CSSComment(self)
def _setCssText(self, cssText):
"""
:param cssText:
textual text to set or tokenlist which is not tokenized
anymore. May also be a single token for this rule
:exceptions:
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error and
is unparsable.
- :exc:`~xml.dom.InvalidModificationErr`:
Raised if the specified CSS string value represents a different
type of rule than the current one.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if the rule is readonly.
"""
super(CSSComment, self)._setCssText(cssText)
tokenizer = self._tokenize2(cssText)
commenttoken = self._nexttoken(tokenizer)
unexpected = self._nexttoken(tokenizer)
if not commenttoken or\
self._type(commenttoken) != self._prods.COMMENT or\
unexpected:
self._log.error('CSSComment: Not a CSSComment: %r' %
self._valuestr(cssText),
error=xml.dom.InvalidModificationErr)
else:
self._cssText = self._tokenvalue(commenttoken)
cssText = property(_getCssText, _setCssText,
doc="The parsable textual representation of this rule.")
type = property(lambda self: self.COMMENT,
doc="The type of this rule, as defined by a CSSRule "
"type constant.")
# constant but needed:
wellformed = property(lambda self: True)
css-parser-1.0.4/src/css_parser/css/cssfontfacerule.py 0000644 0001750 0001750 00000016011 13407702010 023324 0 ustar kovid kovid 0000000 0000000 from __future__ import unicode_literals, division, absolute_import, print_function
import xml.dom
import css_parser
from . import cssrule
from .cssstyledeclaration import CSSStyleDeclaration
"""CSSFontFaceRule implements DOM Level 2 CSS CSSFontFaceRule.
From css_parser 0.9.6 additions from CSS Fonts Module Level 3 are
added http://www.w3.org/TR/css3-fonts/.
"""
__all__ = ['CSSFontFaceRule']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
import sys
if sys.version_info[0] >= 3:
string_type = str
else:
string_type = basestring
class CSSFontFaceRule(cssrule.CSSRule):
"""
The CSSFontFaceRule interface represents a @font-face rule in a CSS
style sheet. The @font-face rule is used to hold a set of font
descriptions.
Format::
font_face
: FONT_FACE_SYM S*
'{' S* declaration [ ';' S* declaration ]* '}' S*
;
css_parser uses a :class:`~css_parser.css.CSSStyleDeclaration` to
represent the font descriptions. For validation a specific profile
is used though were some properties have other valid values than
when used in e.g. a :class:`~css_parser.css.CSSStyleRule`.
"""
def __init__(self, style=None, parentRule=None,
parentStyleSheet=None, readonly=False):
"""
If readonly allows setting of properties in constructor only.
:param style:
CSSStyleDeclaration used to hold any font descriptions
for this CSSFontFaceRule
"""
super(CSSFontFaceRule, self).__init__(parentRule=parentRule,
parentStyleSheet=parentStyleSheet)
self._atkeyword = '@font-face'
if style:
self.style = style
else:
self.style = CSSStyleDeclaration()
self._readonly = readonly
def __repr__(self):
return "css_parser.css.%s(style=%r)" % (
self.__class__.__name__,
self.style.cssText)
def __str__(self):
return "" % (
self.__class__.__name__,
self.style.cssText,
self.valid,
id(self))
def _getCssText(self):
"""Return serialized property cssText."""
return css_parser.ser.do_CSSFontFaceRule(self)
def _setCssText(self, cssText):
"""
:exceptions:
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error and
is unparsable.
- :exc:`~xml.dom.InvalidModificationErr`:
Raised if the specified CSS string value represents a different
type of rule than the current one.
- :exc:`~xml.dom.HierarchyRequestErr`:
Raised if the rule cannot be inserted at this point in the
style sheet.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if the rule is readonly.
"""
super(CSSFontFaceRule, self)._setCssText(cssText)
tokenizer = self._tokenize2(cssText)
attoken = self._nexttoken(tokenizer, None)
if self._type(attoken) != self._prods.FONT_FACE_SYM:
self._log.error('CSSFontFaceRule: No CSSFontFaceRule found: %s' %
self._valuestr(cssText),
error=xml.dom.InvalidModificationErr)
else:
newStyle = CSSStyleDeclaration(parentRule=self)
ok = True
beforetokens, brace = self._tokensupto2(tokenizer,
blockstartonly=True,
separateEnd=True)
if self._tokenvalue(brace) != '{':
ok = False
self._log.error('CSSFontFaceRule: No start { of style '
'declaration found: %r'
% self._valuestr(cssText), brace)
# parse stuff before { which should be comments and S only
new = {'wellformed': True}
newseq = self._tempSeq()
beforewellformed, expected = self._parse(expected=':',
seq=newseq, tokenizer=self._tokenize2(beforetokens),
productions={})
ok = ok and beforewellformed and new['wellformed']
styletokens, braceorEOFtoken = self._tokensupto2(tokenizer,
blockendonly=True,
separateEnd=True)
val, type_ = self._tokenvalue(braceorEOFtoken),\
self._type(braceorEOFtoken)
if val != '}' and type_ != 'EOF':
ok = False
self._log.error('CSSFontFaceRule: No "}" after style '
'declaration found: %r'
% self._valuestr(cssText))
nonetoken = self._nexttoken(tokenizer)
if nonetoken:
ok = False
self._log.error('CSSFontFaceRule: Trailing content found.',
token=nonetoken)
if 'EOF' == type_:
# add again as style needs it
styletokens.append(braceorEOFtoken)
# SET, may raise:
newStyle.cssText = styletokens
if ok:
# contains probably comments only (upto ``{``)
self._setSeq(newseq)
self.style = newStyle
cssText = property(_getCssText, _setCssText,
doc="(DOM) The parsable textual representation of this "
"rule.")
def _setStyle(self, style):
"""
:param style:
a CSSStyleDeclaration or string
"""
self._checkReadonly()
# under Python 2.X this was basestring but given unicode literals ...
if isinstance(style, string_type):
self._style = CSSStyleDeclaration(cssText=style, parentRule=self)
else:
style._parentRule = self
self._style = style
style = property(lambda self: self._style, _setStyle,
doc="(DOM) The declaration-block of this rule set, "
"a :class:`~css_parser.css.CSSStyleDeclaration`.")
type = property(lambda self: self.FONT_FACE_RULE,
doc="The type of this rule, as defined by a CSSRule "
"type constant.")
def _getValid(self):
needed = ['font-family', 'src']
for p in self.style.getProperties(all=True):
if not p.valid:
return False
try:
needed.remove(p.name)
except ValueError:
pass
return not bool(needed)
valid = property(_getValid,
doc="CSSFontFace is valid if properties `font-family` "
"and `src` are set and all properties are valid.")
# constant but needed:
wellformed = property(lambda self: True)
css-parser-1.0.4/src/css_parser/css/cssimportrule.py 0000644 0001750 0001750 00000035672 13407702010 023067 0 ustar kovid kovid 0000000 0000000 from __future__ import unicode_literals, division, absolute_import, print_function
import xml.dom
"""CSSImportRule implements DOM Level 2 CSS CSSImportRule plus the
``name`` property from http://www.w3.org/TR/css3-cascade/#cascading."""
__all__ = ['CSSImportRule']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
from . import cssrule
import css_parser
import os
import sys
if sys.version_info[0] >= 3:
string_type = str
from urllib.parse import urljoin as urllib_urljoin
else:
string_type = basestring
from urlparse import urljoin as urllib_urljoin
class CSSImportRule(cssrule.CSSRule):
"""
Represents an @import rule within a CSS style sheet. The @import rule
is used to import style rules from other style sheets.
Format::
import
: IMPORT_SYM S*
[STRING|URI] S* [ medium [ COMMA S* medium]* ]? S* STRING? S* ';' S*
;
"""
def __init__(self, href=None, mediaText=None, name=None,
parentRule=None, parentStyleSheet=None, readonly=False):
"""
If readonly allows setting of properties in constructor only
:param href:
location of the style sheet to be imported.
:param mediaText:
A list of media types for which this style sheet may be used
as a string
:param name:
Additional name of imported style sheet
"""
super(CSSImportRule, self).__init__(parentRule=parentRule,
parentStyleSheet=parentStyleSheet)
self._atkeyword = '@import'
self._styleSheet = None
# string or uri used for reserialization
self.hreftype = None
# prepare seq
seq = self._tempSeq()
seq.append(None, 'href')
# seq.append(None, 'media')
seq.append(None, 'name')
self._setSeq(seq)
# 1. media
if mediaText:
self.media = mediaText
else:
# must be all for @import
self.media = css_parser.stylesheets.MediaList(mediaText='all')
# 2. name
self.name = name
# 3. href and styleSheet
self.href = href
self._readonly = readonly
def __repr__(self):
if self._usemedia:
mediaText = self.media.mediaText
else:
mediaText = None
return "css_parser.css.%s(href=%r, mediaText=%r, name=%r)" % (
self.__class__.__name__,
self.href,
mediaText,
self.name)
def __str__(self):
if self._usemedia:
mediaText = self.media.mediaText
else:
mediaText = None
return ""\
% (self.__class__.__name__,
self.href,
mediaText,
self.name,
id(self))
_usemedia = property(lambda self: self.media.mediaText not in ('', 'all'),
doc="if self.media is used (or simply empty)")
def _getCssText(self):
"""Return serialized property cssText."""
return css_parser.ser.do_CSSImportRule(self)
def _setCssText(self, cssText):
"""
:exceptions:
- :exc:`~xml.dom.HierarchyRequestErr`:
Raised if the rule cannot be inserted at this point in the
style sheet.
- :exc:`~xml.dom.InvalidModificationErr`:
Raised if the specified CSS string value represents a different
type of rule than the current one.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if the rule is readonly.
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error and
is unparsable.
"""
super(CSSImportRule, self)._setCssText(cssText)
tokenizer = self._tokenize2(cssText)
attoken = self._nexttoken(tokenizer, None)
if self._type(attoken) != self._prods.IMPORT_SYM:
self._log.error('CSSImportRule: No CSSImportRule found: %s' %
self._valuestr(cssText),
error=xml.dom.InvalidModificationErr)
else:
# for closures: must be a mutable
new = {'keyword': self._tokenvalue(attoken),
'href': None,
'hreftype': None,
'media': None,
'name': None,
'wellformed': True
}
def __doname(seq, token):
# called by _string or _ident
new['name'] = self._stringtokenvalue(token)
seq.append(new['name'], 'name')
return ';'
def _string(expected, seq, token, tokenizer=None):
if 'href' == expected:
# href
new['href'] = self._stringtokenvalue(token)
new['hreftype'] = 'string'
seq.append(new['href'], 'href')
return 'media name ;'
elif 'name' in expected:
# name
return __doname(seq, token)
else:
new['wellformed'] = False
self._log.error(
'CSSImportRule: Unexpected string.', token)
return expected
def _uri(expected, seq, token, tokenizer=None):
# href
if 'href' == expected:
uri = self._uritokenvalue(token)
new['hreftype'] = 'uri'
new['href'] = uri
seq.append(new['href'], 'href')
return 'media name ;'
else:
new['wellformed'] = False
self._log.error(
'CSSImportRule: Unexpected URI.', token)
return expected
def _ident(expected, seq, token, tokenizer=None):
# medialist ending with ; which is checked upon too
if expected.startswith('media'):
mediatokens = self._tokensupto2(
tokenizer, importmediaqueryendonly=True)
mediatokens.insert(0, token) # push found token
last = mediatokens.pop() # retrieve ;
lastval, lasttyp = self._tokenvalue(last), self._type(last)
if lastval != ';' and lasttyp not in ('EOF',
self._prods.STRING):
new['wellformed'] = False
self._log.error('CSSImportRule: No ";" found: %s' %
self._valuestr(cssText), token=token)
newMedia = css_parser.stylesheets.MediaList(parentRule=self)
newMedia.mediaText = mediatokens
if newMedia.wellformed:
new['media'] = newMedia
seq.append(newMedia, 'media')
else:
new['wellformed'] = False
self._log.error('CSSImportRule: Invalid MediaList: %s' %
self._valuestr(cssText), token=token)
if lasttyp == self._prods.STRING:
# name
return __doname(seq, last)
else:
return 'EOF' # ';' is token "last"
else:
new['wellformed'] = False
self._log.error('CSSImportRule: Unexpected ident.', token)
return expected
def _char(expected, seq, token, tokenizer=None):
# final ;
val = self._tokenvalue(token)
if expected.endswith(';') and ';' == val:
return 'EOF'
else:
new['wellformed'] = False
self._log.error(
'CSSImportRule: Unexpected char.', token)
return expected
# import : IMPORT_SYM S* [STRING|URI]
# S* [ medium [ ',' S* medium]* ]? ';' S*
# STRING? # see http://www.w3.org/TR/css3-cascade/#cascading
# ;
newseq = self._tempSeq()
wellformed, expected = self._parse(expected='href',
seq=newseq, tokenizer=tokenizer,
productions={'STRING': _string,
'URI': _uri,
'IDENT': _ident,
'CHAR': _char},
new=new)
# wellformed set by parse
ok = wellformed and new['wellformed']
# post conditions
if not new['href']:
ok = False
self._log.error('CSSImportRule: No href found: %s' %
self._valuestr(cssText))
if expected != 'EOF':
ok = False
self._log.error('CSSImportRule: No ";" found: %s' %
self._valuestr(cssText))
# set all
if ok:
self._setSeq(newseq)
self.atkeyword = new['keyword']
self.hreftype = new['hreftype']
self.name = new['name']
if new['media']:
self.media = new['media']
else:
# must be all for @import
self.media = css_parser.stylesheets.MediaList(mediaText='all')
# needs new self.media
self.href = new['href']
cssText = property(fget=_getCssText, fset=_setCssText,
doc="(DOM) The parsable textual representation of this rule.")
def _setHref(self, href):
# set new href
self._href = href
# update seq
for i, item in enumerate(self.seq):
type_ = item.type
if 'href' == type_:
self._seq[i] = (href, type_, item.line, item.col)
break
importedSheet = css_parser.css.CSSStyleSheet(media=self.media, ownerRule=self, title=self.name)
self.hrefFound = False
# set styleSheet
if href and self.parentStyleSheet:
# loading errors are all catched!
# relative href
parentHref = self.parentStyleSheet.href
if parentHref is None:
# use cwd instead
parentHref = css_parser.helper.path2url(os.getcwd()) + '/'
fullhref = urllib_urljoin(parentHref, self.href)
# all possible exceptions are ignored
try:
usedEncoding, enctype, cssText = \
self.parentStyleSheet._resolveImport(fullhref)
if cssText is None:
# catched in next except below!
raise IOError('Cannot read Stylesheet.')
# contentEncoding with parentStyleSheet.overrideEncoding,
# HTTP or parent
encodingOverride, encoding = None, None
if enctype == 0:
encodingOverride = usedEncoding
elif 0 < enctype < 5:
encoding = usedEncoding
# inherit fetcher for @imports in styleSheet
importedSheet._href = fullhref
importedSheet._setFetcher(self.parentStyleSheet._fetcher)
importedSheet._setCssTextWithEncodingOverride(
cssText,
encodingOverride=encodingOverride,
encoding=encoding)
except (OSError, IOError, ValueError) as e:
self._log.warn('CSSImportRule: While processing imported '
'style sheet href=%s: %r'
% (self.href, e), neverraise=True)
else:
# used by resolveImports if to keep unprocessed href
self.hrefFound = True
self._styleSheet = importedSheet
_href = None # needs to be set
href = property(lambda self: self._href, _setHref,
doc="Location of the style sheet to be imported.")
def _setMedia(self, media):
"""
:param media:
a :class:`~css_parser.stylesheets.MediaList` or string
"""
self._checkReadonly()
# Under Python 2.X this was basestring but given unicode literals ...
if isinstance(media, string_type):
self._media = css_parser.stylesheets.MediaList(mediaText=media, parentRule=self)
else:
media._parentRule = self
self._media = media
# update seq
ihref = 0
for i, item in enumerate(self.seq):
if item.type == 'href':
ihref = i
elif item.type == 'media':
self.seq[i] = (self._media, 'media', None, None)
break
else:
# if no media until now add after href
self.seq.insert(ihref+1,
self._media, 'media', None, None)
media = property(lambda self: self._media, _setMedia,
doc="(DOM) A list of media types for this rule "
"of type :class:`~css_parser.stylesheets.MediaList`.")
def _setName(self, name=''):
"""Raises xml.dom.SyntaxErr if name is not a string."""
# Under Python 2.X this was basestring but given unicode literals ...
if name is None or isinstance(name, string_type):
# "" or '' handled as None
if not name:
name = None
# save name
self._name = name
# update seq
for i, item in enumerate(self.seq):
typ = item.type
if 'name' == typ:
self._seq[i] = (name, typ, item.line, item.col)
break
# set title of imported sheet
if self.styleSheet:
self.styleSheet.title = name
else:
self._log.error('CSSImportRule: Not a valid name: %s' % name)
name = property(lambda self: self._name, _setName,
doc="An optional name for the imported sheet.")
styleSheet = property(lambda self: self._styleSheet,
doc="(readonly) The style sheet referred to by this "
"rule.")
type = property(lambda self: self.IMPORT_RULE,
doc="The type of this rule, as defined by a CSSRule "
"type constant.")
def _getWellformed(self):
"Depending on if media is used at all."
if self._usemedia:
return bool(self.href and self.media.wellformed)
else:
return bool(self.href)
wellformed = property(_getWellformed)
css-parser-1.0.4/src/css_parser/css/cssmediarule.py 0000644 0001750 0001750 00000031155 13407703115 022633 0 ustar kovid kovid 0000000 0000000 """CSSMediaRule implements DOM Level 2 CSS CSSMediaRule."""
from __future__ import unicode_literals, division, absolute_import, print_function
__all__ = ['CSSMediaRule']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
from . import cssrule
import css_parser
import xml.dom
import sys
if sys.version_info[0] >= 3:
string_type = str
else:
string_type = basestring
class CSSMediaRule(cssrule.CSSRuleRules):
"""
Objects implementing the CSSMediaRule interface can be identified by the
MEDIA_RULE constant. On these objects the type attribute must return the
value of that constant.
Format::
: MEDIA_SYM S* medium [ COMMA S* medium ]*
STRING? # the name
LBRACE S* ruleset* '}' S*;
``cssRules``
All Rules in this media rule, a :class:`~css_parser.css.CSSRuleList`.
"""
def __init__(self, mediaText='all', name=None,
parentRule=None, parentStyleSheet=None, readonly=False):
"""constructor"""
super(CSSMediaRule, self).__init__(parentRule=parentRule,
parentStyleSheet=parentStyleSheet)
self._atkeyword = '@media'
# 1. media
if mediaText:
self.media = mediaText
else:
self.media = css_parser.stylesheets.MediaList()
self.name = name
self._readonly = readonly
def __repr__(self):
return "css_parser.css.%s(mediaText=%r)" % (
self.__class__.__name__,
self.media.mediaText)
def __str__(self):
return "" % (
self.__class__.__name__,
self.media.mediaText,
id(self))
def _getCssText(self):
"""Return serialized property cssText."""
return css_parser.ser.do_CSSMediaRule(self)
def _setCssText(self, cssText):
"""
:param cssText:
a parseable string or a tuple of (cssText, dict-of-namespaces)
:Exceptions:
- :exc:`~xml.dom.NamespaceErr`:
Raised if a specified selector uses an unknown namespace
prefix.
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error and
is unparsable.
- :exc:`~xml.dom.InvalidModificationErr`:
Raised if the specified CSS string value represents a different
type of rule than the current one.
- :exc:`~xml.dom.HierarchyRequestErr`:
Raised if the rule cannot be inserted at this point in the
style sheet.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if the rule is readonly.
"""
# media "name"? { cssRules }
super(CSSMediaRule, self)._setCssText(cssText)
# might be (cssText, namespaces)
cssText, namespaces = self._splitNamespacesOff(cssText)
tokenizer = self._tokenize2(cssText)
attoken = self._nexttoken(tokenizer, None)
if self._type(attoken) != self._prods.MEDIA_SYM:
self._log.error('CSSMediaRule: No CSSMediaRule found: %s' %
self._valuestr(cssText),
error=xml.dom.InvalidModificationErr)
else:
# save if parse goes wrong
oldMedia = self._media
oldCssRules = self._cssRules
ok = True
# media
mediatokens, end = self._tokensupto2(tokenizer,
mediaqueryendonly=True,
separateEnd=True)
if '{' == self._tokenvalue(end)\
or self._prods.STRING == self._type(end):
self.media = css_parser.stylesheets.MediaList(parentRule=self)
# TODO: remove special case
self.media.mediaText = mediatokens
ok = ok and self.media.wellformed
else:
ok = False
# name (optional)
name = None
nameseq = self._tempSeq()
if self._prods.STRING == self._type(end):
name = self._stringtokenvalue(end)
# TODO: for now comments are lost after name
nametokens, end = self._tokensupto2(tokenizer,
blockstartonly=True,
separateEnd=True)
wellformed, expected = self._parse(None,
nameseq,
nametokens,
{})
if not wellformed:
ok = False
self._log.error('CSSMediaRule: Syntax Error: %s' %
self._valuestr(cssText))
# check for {
if '{' != self._tokenvalue(end):
self._log.error('CSSMediaRule: No "{" found: %s' %
self._valuestr(cssText))
return
# cssRules
cssrulestokens, braceOrEOF = self._tokensupto2(tokenizer,
mediaendonly=True,
separateEnd=True)
nonetoken = self._nexttoken(tokenizer, None)
if 'EOF' == self._type(braceOrEOF):
# HACK!!!
# TODO: Not complete, add EOF to rule and } to @media
cssrulestokens.append(braceOrEOF)
braceOrEOF = ('CHAR', '}', 0, 0)
self._log.debug('CSSMediaRule: Incomplete, adding "}".',
token=braceOrEOF, neverraise=True)
if '}' != self._tokenvalue(braceOrEOF):
self._log.error('CSSMediaRule: No "}" found.',
token=braceOrEOF)
elif nonetoken:
self._log.error('CSSMediaRule: Trailing content found.',
token=nonetoken)
else:
# for closures: must be a mutable
new = {'wellformed': True}
def COMMENT(expected, seq, token, tokenizer=None):
self.insertRule(css_parser.css.CSSComment(
[token],
parentRule=self,
parentStyleSheet=self.parentStyleSheet))
return expected
def ruleset(expected, seq, token, tokenizer):
rule = css_parser.css.CSSStyleRule(
parentRule=self,
parentStyleSheet=self.parentStyleSheet)
rule.cssText = self._tokensupto2(tokenizer, token)
if rule.wellformed:
self.insertRule(rule)
return expected
def atrule(expected, seq, token, tokenizer):
# TODO: get complete rule!
tokens = self._tokensupto2(tokenizer, token)
atval = self._tokenvalue(token)
factories = {
'@page': css_parser.css.CSSPageRule,
'@media': CSSMediaRule,
}
if atval in ('@charset ', '@font-face', '@import',
'@namespace', '@variables'):
self._log.error('CSSMediaRule: This rule is not '
'allowed in CSSMediaRule - ignored: '
'%s.' % self._valuestr(tokens),
token=token,
error=xml.dom.HierarchyRequestErr)
elif atval in factories:
rule = factories[atval](
parentRule=self,
parentStyleSheet=self.parentStyleSheet)
rule.cssText = tokens
if rule.wellformed:
self.insertRule(rule)
else:
rule = css_parser.css.CSSUnknownRule(
tokens,
parentRule=self,
parentStyleSheet=self.parentStyleSheet)
if rule.wellformed:
self.insertRule(rule)
return expected
# save for possible reset
oldCssRules = self.cssRules
self.cssRules = css_parser.css.CSSRuleList()
seq = [] # not used really
tokenizer = iter(cssrulestokens)
wellformed, expected = self._parse(braceOrEOF,
seq,
tokenizer, {
'COMMENT': COMMENT,
'CHARSET_SYM': atrule,
'FONT_FACE_SYM': atrule,
'IMPORT_SYM': atrule,
'NAMESPACE_SYM': atrule,
'PAGE_SYM': atrule,
'MEDIA_SYM': atrule,
'ATKEYWORD': atrule
},
default=ruleset,
new=new)
ok = ok and wellformed
if ok:
self.name = name
self._setSeq(nameseq)
else:
self._media = oldMedia
self._cssRules = oldCssRules
cssText = property(_getCssText, _setCssText,
doc="(DOM) The parsable textual representation of this "
"rule.")
def _setName(self, name):
# Under Python 2.x this was basestring but given unicode literals ...
if isinstance(name, string_type) or name is None:
# "" or ''
if not name:
name = None
self._name = name
else:
self._log.error('CSSImportRule: Not a valid name: %s' % name)
name = property(lambda self: self._name, _setName,
doc="An optional name for this media rule.")
def _setMedia(self, media):
"""
:param media:
a :class:`~css_parser.stylesheets.MediaList` or string
"""
self._checkReadonly()
# Under Python 2.x this was basestring but given unicode literals ...
if isinstance(media, string_type):
self._media = css_parser.stylesheets.MediaList(
mediaText=media, parentRule=self)
else:
media._parentRule = self
self._media = media
# NOT IN @media seq at all?!
# # update seq
# for i, item in enumerate(self.seq):
# if item.type == 'media':
# self._seq[i] = (self._media, 'media', None, None)
# break
# else:
# # insert after @media if not in seq at all
# self.seq.insert(0,
# self._media, 'media', None, None)
media = property(lambda self: self._media, _setMedia,
doc="(DOM) A list of media types for this rule "
"of type :class:`~css_parser.stylesheets.MediaList`.")
def insertRule(self, rule, index=None):
"""Implements base ``insertRule``."""
rule, index = self._prepareInsertRule(rule, index)
if rule is False or rule is True:
# done or error
return
# check hierarchy
if isinstance(rule, css_parser.css.CSSCharsetRule) or \
isinstance(rule, css_parser.css.CSSFontFaceRule) or \
isinstance(rule, css_parser.css.CSSImportRule) or \
isinstance(rule, css_parser.css.CSSNamespaceRule) or \
isinstance(rule, css_parser.css.MarginRule):
self._log.error('%s: This type of rule is not allowed here: %s'
% (self.__class__.__name__, rule.cssText),
error=xml.dom.HierarchyRequestErr)
return
return self._finishInsertRule(rule, index)
type = property(lambda self: self.MEDIA_RULE,
doc="The type of this rule, as defined by a CSSRule "
"type constant.")
wellformed = property(lambda self: self.media.wellformed)
css-parser-1.0.4/src/css_parser/css/cssnamespacerule.py 0000644 0001750 0001750 00000026535 13407702010 023507 0 ustar kovid kovid 0000000 0000000 """CSSNamespaceRule currently implements http://dev.w3.org/csswg/css3-namespace/
"""
from __future__ import unicode_literals, division, absolute_import, print_function
__all__ = ['CSSNamespaceRule']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
from . import cssrule
import css_parser
import xml.dom
class CSSNamespaceRule(cssrule.CSSRule):
"""
Represents an @namespace rule within a CSS style sheet.
The @namespace at-rule declares a namespace prefix and associates
it with a given namespace (a string). This namespace prefix can then be
used in namespace-qualified names such as those described in the
Selectors Module [SELECT] or the Values and Units module [CSS3VAL].
Dealing with these rules directly is not needed anymore, easier is
the use of :attr:`css_parser.css.CSSStyleSheet.namespaces`.
Format::
namespace
: NAMESPACE_SYM S* [namespace_prefix S*]? [STRING|URI] S* ';' S*
;
namespace_prefix
: IDENT
;
"""
def __init__(self, namespaceURI=None, prefix=None, cssText=None,
parentRule=None, parentStyleSheet=None, readonly=False):
"""
:Parameters:
namespaceURI
The namespace URI (a simple string!) which is bound to the
given prefix. If no prefix is set
(``CSSNamespaceRule.prefix==''``) the namespace defined by
namespaceURI is set as the default namespace
prefix
The prefix used in the stylesheet for the given
``CSSNamespaceRule.uri``.
cssText
if no namespaceURI is given cssText must be given to set
a namespaceURI as this is readonly later on
parentStyleSheet
sheet where this rule belongs to
Do not use as positional but as keyword parameters only!
If readonly allows setting of properties in constructor only
format namespace::
namespace
: NAMESPACE_SYM S* [namespace_prefix S*]? [STRING|URI] S* ';' S*
;
namespace_prefix
: IDENT
;
"""
super(CSSNamespaceRule, self).__init__(parentRule=parentRule,
parentStyleSheet=parentStyleSheet)
self._atkeyword = '@namespace'
self._prefix = ''
self._namespaceURI = None
if namespaceURI:
self.namespaceURI = namespaceURI
self.prefix = prefix
tempseq = self._tempSeq()
tempseq.append(self.prefix, 'prefix')
tempseq.append(self.namespaceURI, 'namespaceURI')
self._setSeq(tempseq)
elif cssText is not None:
self.cssText = cssText
if parentStyleSheet:
self._parentStyleSheet = parentStyleSheet
self._readonly = readonly
def __repr__(self):
return "css_parser.css.%s(namespaceURI=%r, prefix=%r)" % (
self.__class__.__name__,
self.namespaceURI,
self.prefix)
def __str__(self):
return "" % (
self.__class__.__name__,
self.namespaceURI,
self.prefix,
id(self))
def _getCssText(self):
"""Return serialized property cssText"""
return css_parser.ser.do_CSSNamespaceRule(self)
def _setCssText(self, cssText):
"""
:param cssText: initial value for this rules cssText which is parsed
:exceptions:
- :exc:`~xml.dom.HierarchyRequestErr`:
Raised if the rule cannot be inserted at this point in the
style sheet.
- :exc:`~xml.dom.InvalidModificationErr`:
Raised if the specified CSS string value represents a different
type of rule than the current one.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if the rule is readonly.
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error and
is unparsable.
"""
super(CSSNamespaceRule, self)._setCssText(cssText)
tokenizer = self._tokenize2(cssText)
attoken = self._nexttoken(tokenizer, None)
if self._type(attoken) != self._prods.NAMESPACE_SYM:
self._log.error('CSSNamespaceRule: No CSSNamespaceRule found: %s' %
self._valuestr(cssText),
error=xml.dom.InvalidModificationErr)
else:
# for closures: must be a mutable
new = {'keyword': self._tokenvalue(attoken),
'prefix': '',
'uri': None,
'wellformed': True
}
def _ident(expected, seq, token, tokenizer=None):
# the namespace prefix, optional
if 'prefix or uri' == expected:
new['prefix'] = self._tokenvalue(token)
seq.append(new['prefix'], 'prefix')
return 'uri'
else:
new['wellformed'] = False
self._log.error(
'CSSNamespaceRule: Unexpected ident.', token)
return expected
def _string(expected, seq, token, tokenizer=None):
# the namespace URI as a STRING
if expected.endswith('uri'):
new['uri'] = self._stringtokenvalue(token)
seq.append(new['uri'], 'namespaceURI')
return ';'
else:
new['wellformed'] = False
self._log.error(
'CSSNamespaceRule: Unexpected string.', token)
return expected
def _uri(expected, seq, token, tokenizer=None):
# the namespace URI as URI which is DEPRECATED
if expected.endswith('uri'):
uri = self._uritokenvalue(token)
new['uri'] = uri
seq.append(new['uri'], 'namespaceURI')
return ';'
else:
new['wellformed'] = False
self._log.error(
'CSSNamespaceRule: Unexpected URI.', token)
return expected
def _char(expected, seq, token, tokenizer=None):
# final ;
val = self._tokenvalue(token)
if ';' == expected and ';' == val:
return 'EOF'
else:
new['wellformed'] = False
self._log.error(
'CSSNamespaceRule: Unexpected char.', token)
return expected
# "NAMESPACE_SYM S* [namespace_prefix S*]? [STRING|URI] S* ';' S*"
newseq = self._tempSeq()
wellformed, expected = self._parse(expected='prefix or uri',
seq=newseq, tokenizer=tokenizer,
productions={'IDENT': _ident,
'STRING': _string,
'URI': _uri,
'CHAR': _char},
new=new)
# wellformed set by parse
wellformed = wellformed and new['wellformed']
# post conditions
if new['uri'] is None:
wellformed = False
self._log.error('CSSNamespaceRule: No namespace URI found: %s'
% self._valuestr(cssText))
if expected != 'EOF':
wellformed = False
self._log.error('CSSNamespaceRule: No ";" found: %s' %
self._valuestr(cssText))
# set all
if wellformed:
self.atkeyword = new['keyword']
self._prefix = new['prefix']
self.namespaceURI = new['uri']
self._setSeq(newseq)
cssText = property(fget=_getCssText, fset=_setCssText,
doc="(DOM) The parsable textual representation of this "
"rule.")
def _setNamespaceURI(self, namespaceURI):
"""
:param namespaceURI: the initial value for this rules namespaceURI
:exceptions:
- :exc:`~xml.dom.NoModificationAllowedErr`:
(CSSRule) Raised if this rule is readonly or a namespaceURI is
already set in this rule.
"""
self._checkReadonly()
if not self._namespaceURI:
# initial setting
self._namespaceURI = namespaceURI
tempseq = self._tempSeq()
tempseq.append(namespaceURI, 'namespaceURI')
self._setSeq(tempseq) # makes seq readonly!
elif self._namespaceURI != namespaceURI:
self._log.error('CSSNamespaceRule: namespaceURI is readonly.',
error=xml.dom.NoModificationAllowedErr)
namespaceURI = property(lambda self: self._namespaceURI, _setNamespaceURI,
doc="URI (handled as simple string) of the defined namespace.")
def _replaceNamespaceURI(self, namespaceURI):
"""Used during parse of new sheet only!
:param namespaceURI: the new value for this rules namespaceURI
"""
self._namespaceURI = namespaceURI
for i, x in enumerate(self._seq):
if 'namespaceURI' == x.type:
self._seq._readonly = False
self._seq.replace(i, namespaceURI, 'namespaceURI')
self._seq._readonly = True
break
def _setPrefix(self, prefix=None):
"""
:param prefix: the new prefix
:exceptions:
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error and
is unparsable.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this rule is readonly.
"""
self._checkReadonly()
if not prefix:
prefix = ''
else:
tokenizer = self._tokenize2(prefix)
prefixtoken = self._nexttoken(tokenizer, None)
if not prefixtoken or self._type(prefixtoken) != self._prods.IDENT:
self._log.error('CSSNamespaceRule: No valid prefix "%s".' %
self._valuestr(prefix),
error=xml.dom.SyntaxErr)
return
else:
prefix = self._tokenvalue(prefixtoken)
# update seq
for i, x in enumerate(self._seq):
if x == self._prefix:
self._seq[i] = (prefix, 'prefix', None, None)
break
else:
# put prefix at the beginning!
self._seq[0] = (prefix, 'prefix', None, None)
# set new prefix
self._prefix = prefix
prefix = property(lambda self: self._prefix, _setPrefix,
doc="Prefix used for the defined namespace.")
type = property(lambda self: self.NAMESPACE_RULE,
doc="The type of this rule, as defined by a CSSRule "
"type constant.")
wellformed = property(lambda self: self.namespaceURI is not None)
css-parser-1.0.4/src/css_parser/css/csspagerule.py 0000644 0001750 0001750 00000037621 13407702010 022465 0 ustar kovid kovid 0000000 0000000 from __future__ import unicode_literals, division, absolute_import, print_function
import xml.dom
import css_parser
from . import cssrule
from .marginrule import MarginRule
from .cssstyledeclaration import CSSStyleDeclaration
from itertools import chain
"""CSSPageRule implements DOM Level 2 CSS CSSPageRule."""
__all__ = ['CSSPageRule']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
import sys
if sys.version_info[0] >= 3:
string_type = str
else:
string_type = basestring
def as_list(p):
if isinstance(p, list):
return p
return list(p)
class CSSPageRule(cssrule.CSSRuleRules):
"""
The CSSPageRule interface represents a @page rule within a CSS style
sheet. The @page rule is used to specify the dimensions, orientation,
margins, etc. of a page box for paged media.
Format::
page :
PAGE_SYM S* IDENT? pseudo_page? S*
'{' S* [ declaration | margin ]? [ ';' S* [ declaration | margin ]? ]* '}' S*
;
pseudo_page :
':' [ "left" | "right" | "first" ]
;
margin :
margin_sym S* '{' declaration [ ';' S* declaration? ]* '}' S*
;
margin_sym :
TOPLEFTCORNER_SYM |
TOPLEFT_SYM |
TOPCENTER_SYM |
TOPRIGHT_SYM |
TOPRIGHTCORNER_SYM |
BOTTOMLEFTCORNER_SYM |
BOTTOMLEFT_SYM |
BOTTOMCENTER_SYM |
BOTTOMRIGHT_SYM |
BOTTOMRIGHTCORNER_SYM |
LEFTTOP_SYM |
LEFTMIDDLE_SYM |
LEFTBOTTOM_SYM |
RIGHTTOP_SYM |
RIGHTMIDDLE_SYM |
RIGHTBOTTOM_SYM
;
`cssRules` contains a list of `MarginRule` objects.
"""
def __init__(self, selectorText=None, style=None, parentRule=None,
parentStyleSheet=None, readonly=False):
"""
If readonly allows setting of properties in constructor only.
:param selectorText:
type string
:param style:
CSSStyleDeclaration for this CSSStyleRule
"""
super(CSSPageRule, self).__init__(parentRule=parentRule,
parentStyleSheet=parentStyleSheet)
self._atkeyword = '@page'
self._specificity = (0, 0, 0)
tempseq = self._tempSeq()
if selectorText:
self.selectorText = selectorText
tempseq.append(self.selectorText, 'selectorText')
else:
self._selectorText = self._tempSeq()
if style:
self.style = style
else:
self.style = CSSStyleDeclaration()
tempseq.append(self.style, 'style')
self._setSeq(tempseq)
self._readonly = readonly
def __repr__(self):
return "css_parser.css.%s(selectorText=%r, style=%r)" % (
self.__class__.__name__,
self.selectorText,
self.style.cssText)
def __str__(self):
return ("") % (
self.__class__.__name__,
self.selectorText,
self.specificity,
self.style.cssText,
len(self.cssRules),
id(self))
def __contains__(self, margin):
"""Check if margin is set in the rule."""
return margin in as_list(self.keys())
def keys(self):
"Return list of all set margins (MarginRule)."
return as_list(r.margin for r in self.cssRules)
def __getitem__(self, margin):
"""Retrieve the style (of MarginRule)
for `margin` (which must be normalized).
"""
for r in self.cssRules:
if r.margin == margin:
return r.style
def __setitem__(self, margin, style):
"""Set the style (of MarginRule)
for `margin` (which must be normalized).
"""
for i, r in enumerate(self.cssRules):
if r.margin == margin:
r.style = style
return i
else:
return self.add(MarginRule(margin, style))
def __delitem__(self, margin):
"""Delete the style (the MarginRule)
for `margin` (which must be normalized).
"""
for r in self.cssRules:
if r.margin == margin:
self.deleteRule(r)
def __parseSelectorText(self, selectorText):
"""
Parse `selectorText` which may also be a list of tokens
and returns (selectorText, seq).
see _setSelectorText for details
"""
# for closures: must be a mutable
new = {'wellformed': True, 'last-S': False,
'name': 0, 'first': 0, 'lr': 0}
def _char(expected, seq, token, tokenizer=None):
# pseudo_page, :left, :right or :first
val = self._tokenvalue(token)
if not new['last-S'] and expected in ['page', ': or EOF']\
and ':' == val:
try:
identtoken = next(tokenizer)
except StopIteration:
self._log.error(
'CSSPageRule selectorText: No IDENT found.', token)
else:
ival, ityp = self._tokenvalue(identtoken),\
self._type(identtoken)
if self._prods.IDENT != ityp:
self._log.error('CSSPageRule selectorText: Expected '
'IDENT but found: %r' % ival, token)
else:
if ival not in ('first', 'left', 'right'):
self._log.warn('CSSPageRule: Unknown @page '
'selector: %r'
% (':'+ival,), neverraise=True)
if ival == 'first':
new['first'] = 1
else:
new['lr'] = 1
seq.append(val + ival, 'pseudo')
return 'EOF'
return expected
else:
new['wellformed'] = False
self._log.error('CSSPageRule selectorText: Unexpected CHAR: %r'
% val, token)
return expected
def S(expected, seq, token, tokenizer=None):
"Does not raise if EOF is found."
if expected == ': or EOF':
# pseudo must directly follow IDENT if given
new['last-S'] = True
return expected
def IDENT(expected, seq, token, tokenizer=None):
""
val = self._tokenvalue(token)
if 'page' == expected:
if self._normalize(val) == 'auto':
self._log.error('CSSPageRule selectorText: Invalid pagename.',
token)
else:
new['name'] = 1
seq.append(val, 'IDENT')
return ': or EOF'
else:
new['wellformed'] = False
self._log.error('CSSPageRule selectorText: Unexpected IDENT: '
'%r' % val, token)
return expected
def COMMENT(expected, seq, token, tokenizer=None):
"Does not raise if EOF is found."
seq.append(css_parser.css.CSSComment([token]), 'COMMENT')
return expected
newseq = self._tempSeq()
wellformed, expected = self._parse(expected='page',
seq=newseq, tokenizer=self._tokenize2(selectorText),
productions={'CHAR': _char,
'IDENT': IDENT,
'COMMENT': COMMENT,
'S': S},
new=new)
wellformed = wellformed and new['wellformed']
# post conditions
if expected == 'ident':
self._log.error(
'CSSPageRule selectorText: No valid selector: %r' %
self._valuestr(selectorText))
return wellformed, newseq, (new['name'], new['first'], new['lr'])
def __parseMarginAndStyle(self, tokens):
"tokens is a list, no generator (yet)"
g = iter(tokens)
styletokens = []
# new rules until parse done
cssRules = []
for token in g:
if token[0] == 'ATKEYWORD' and \
self._normalize(token[1]) in MarginRule.margins:
# MarginRule
m = MarginRule(parentRule=self,
parentStyleSheet=self.parentStyleSheet)
m.cssText = chain([token], g)
# merge if margin set more than once
for r in cssRules:
if r.margin == m.margin:
for p in m.style:
r.style.setProperty(p, replace=False)
break
else:
cssRules.append(m)
continue
# TODO: Properties?
styletokens.append(token)
return cssRules, styletokens
def _getCssText(self):
"""Return serialized property cssText."""
return css_parser.ser.do_CSSPageRule(self)
def _setCssText(self, cssText):
"""
:exceptions:
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error and
is unparsable.
- :exc:`~xml.dom.InvalidModificationErr`:
Raised if the specified CSS string value represents a different
type of rule than the current one.
- :exc:`~xml.dom.HierarchyRequestErr`:
Raised if the rule cannot be inserted at this point in the
style sheet.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if the rule is readonly.
"""
super(CSSPageRule, self)._setCssText(cssText)
tokenizer = self._tokenize2(cssText)
if self._type(self._nexttoken(tokenizer)) != self._prods.PAGE_SYM:
self._log.error('CSSPageRule: No CSSPageRule found: %s' %
self._valuestr(cssText),
error=xml.dom.InvalidModificationErr)
else:
newStyle = CSSStyleDeclaration(parentRule=self)
ok = True
selectortokens, startbrace = self._tokensupto2(tokenizer,
blockstartonly=True,
separateEnd=True)
styletokens, braceorEOFtoken = self._tokensupto2(tokenizer,
blockendonly=True,
separateEnd=True)
nonetoken = self._nexttoken(tokenizer)
if self._tokenvalue(startbrace) != '{':
ok = False
self._log.error('CSSPageRule: No start { of style declaration '
'found: %r' %
self._valuestr(cssText), startbrace)
elif nonetoken:
ok = False
self._log.error('CSSPageRule: Trailing content found.',
token=nonetoken)
selok, newselseq, specificity = self.__parseSelectorText(selectortokens)
ok = ok and selok
val, type_ = self._tokenvalue(braceorEOFtoken),\
self._type(braceorEOFtoken)
if val != '}' and type_ != 'EOF':
ok = False
self._log.error(
'CSSPageRule: No "}" after style declaration found: %r' %
self._valuestr(cssText))
else:
if 'EOF' == type_:
# add again as style needs it
styletokens.append(braceorEOFtoken)
# filter pagemargin rules out first
cssRules, styletokens = self.__parseMarginAndStyle(styletokens)
# SET, may raise:
newStyle.cssText = styletokens
if ok:
self._selectorText = newselseq
self._specificity = specificity
self.style = newStyle
self.cssRules = css_parser.css.CSSRuleList()
for r in cssRules:
self.cssRules.append(r)
cssText = property(_getCssText, _setCssText,
doc="(DOM) The parsable textual representation of this rule.")
def _getSelectorText(self):
"""Wrapper for css_parser Selector object."""
return css_parser.ser.do_CSSPageRuleSelector(self._selectorText)
def _setSelectorText(self, selectorText):
"""Wrapper for css_parser Selector object.
:param selectorText:
DOM String, in CSS 2.1 one of
- :first
- :left
- :right
- empty
:exceptions:
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error
and is unparsable.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this rule is readonly.
"""
self._checkReadonly()
# may raise SYNTAX_ERR
wellformed, newseq, specificity = self.__parseSelectorText(selectorText)
if wellformed:
self._selectorText = newseq
self._specificity = specificity
selectorText = property(_getSelectorText, _setSelectorText,
doc="(DOM) The parsable textual representation of "
"the page selector for the rule.")
def _setStyle(self, style):
"""
:param style:
a CSSStyleDeclaration or string
"""
self._checkReadonly()
# Under Python2.x this was basestring but given unicode literals ...
if isinstance(style, string_type):
self._style = CSSStyleDeclaration(cssText=style, parentRule=self)
else:
style._parentRule = self
self._style = style
style = property(lambda self: self._style, _setStyle,
doc="(DOM) The declaration-block of this rule set, "
"a :class:`~css_parser.css.CSSStyleDeclaration`.")
def insertRule(self, rule, index=None):
"""Implements base ``insertRule``."""
rule, index = self._prepareInsertRule(rule, index)
if rule is False or rule is True:
# done or error
return
# check hierarchy
if isinstance(rule, css_parser.css.CSSCharsetRule) or \
isinstance(rule, css_parser.css.CSSFontFaceRule) or \
isinstance(rule, css_parser.css.CSSImportRule) or \
isinstance(rule, css_parser.css.CSSNamespaceRule) or \
isinstance(rule, CSSPageRule) or \
isinstance(rule, css_parser.css.CSSMediaRule):
self._log.error('%s: This type of rule is not allowed here: %s'
% (self.__class__.__name__, rule.cssText),
error=xml.dom.HierarchyRequestErr)
return
return self._finishInsertRule(rule, index)
specificity = property(lambda self: self._specificity,
doc="""Specificity of this page rule (READONLY).
Tuple of (f, g, h) where:
- if the page selector has a named page, f=1; else f=0
- if the page selector has a ':first' pseudo-class, g=1; else g=0
- if the page selector has a ':left' or ':right' pseudo-class, h=1; else h=0
""")
type = property(lambda self: self.PAGE_RULE,
doc="The type of this rule, as defined by a CSSRule "
"type constant.")
# constant but needed:
wellformed = property(lambda self: True)
css-parser-1.0.4/src/css_parser/css/cssproperties.py 0000644 0001750 0001750 00000012125 13407702010 023045 0 ustar kovid kovid 0000000 0000000 """CSS2Properties (partly!) implements DOM Level 2 CSS CSS2Properties used
by CSSStyleDeclaration
TODO: CSS2Properties
If an implementation does implement this interface, it is expected to
understand the specific syntax of the shorthand properties, and apply
their semantics; when the margin property is set, for example, the
marginTop, marginRight, marginBottom and marginLeft properties are
actually being set by the underlying implementation.
When dealing with CSS "shorthand" properties, the shorthand properties
should be decomposed into their component longhand properties as
appropriate, and when querying for their value, the form returned
should be the shortest form exactly equivalent to the declarations made
in the ruleset. However, if there is no shorthand declaration that
could be added to the ruleset without changing in any way the rules
already declared in the ruleset (i.e., by adding longhand rules that
were previously not declared in the ruleset), then the empty string
should be returned for the shorthand property.
For example, querying for the font property should not return
"normal normal normal 14pt/normal Arial, sans-serif", when
"14pt Arial, sans-serif" suffices. (The normals are initial values, and
are implied by use of the longhand property.)
If the values for all the longhand properties that compose a particular
string are the initial values, then a string consisting of all the
initial values should be returned (e.g. a border-width value of
"medium" should be returned as such, not as "").
For some shorthand properties that take missing values from other
sides, such as the margin, padding, and border-[width|style|color]
properties, the minimum number of sides possible should be used; i.e.,
"0px 10px" will be returned instead of "0px 10px 0px 10px".
If the value of a shorthand property can not be decomposed into its
component longhand properties, as is the case for the font property
with a value of "menu", querying for the values of the component
longhand properties should return the empty string.
TODO: CSS2Properties DOMImplementation
The interface found within this section are not mandatory. A DOM
application can use the hasFeature method of the DOMImplementation
interface to determine whether it is supported or not. The feature
string for this extended interface listed in this section is "CSS2"
and the version is "2.0".
"""
__all__ = ['CSS2Properties']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
import css_parser.profiles
import re
class CSS2Properties(object):
"""The CSS2Properties interface represents a convenience mechanism
for retrieving and setting properties within a CSSStyleDeclaration.
The attributes of this interface correspond to all the properties
specified in CSS2. Getting an attribute of this interface is
equivalent to calling the getPropertyValue method of the
CSSStyleDeclaration interface. Setting an attribute of this
interface is equivalent to calling the setProperty method of the
CSSStyleDeclaration interface.
css_parser actually also allows usage of ``del`` to remove a CSS property
from a CSSStyleDeclaration.
This is an abstract class, the following functions need to be present
in inheriting class:
- ``_getP``
- ``_setP``
- ``_delP``
"""
# actual properties are set after the class definition!
def _getP(self, CSSname): pass
def _setP(self, CSSname, value): pass
def _delP(self, CSSname): pass
_reCSStoDOMname = re.compile('-[a-z]', re.I)
def _toDOMname(CSSname):
"""Returns DOMname for given CSSname e.g. for CSSname 'font-style' returns
'fontStyle'.
"""
def _doCSStoDOMname2(m): return m.group(0)[1].capitalize()
return _reCSStoDOMname.sub(_doCSStoDOMname2, CSSname)
_reDOMtoCSSname = re.compile('([A-Z])[a-z]+')
def _toCSSname(DOMname):
"""Return CSSname for given DOMname e.g. for DOMname 'fontStyle' returns
'font-style'.
"""
def _doDOMtoCSSname2(m): return '-' + m.group(0).lower()
return _reDOMtoCSSname.sub(_doDOMtoCSSname2, DOMname)
# add list of DOMname properties to CSS2Properties
# used for CSSStyleDeclaration to check if allowed properties
# but somehow doubled, any better way?
CSS2Properties._properties = []
for group in css_parser.profiles.properties:
for name in css_parser.profiles.properties[group]:
CSS2Properties._properties.append(_toDOMname(name))
# add CSS2Properties to CSSStyleDeclaration:
def __named_property_def(DOMname):
"""
Closure to keep name known in each properties accessor function
DOMname is converted to CSSname here, so actual calls use CSSname.
"""
CSSname = _toCSSname(DOMname)
def _get(self): return self._getP(CSSname)
def _set(self, value): self._setP(CSSname, value)
def _del(self): self._delP(CSSname)
return _get, _set, _del
# add all CSS2Properties to CSSStyleDeclaration
for DOMname in CSS2Properties._properties:
setattr(CSS2Properties, DOMname,
property(*__named_property_def(DOMname)))
css-parser-1.0.4/src/css_parser/css/cssrule.py 0000644 0001750 0001750 00000030111 13407702010 021613 0 ustar kovid kovid 0000000 0000000 from __future__ import unicode_literals, division, absolute_import, print_function
import xml.dom
import css_parser
"""CSSRule implements DOM Level 2 CSS CSSRule."""
__all__ = ['CSSRule']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
import sys
if sys.version_info[0] >= 3:
string_type = str
else:
string_type = basestring
class CSSRule(css_parser.util.Base2):
"""Abstract base interface for any type of CSS statement. This includes
both rule sets and at-rules. An implementation is expected to preserve
all rules specified in a CSS style sheet, even if the rule is not
recognized by the parser. Unrecognized rules are represented using the
:class:`CSSUnknownRule` interface.
"""
"""
CSSRule type constants.
An integer indicating which type of rule this is.
"""
UNKNOWN_RULE = 0
":class:`css_parser.css.CSSUnknownRule` (not used in CSSOM anymore)"
STYLE_RULE = 1
":class:`css_parser.css.CSSStyleRule`"
CHARSET_RULE = 2
":class:`css_parser.css.CSSCharsetRule` (not used in CSSOM anymore)"
IMPORT_RULE = 3
":class:`css_parser.css.CSSImportRule`"
MEDIA_RULE = 4
":class:`css_parser.css.CSSMediaRule`"
FONT_FACE_RULE = 5
":class:`css_parser.css.CSSFontFaceRule`"
PAGE_RULE = 6
":class:`css_parser.css.CSSPageRule`"
NAMESPACE_RULE = 10
""":class:`css_parser.css.CSSNamespaceRule`,
Value has changed in 0.9.7a3 due to a change in the CSSOM spec."""
COMMENT = 1001 # was -1, css_parser only
""":class:`css_parser.css.CSSComment` - not in the offical spec,
Value has changed in 0.9.7a3"""
VARIABLES_RULE = 1008
""":class:`css_parser.css.CSSVariablesRule` - experimental rule
not in the offical spec"""
MARGIN_RULE = 1006
""":class:`css_parser.css.MarginRule` - experimental rule
not in the offical spec"""
_typestrings = {UNKNOWN_RULE: 'UNKNOWN_RULE',
STYLE_RULE: 'STYLE_RULE',
CHARSET_RULE: 'CHARSET_RULE',
IMPORT_RULE: 'IMPORT_RULE',
MEDIA_RULE: 'MEDIA_RULE',
FONT_FACE_RULE: 'FONT_FACE_RULE',
PAGE_RULE: 'PAGE_RULE',
NAMESPACE_RULE: 'NAMESPACE_RULE',
COMMENT: 'COMMENT',
VARIABLES_RULE: 'VARIABLES_RULE',
MARGIN_RULE: 'MARGIN_RULE'
}
def __init__(self, parentRule=None, parentStyleSheet=None, readonly=False):
"""Set common attributes for all rules."""
super(CSSRule, self).__init__()
self._parent = parentRule
self._parentRule = parentRule
self._parentStyleSheet = parentStyleSheet
self._setSeq(self._tempSeq())
# self._atkeyword = None
# must be set after initialization of #inheriting rule is done
self._readonly = False
def _setAtkeyword(self, keyword):
"""Check if new keyword fits the rule it is used for."""
atkeyword = self._normalize(keyword)
if not self.atkeyword or (self.atkeyword == atkeyword):
self._atkeyword = atkeyword
self._keyword = keyword
else:
self._log.error('%s: Invalid atkeyword for this rule: %r' %
(self.atkeyword, keyword),
error=xml.dom.InvalidModificationErr)
atkeyword = property(lambda self: self._atkeyword, _setAtkeyword,
doc="Normalized keyword of an @rule (e.g. ``@import``).")
def _setCssText(self, cssText):
"""
:param cssText:
A parsable DOMString.
:exceptions:
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error and
is unparsable.
- :exc:`~xml.dom.InvalidModificationErr`:
Raised if the specified CSS string value represents a different
type of rule than the current one.
- :exc:`~xml.dom.HierarchyRequestErr`:
Raised if the rule cannot be inserted at this point in the
style sheet.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if the rule is readonly.
"""
self._checkReadonly()
cssText = property(lambda self: '', _setCssText,
doc="(DOM) The parsable textual representation of the "
"rule. This reflects the current state of the rule "
"and not its initial value.")
parent = property(lambda self: self._parent,
doc="The Parent Node of this CSSRule or None.")
parentRule = property(lambda self: self._parentRule,
doc="If this rule is contained inside another rule "
"(e.g. a style rule inside an @media block), this "
"is the containing rule. If this rule is not nested "
"inside any other rules, this returns None.")
def _getParentStyleSheet(self):
# rules contained in other rules (@media) use that rules parent
if (self.parentRule):
return self.parentRule._parentStyleSheet
else:
return self._parentStyleSheet
parentStyleSheet = property(_getParentStyleSheet,
doc="The style sheet that contains this rule.")
type = property(lambda self: self.UNKNOWN_RULE,
doc="The type of this rule, as defined by a CSSRule "
"type constant.")
typeString = property(lambda self: CSSRule._typestrings[self.type],
doc="Descriptive name of this rule's type.")
wellformed = property(lambda self: False,
doc="If the rule is wellformed.")
class CSSRuleRules(CSSRule):
"""Abstract base interface for rules that contain other rules
like @media or @page. Methods may be overwritten if a rule
has specific stuff to do like checking the order of insertion like
@media does.
"""
def __init__(self, parentRule=None, parentStyleSheet=None):
super(CSSRuleRules, self).__init__(parentRule=parentRule,
parentStyleSheet=parentStyleSheet)
self.cssRules = css_parser.css.CSSRuleList()
def __iter__(self):
"""Generator iterating over these rule's cssRules."""
for rule in self._cssRules:
yield rule
def _setCssRules(self, cssRules):
"Set new cssRules and update contained rules refs."
cssRules.append = self.insertRule
cssRules.extend = self.insertRule
cssRules.__delitem__ == self.deleteRule
for rule in cssRules:
rule._parentRule = self
rule._parentStyleSheet = None
self._cssRules = cssRules
cssRules = property(lambda self: self._cssRules, _setCssRules,
"All Rules in this style sheet, a "
":class:`~css_parser.css.CSSRuleList`.")
def deleteRule(self, index):
"""
Delete the rule at `index` from rules ``cssRules``.
:param index:
The `index` of the rule to be removed from the rules cssRules
list. For an `index` < 0 **no** :exc:`~xml.dom.IndexSizeErr` is
raised but rules for normal Python lists are used. E.g.
``deleteRule(-1)`` removes the last rule in cssRules.
`index` may also be a CSSRule object which will then be removed.
:Exceptions:
- :exc:`~xml.dom.IndexSizeErr`:
Raised if the specified index does not correspond to a rule in
the media rule list.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this media rule is readonly.
"""
self._checkReadonly()
if isinstance(index, CSSRule):
for i, r in enumerate(self.cssRules):
if index == r:
index = i
break
else:
raise xml.dom.IndexSizeErr("%s: Not a rule in "
"this rule'a cssRules list: %s"
% (self.__class__.__name__, index))
try:
# detach
self._cssRules[index]._parentRule = None
del self._cssRules[index]
except IndexError:
raise xml.dom.IndexSizeErr('%s: %s is not a valid index '
'in the rulelist of length %i'
% (self.__class__.__name__,
index, self._cssRules.length))
def _prepareInsertRule(self, rule, index=None):
"return checked `index` and optional parsed `rule`"
self._checkReadonly()
# check index
if index is None:
index = len(self._cssRules)
elif index < 0 or index > self._cssRules.length:
raise xml.dom.IndexSizeErr('%s: Invalid index %s for '
'CSSRuleList with a length of %s.'
% (self.__class__.__name__,
index, self._cssRules.length))
# check and optionally parse rule
# Under Python 2.x this was basestring but ...
if isinstance(rule, string_type):
tempsheet = css_parser.css.CSSStyleSheet()
tempsheet.cssText = rule
if len(tempsheet.cssRules) != 1 or (tempsheet.cssRules and
not isinstance(tempsheet.cssRules[0], css_parser.css.CSSRule)):
self._log.error('%s: Invalid Rule: %s' % (self.__class__.__name__,
rule))
return False, False
rule = tempsheet.cssRules[0]
elif isinstance(rule, css_parser.css.CSSRuleList):
# insert all rules
for i, r in enumerate(rule):
self.insertRule(r, index + i)
return True, True
elif not isinstance(rule, css_parser.css.CSSRule):
self._log.error('%s: Not a CSSRule: %s' % (rule,
self.__class__.__name__))
return False, False
return rule, index
def _finishInsertRule(self, rule, index):
"add `rule` at `index`"
rule._parentRule = self
rule._parentStyleSheet = None
self._cssRules.insert(index, rule)
return index
def add(self, rule):
"""Add `rule` to page rule. Same as ``insertRule(rule)``."""
return self.insertRule(rule)
def insertRule(self, rule, index=None):
"""
Insert `rule` into the rules ``cssRules``.
:param rule:
the parsable text representing the `rule` to be inserted. For rule
sets this contains both the selector and the style declaration.
For at-rules, this specifies both the at-identifier and the rule
content.
css_parser also allows rule to be a valid
:class:`~css_parser.css.CSSRule` object.
:param index:
before the `index` the specified `rule` will be inserted.
If the specified `index` is equal to the length of the rules
rule collection, the rule will be added to the end of the rule.
If index is not given or None rule will be appended to rule
list.
:returns:
the index of the newly inserted rule.
:exceptions:
- :exc:`~xml.dom.HierarchyRequestErr`:
Raised if the `rule` cannot be inserted at the specified `index`,
e.g., if an @import rule is inserted after a standard rule set
or other at-rule.
- :exc:`~xml.dom.IndexSizeErr`:
Raised if the specified `index` is not a valid insertion point.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this rule is readonly.
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified `rule` has a syntax error and is
unparsable.
"""
return self._prepareInsertRule(rule, index)
css-parser-1.0.4/src/css_parser/css/cssrulelist.py 0000644 0001750 0001750 00000004077 13407702010 022523 0 ustar kovid kovid 0000000 0000000 """CSSRuleList implements DOM Level 2 CSS CSSRuleList.
Partly also http://dev.w3.org/csswg/cssom/#the-cssrulelist."""
from __future__ import unicode_literals, division, absolute_import, print_function
__all__ = ['CSSRuleList']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
class CSSRuleList(list):
"""The CSSRuleList object represents an (ordered) list of statements.
The items in the CSSRuleList are accessible via an integral index,
starting from 0.
Subclasses a standard Python list so theoretically all standard list
methods are available. Setting methods like ``__init__``, ``append``,
``extend`` or ``__setslice__`` are added later on instances of this
class if so desired.
E.g. CSSStyleSheet adds ``append`` which is not available in a simple
instance of this class!
"""
def __init__(self, *ignored):
"Nothing is set as this must also be defined later."
pass
def __notimplemented(self, *ignored):
"Implemented in class using a CSSRuleList only."
raise NotImplementedError(
'Must be implemented by class using an instance of this class.')
append = extend = __setitem__ = __setslice__ = __notimplemented
def item(self, index):
"""(DOM) Retrieve a CSS rule by ordinal `index`. The order in this
collection represents the order of the rules in the CSS style
sheet. If index is greater than or equal to the number of rules in
the list, this returns None.
Returns CSSRule, the style rule at the index position in the
CSSRuleList, or None if that is not a valid index.
"""
try:
return self[index]
except IndexError:
return None
length = property(lambda self: len(self),
doc="(DOM) The number of CSSRules in the list.")
def rulesOfType(self, type):
"""Yield the rules which have the given `type` only, one of the
constants defined in :class:`css_parser.css.CSSRule`."""
for r in self:
if r.type == type:
yield r
css-parser-1.0.4/src/css_parser/css/cssstyledeclaration.py 0000644 0001750 0001750 00000065564 13407702010 024236 0 ustar kovid kovid 0000000 0000000 """CSSStyleDeclaration implements DOM Level 2 CSS CSSStyleDeclaration and
extends CSS2Properties
see
http://www.w3.org/TR/1998/REC-CSS2-19980512/syndata.html#parsing-errors
Unknown properties
------------------
User agents must ignore a declaration with an unknown property.
For example, if the style sheet is::
H1 { color: red; rotation: 70minutes }
the user agent will treat this as if the style sheet had been::
H1 { color: red }
css_parser gives a message about any unknown properties but
keeps any property (if syntactically correct).
Illegal values
--------------
User agents must ignore a declaration with an illegal value. For example::
IMG { float: left } /* correct CSS2 */
IMG { float: left here } /* "here" is not a value of 'float' */
IMG { background: "red" } /* keywords cannot be quoted in CSS2 */
IMG { border-width: 3 } /* a unit must be specified for length values */
A CSS2 parser would honor the first rule and ignore the rest, as if the
style sheet had been::
IMG { float: left }
IMG { }
IMG { }
IMG { }
css_parser again will issue a message (WARNING in this case) about invalid
CSS2 property values.
TODO:
This interface is also used to provide a read-only access to the
computed values of an element. See also the ViewCSS interface.
- return computed values and not literal values
- simplify unit pairs/triples/quadruples
2px 2px 2px 2px -> 2px for border/padding...
- normalize compound properties like:
background: no-repeat left url() #fff
-> background: #fff url() no-repeat left
"""
from __future__ import unicode_literals, division, absolute_import, print_function
__all__ = ['CSSStyleDeclaration', 'Property']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
from .cssproperties import CSS2Properties
from .property import Property
import css_parser
class CSSStyleDeclaration(CSS2Properties, css_parser.util.Base2):
"""The CSSStyleDeclaration class represents a single CSS declaration
block. This class may be used to determine the style properties
currently set in a block or to set style properties explicitly
within the block.
While an implementation may not recognize all CSS properties within
a CSS declaration block, it is expected to provide access to all
specified properties in the style sheet through the
CSSStyleDeclaration interface.
Furthermore, implementations that support a specific level of CSS
should correctly handle CSS shorthand properties for that level. For
a further discussion of shorthand properties, see the CSS2Properties
interface.
Additionally the CSS2Properties interface is implemented.
$css2propertyname
All properties defined in the CSS2Properties class are available
as direct properties of CSSStyleDeclaration with their respective
DOM name, so e.g. ``fontStyle`` for property 'font-style'.
These may be used as::
>>> style = CSSStyleDeclaration(cssText='color: red')
>>> style.color = 'green'
>>> print style.color
green
>>> del style.color
>>> print style.color
Format::
[Property: Value Priority?;]* [Property: Value Priority?]?
"""
def __init__(self, cssText='', parentRule=None, readonly=False,
validating=None):
"""
:param cssText:
Shortcut, sets CSSStyleDeclaration.cssText
:param parentRule:
The CSS rule that contains this declaration block or
None if this CSSStyleDeclaration is not attached to a CSSRule.
:param readonly:
defaults to False
:param validating:
a flag defining if this sheet should be validated on change.
Defaults to None, which means defer to the parent stylesheet.
"""
super(CSSStyleDeclaration, self).__init__()
self._parentRule = parentRule
self.validating = validating
self.cssText = cssText
self._readonly = readonly
def __contains__(self, nameOrProperty):
"""Check if a property (or a property with given name) is in style.
:param name:
a string or Property, uses normalized name and not literalname
"""
if isinstance(nameOrProperty, Property):
name = nameOrProperty.name
else:
name = self._normalize(nameOrProperty)
return name in self.__nnames()
def __iter__(self):
"""Iterator of set Property objects with different normalized names."""
def properties():
for name in self.__nnames():
yield self.getProperty(name)
return properties()
def keys(self):
"""Analoguous to standard dict returns property names which are set in
this declaration."""
return list(self.__nnames())
def __getitem__(self, CSSName):
"""Retrieve the value of property ``CSSName`` from this declaration.
``CSSName`` will be always normalized.
"""
return self.getPropertyValue(CSSName)
def __setitem__(self, CSSName, value):
"""Set value of property ``CSSName``. ``value`` may also be a tuple of
(value, priority), e.g. style['color'] = ('red', 'important')
``CSSName`` will be always normalized.
"""
priority = None
if isinstance(value, tuple):
value, priority = value
return self.setProperty(CSSName, value, priority)
def __delitem__(self, CSSName):
"""Delete property ``CSSName`` from this declaration.
If property is not in this declaration return u'' just like
removeProperty.
``CSSName`` will be always normalized.
"""
return self.removeProperty(CSSName)
def __setattr__(self, n, v):
"""Prevent setting of unknown properties on CSSStyleDeclaration
which would not work anyway. For these
``CSSStyleDeclaration.setProperty`` MUST be called explicitly!
TODO:
implementation of known is not really nice, any alternative?
"""
known = ['_tokenizer', '_log', '_ttypes',
'_seq', 'seq', 'parentRule', '_parentRule', 'cssText',
'valid', 'wellformed', 'validating',
'_readonly', '_profiles', '_validating']
known.extend(CSS2Properties._properties)
if n in known:
super(CSSStyleDeclaration, self).__setattr__(n, v)
else:
raise AttributeError('Unknown CSS Property, '
'``CSSStyleDeclaration.setProperty("%s", '
'...)`` MUST be used.' % n)
def __repr__(self):
return "css_parser.css.%s(cssText=%r)" % (
self.__class__.__name__,
self.getCssText(separator=' '))
def __str__(self):
return "" % (
self.__class__.__name__,
self.length,
len(self.getProperties(all=True)),
id(self))
def __nnames(self):
"""Return iterator for all different names in order as set
if names are set twice the last one is used (double reverse!)
"""
names = []
for item in reversed(self.seq):
val = item.value
if isinstance(val, Property) and val.name not in names:
names.append(val.name)
return reversed(names)
# overwritten accessor functions for CSS2Properties' properties
def _getP(self, CSSName):
"""(DOM CSS2Properties) Overwritten here and effectively the same as
``self.getPropertyValue(CSSname)``.
Parameter is in CSSname format ('font-style'), see CSS2Properties.
Example::
>>> style = CSSStyleDeclaration(cssText='font-style:italic;')
>>> print style.fontStyle
italic
"""
return self.getPropertyValue(CSSName)
def _setP(self, CSSName, value):
"""(DOM CSS2Properties) Overwritten here and effectively the same as
``self.setProperty(CSSname, value)``.
Only known CSS2Properties may be set this way, otherwise an
AttributeError is raised.
For these unknown properties ``setPropertyValue(CSSname, value)``
has to be called explicitly.
Also setting the priority of properties needs to be done with a
call like ``setPropertyValue(CSSname, value, priority)``.
Example::
>>> style = CSSStyleDeclaration()
>>> style.fontStyle = 'italic'
>>> # or
>>> style.setProperty('font-style', 'italic', '!important')
"""
self.setProperty(CSSName, value)
# TODO: Shorthand ones
def _delP(self, CSSName):
"""(css_parser only) Overwritten here and effectively the same as
``self.removeProperty(CSSname)``.
Example::
>>> style = CSSStyleDeclaration(cssText='font-style:italic;')
>>> del style.fontStyle
>>> print style.fontStyle
"""
self.removeProperty(CSSName)
def children(self):
"""Generator yielding any known child in this declaration including
*all* properties, comments or CSSUnknownrules.
"""
for item in self._seq:
yield item.value
def _getCssText(self):
"""Return serialized property cssText."""
return css_parser.ser.do_css_CSSStyleDeclaration(self)
def _setCssText(self, cssText):
"""Setting this attribute will result in the parsing of the new value
and resetting of all the properties in the declaration block
including the removal or addition of properties.
:exceptions:
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this declaration is readonly or a property is readonly.
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error and
is unparsable.
"""
self._checkReadonly()
tokenizer = self._tokenize2(cssText)
def ident(expected, seq, token, tokenizer=None):
# a property
tokens = self._tokensupto2(tokenizer, starttoken=token,
semicolon=True)
if self._tokenvalue(tokens[-1]) == ';':
tokens.pop()
property = Property(parent=self)
property.cssText = tokens
if property.wellformed:
seq.append(property, 'Property')
else:
self._log.error('CSSStyleDeclaration: Syntax Error in '
'Property: %s' % self._valuestr(tokens))
# does not matter in this case
return expected
def unexpected(expected, seq, token, tokenizer=None):
# error, find next ; or } to omit upto next property
ignored = self._tokenvalue(token) + self._valuestr(
self._tokensupto2(tokenizer,
propertyvalueendonly=True))
self._log.error('CSSStyleDeclaration: Unexpected token, ignoring '
'upto %r.' % ignored, token)
# does not matter in this case
return expected
def char(expected, seq, token, tokenizer=None):
# a standalone ; or error...
if self._tokenvalue(token) == ';':
self._log.info('CSSStyleDeclaration: Stripped standalone semicolon'
': %s' % self._valuestr([token]), neverraise=True)
return expected
else:
return unexpected(expected, seq, token, tokenizer)
# [Property: Value;]* Property: Value?
newseq = self._tempSeq()
wellformed, expected = self._parse(expected=None,
seq=newseq, tokenizer=tokenizer,
productions={'IDENT': ident, 'CHAR': char},
default=unexpected)
# wellformed set by parse
for item in newseq:
item.value._parent = self
# do not check wellformed as invalid things are removed anyway
self._setSeq(newseq)
cssText = property(_getCssText, _setCssText,
doc="(DOM) A parsable textual representation of the "
"declaration block excluding the surrounding curly "
"braces.")
def getCssText(self, separator=None):
"""
:returns:
serialized property cssText, each property separated by
given `separator` which may e.g. be ``u''`` to be able to use
cssText directly in an HTML style attribute. ``;`` is part of
each property (except the last one) and **cannot** be set with
separator!
"""
return css_parser.ser.do_css_CSSStyleDeclaration(self, separator)
def _setParentRule(self, parentRule):
self._parentRule = parentRule
# for x in self.children():
# x.parent = self
parentRule = property(lambda self: self._parentRule, _setParentRule,
doc="(DOM) The CSS rule that contains this declaration block or "
"None if this CSSStyleDeclaration is not attached to a CSSRule.")
def getProperties(self, name=None, all=False):
"""
:param name:
optional `name` of properties which are requested.
Only properties with this **always normalized** `name` are returned.
If `name` is ``None`` all properties are returned (at least one for
each set name depending on parameter `all`).
:param all:
if ``False`` (DEFAULT) only the effective properties are returned.
If name is given a list with only one property is returned.
if ``True`` all properties including properties set multiple times
with different values or priorities for different UAs are returned.
The order of the properties is fully kept as in the original
stylesheet.
:returns:
a list of :class:`~css_parser.css.Property` objects set in
this declaration.
"""
if name and not all:
# single prop but list
p = self.getProperty(name)
if p:
return [p]
else:
return []
elif not all:
# effective Properties in name order
return [self.getProperty(name_) for name_ in self.__nnames()]
else:
# all properties or all with this name
nname = self._normalize(name)
properties = []
for item in self.seq:
val = item.value
if isinstance(val, Property) and (
(bool(nname) is False) or (val.name == nname)):
properties.append(val)
return properties
def getProperty(self, name, normalize=True):
r"""
:param name:
of the CSS property, always lowercase (even if not normalized)
:param normalize:
if ``True`` (DEFAULT) name will be normalized (lowercase, no simple
escapes) so "color", "COLOR" or "C\olor" will all be equivalent
If ``False`` may return **NOT** the effective value but the
effective for the unnormalized name.
:returns:
the effective :class:`~css_parser.css.Property` object.
"""
nname = self._normalize(name)
found = None
for item in reversed(self.seq):
val = item.value
if isinstance(val, Property):
if (normalize and nname == val.name) or name == val.literalname:
if val.priority:
return val
elif not found:
found = val
return found
def getPropertyCSSValue(self, name, normalize=True):
r"""
:param name:
of the CSS property, always lowercase (even if not normalized)
:param normalize:
if ``True`` (DEFAULT) name will be normalized (lowercase, no simple
escapes) so "color", "COLOR" or "C\olor" will all be equivalent
If ``False`` may return **NOT** the effective value but the
effective for the unnormalized name.
:returns:
:class:`~css_parser.css.CSSValue`, the value of the effective
property if it has been explicitly set for this declaration block.
(DOM)
Used to retrieve the object representation of the value of a CSS
property if it has been explicitly set within this declaration
block. Returns None if the property has not been set.
(This method returns None if the property is a shorthand
property. Shorthand property values can only be accessed and
modified as strings, using the getPropertyValue and setProperty
methods.)
**css_parser currently always returns a CSSValue if the property is
set.**
for more on shorthand properties see
http://www.dustindiaz.com/css-shorthand/
"""
nname = self._normalize(name)
if nname in self._SHORTHANDPROPERTIES:
self._log.info('CSSValue for shorthand property "%s" should be '
'None, this may be implemented later.' %
nname, neverraise=True)
p = self.getProperty(name, normalize)
if p:
return p.propertyValue
else:
return None
def getPropertyValue(self, name, normalize=True):
r"""
:param name:
of the CSS property, always lowercase (even if not normalized)
:param normalize:
if ``True`` (DEFAULT) name will be normalized (lowercase, no simple
escapes) so "color", "COLOR" or "C\olor" will all be equivalent
If ``False`` may return **NOT** the effective value but the
effective for the unnormalized name.
:returns:
the value of the effective property if it has been explicitly set
for this declaration block. Returns the empty string if the
property has not been set.
"""
p = self.getProperty(name, normalize)
if p:
return p.value
else:
return ''
def getPropertyPriority(self, name, normalize=True):
r"""
:param name:
of the CSS property, always lowercase (even if not normalized)
:param normalize:
if ``True`` (DEFAULT) name will be normalized (lowercase, no simple
escapes) so "color", "COLOR" or "C\olor" will all be equivalent
If ``False`` may return **NOT** the effective value but the
effective for the unnormalized name.
:returns:
the priority of the effective CSS property (e.g. the
"important" qualifier) if the property has been explicitly set in
this declaration block. The empty string if none exists.
"""
p = self.getProperty(name, normalize)
if p:
return p.priority
else:
return ''
def removeProperty(self, name, normalize=True):
r"""
(DOM)
Used to remove a CSS property if it has been explicitly set within
this declaration block.
:param name:
of the CSS property
:param normalize:
if ``True`` (DEFAULT) name will be normalized (lowercase, no simple
escapes) so "color", "COLOR" or "C\olor" will all be equivalent.
The effective Property value is returned and *all* Properties
with ``Property.name == name`` are removed.
If ``False`` may return **NOT** the effective value but the
effective for the unnormalized `name` only. Also only the
Properties with the literal name `name` are removed.
:returns:
the value of the property if it has been explicitly set for
this declaration block. Returns the empty string if the property
has not been set or the property name does not correspond to a
known CSS property
:exceptions:
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this declaration is readonly or the property is
readonly.
"""
self._checkReadonly()
r = self.getPropertyValue(name, normalize=normalize)
newseq = self._tempSeq()
if normalize:
# remove all properties with name == nname
nname = self._normalize(name)
for item in self.seq:
if not (isinstance(item.value, Property)
and item.value.name == nname):
newseq.appendItem(item)
else:
# remove all properties with literalname == name
for item in self.seq:
if not (isinstance(item.value, Property)
and item.value.literalname == name):
newseq.appendItem(item)
self._setSeq(newseq)
return r
def setProperty(self, name, value=None, priority='',
normalize=True, replace=True):
r"""(DOM) Set a property value and priority within this declaration
block.
:param name:
of the CSS property to set (in W3C DOM the parameter is called
"propertyName"), always lowercase (even if not normalized)
If a property with this `name` is present it will be reset.
css_parser also allowed `name` to be a
:class:`~css_parser.css.Property` object, all other
parameter are ignored in this case
:param value:
the new value of the property, ignored if `name` is a Property.
:param priority:
the optional priority of the property (e.g. "important"),
ignored if `name` is a Property.
:param normalize:
if True (DEFAULT) `name` will be normalized (lowercase, no simple
escapes) so "color", "COLOR" or "C\olor" will all be equivalent
:param replace:
if True (DEFAULT) the given property will replace a present
property. If False a new property will be added always.
The difference to `normalize` is that two or more properties with
the same name may be set, useful for e.g. stuff like::
background: red;
background: rgba(255, 0, 0, 0.5);
which defines the same property but only capable UAs use the last
property value, older ones use the first value.
:exceptions:
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified value has a syntax error and is
unparsable.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this declaration is readonly or the property is
readonly.
"""
self._checkReadonly()
if isinstance(name, Property):
newp = name
name = newp.literalname
elif not value:
# empty string or None effectively removed property
return self.removeProperty(name)
else:
newp = Property(name, value, priority, parent=self)
if newp.wellformed:
if replace:
# check if update
nname = self._normalize(name)
properties = self.getProperties(name, all=(not normalize))
for property in reversed(properties):
if normalize and property.name == nname:
property.propertyValue = newp.propertyValue.cssText
property.priority = newp.priority
return
elif property.literalname == name:
property.propertyValue = newp.propertyValue.cssText
property.priority = newp.priority
return
# not yet set or forced omit replace
newp.parent = self
self.seq._readonly = False
self.seq.append(newp, 'Property')
self.seq._readonly = True
else:
self._log.warn('Invalid Property: %s: %s %s'
% (name, value, priority))
def item(self, index):
r"""(DOM) Retrieve the properties that have been explicitly set in
this declaration block. The order of the properties retrieved using
this method does not have to be the order in which they were set.
This method can be used to iterate over all properties in this
declaration block.
:param index:
of the property to retrieve, negative values behave like
negative indexes on Python lists, so -1 is the last element
:returns:
the name of the property at this ordinal position. The
empty string if no property exists at this position.
**ATTENTION:**
Only properties with different names are counted. If two
properties with the same name are present in this declaration
only the effective one is included.
:meth:`item` and :attr:`length` work on the same set here.
"""
names = list(self.__nnames())
try:
return names[index]
except IndexError:
return ''
length = property(lambda self: len(list(self.__nnames())),
doc="(DOM) The number of distinct properties that have "
"been explicitly in this declaration block. The "
"range of valid indices is 0 to length-1 inclusive. "
"These are properties with a different ``name`` "
"only. :meth:`item` and :attr:`length` work on the "
"same set here.")
def _getValidating(self):
try:
# CSSParser.parseX() sets validating of stylesheet
return self.parentRule.parentStyleSheet.validating
except AttributeError:
# CSSParser.parseStyle() sets validating of declaration
if self._validating is not None:
return self._validating
# default
return True
def _setValidating(self, validating):
self._validating = validating
validating = property(_getValidating, _setValidating,
doc="If ``True`` this declaration validates "
"contained properties. The parent StyleSheet "
"validation setting does *always* win though so "
"even if validating is True it may not validate "
"if the StyleSheet defines else!")
def _getValid(self):
"""Check each contained property for validity."""
return all(prop.valid for prop in self.getProperties())
valid = property(_getValid,
doc='``True`` if each property is valid.')
css-parser-1.0.4/src/css_parser/css/cssstylerule.py 0000644 0001750 0001750 00000022606 13407702010 022706 0 ustar kovid kovid 0000000 0000000 from __future__ import unicode_literals, division, absolute_import, print_function
import xml.dom
import css_parser
from . import cssrule
from .selectorlist import SelectorList
from .cssstyledeclaration import CSSStyleDeclaration
"""CSSStyleRule implements DOM Level 2 CSS CSSStyleRule."""
__all__ = ['CSSStyleRule']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
import sys
if sys.version_info[0] >= 3:
string_type = str
else:
string_type = basestring
class CSSStyleRule(cssrule.CSSRule):
"""The CSSStyleRule object represents a ruleset specified (if any) in a CSS
style sheet. It provides access to a declaration block as well as to the
associated group of selectors.
Format::
: selector [ COMMA S* selector ]*
LBRACE S* declaration [ ';' S* declaration ]* '}' S*
;
"""
def __init__(self, selectorText=None, style=None, parentRule=None,
parentStyleSheet=None, readonly=False):
"""
:Parameters:
selectorText
string parsed into selectorList
style
string parsed into CSSStyleDeclaration for this CSSStyleRule
readonly
if True allows setting of properties in constructor only
"""
super(CSSStyleRule, self).__init__(parentRule=parentRule,
parentStyleSheet=parentStyleSheet)
self.selectorList = SelectorList()
if selectorText:
self.selectorText = selectorText
if style:
self.style = style
else:
self.style = CSSStyleDeclaration()
self._readonly = readonly
def __repr__(self):
if self._namespaces:
st = (self.selectorText, self._namespaces)
else:
st = self.selectorText
return "css_parser.css.%s(selectorText=%r, style=%r)" % (
self.__class__.__name__, st, self.style.cssText)
def __str__(self):
return "" % (self.__class__.__name__,
self.selectorText,
self.style.cssText,
self._namespaces,
id(self))
def _getCssText(self):
"""Return serialized property cssText."""
return css_parser.ser.do_CSSStyleRule(self)
def _setCssText(self, cssText):
"""
:param cssText:
a parseable string or a tuple of (cssText, dict-of-namespaces)
:exceptions:
- :exc:`~xml.dom.NamespaceErr`:
Raised if the specified selector uses an unknown namespace
prefix.
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error and
is unparsable.
- :exc:`~xml.dom.InvalidModificationErr`:
Raised if the specified CSS string value represents a different
type of rule than the current one.
- :exc:`~xml.dom.HierarchyRequestErr`:
Raised if the rule cannot be inserted at this point in the
style sheet.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if the rule is readonly.
"""
super(CSSStyleRule, self)._setCssText(cssText)
# might be (cssText, namespaces)
cssText, namespaces = self._splitNamespacesOff(cssText)
try:
# use parent style sheet ones if available
namespaces = self.parentStyleSheet.namespaces
except AttributeError:
pass
tokenizer = self._tokenize2(cssText)
selectortokens = self._tokensupto2(tokenizer, blockstartonly=True)
styletokens = self._tokensupto2(tokenizer, blockendonly=True)
trail = self._nexttoken(tokenizer)
if trail:
self._log.error('CSSStyleRule: Trailing content: %s' %
self._valuestr(cssText), token=trail)
elif not selectortokens:
self._log.error('CSSStyleRule: No selector found: %r' %
self._valuestr(cssText))
elif self._tokenvalue(selectortokens[0]).startswith('@'):
self._log.error('CSSStyleRule: No style rule: %r' %
self._valuestr(cssText),
error=xml.dom.InvalidModificationErr)
else:
newSelectorList = SelectorList(parentRule=self)
newStyle = CSSStyleDeclaration(parentRule=self)
ok = True
bracetoken = selectortokens.pop()
if self._tokenvalue(bracetoken) != '{':
ok = False
self._log.error(
'CSSStyleRule: No start { of style declaration found: %r' %
self._valuestr(cssText), bracetoken)
elif not selectortokens:
ok = False
self._log.error('CSSStyleRule: No selector found: %r.' %
self._valuestr(cssText), bracetoken)
# SET
newSelectorList.selectorText = (selectortokens,
namespaces)
if not styletokens:
ok = False
self._log.error(
'CSSStyleRule: No style declaration or "}" found: %r' %
self._valuestr(cssText))
else:
braceorEOFtoken = styletokens.pop()
val, typ = self._tokenvalue(braceorEOFtoken),\
self._type(braceorEOFtoken)
if val != '}' and typ != 'EOF':
ok = False
self._log.error('CSSStyleRule: No "}" after style '
'declaration found: %r'
% self._valuestr(cssText))
else:
if 'EOF' == typ:
# add again as style needs it
styletokens.append(braceorEOFtoken)
# SET, may raise:
newStyle.cssText = styletokens
if ok:
self.selectorList = newSelectorList
self.style = newStyle
cssText = property(_getCssText, _setCssText,
doc="(DOM) The parsable textual representation of this "
"rule.")
def __getNamespaces(self):
"""Uses children namespaces if not attached to a sheet, else the sheet's
ones."""
try:
return self.parentStyleSheet.namespaces
except AttributeError:
return self.selectorList._namespaces
_namespaces = property(__getNamespaces,
doc="If this Rule is attached to a CSSStyleSheet "
"the namespaces of that sheet are mirrored "
"here. While the Rule is not attached the "
"namespaces of selectorList are used.""")
def _setSelectorList(self, selectorList):
"""
:param selectorList: A SelectorList which replaces the current
selectorList object
"""
self._checkReadonly()
selectorList._parentRule = self
self._selectorList = selectorList
_selectorList = None
selectorList = property(lambda self: self._selectorList, _setSelectorList,
doc="The SelectorList of this rule.")
def _setSelectorText(self, selectorText):
"""
wrapper for css_parser SelectorList object
:param selectorText:
of type string, might also be a comma separated list
of selectors
:exceptions:
- :exc:`~xml.dom.NamespaceErr`:
Raised if the specified selector uses an unknown namespace
prefix.
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error
and is unparsable.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this rule is readonly.
"""
self._checkReadonly()
sl = SelectorList(selectorText=selectorText, parentRule=self)
if sl.wellformed:
self._selectorList = sl
selectorText = property(lambda self: self._selectorList.selectorText,
_setSelectorText,
doc="(DOM) The textual representation of the "
"selector for the rule set.")
def _setStyle(self, style):
"""
:param style: A string or CSSStyleDeclaration which replaces the
current style object.
"""
self._checkReadonly()
if isinstance(style, string_type):
self._style = CSSStyleDeclaration(cssText=style, parentRule=self)
else:
style._parentRule = self
self._style = style
style = property(lambda self: self._style, _setStyle,
doc="(DOM) The declaration-block of this rule set.")
type = property(lambda self: self.STYLE_RULE,
doc="The type of this rule, as defined by a CSSRule "
"type constant.")
wellformed = property(lambda self: self.selectorList.wellformed)
def _getValid(self):
"""Return whether the style declaration is valid."""
return self.style.valid
valid = property(_getValid,
doc='``True`` when the style declaration is true.')
css-parser-1.0.4/src/css_parser/css/cssstylesheet.py 0000644 0001750 0001750 00000102522 13407702010 023043 0 ustar kovid kovid 0000000 0000000 from __future__ import unicode_literals, division, absolute_import, print_function
import xml.dom
import css_parser.stylesheets
from .cssvariablesdeclaration import CSSVariablesDeclaration
from .cssrule import CSSRule
from css_parser.util import _Namespaces, _readUrl
from css_parser.helper import Deprecated
"""CSSStyleSheet implements DOM Level 2 CSS CSSStyleSheet.
Partly also:
- http://dev.w3.org/csswg/cssom/#the-cssstylesheet
- http://www.w3.org/TR/2006/WD-css3-namespace-20060828/
TODO:
- ownerRule and ownerNode
"""
__all__ = ['CSSStyleSheet']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
import sys
if sys.version_info[0] >= 3:
string_type = str
else:
string_type = basestring
def as_list(p):
if isinstance(p, list):
return p
return list(p)
class CSSStyleSheet(css_parser.stylesheets.StyleSheet):
"""CSSStyleSheet represents a CSS style sheet.
Format::
stylesheet
: [ CHARSET_SYM S* STRING S* ';' ]?
[S|CDO|CDC]* [ import [S|CDO|CDC]* ]*
[ namespace [S|CDO|CDC]* ]* # according to @namespace WD
[ [ ruleset | media | page ] [S|CDO|CDC]* ]*
``cssRules``
All Rules in this style sheet, a :class:`~css_parser.css.CSSRuleList`.
"""
def __init__(self, href=None, media=None, title='', disabled=None,
ownerNode=None, parentStyleSheet=None, readonly=False,
ownerRule=None,
validating=True):
"""
For parameters see :class:`~css_parser.stylesheets.StyleSheet`
"""
super(CSSStyleSheet, self).__init__(
'text/css', href, media, title, disabled,
ownerNode, parentStyleSheet,
validating=validating)
self._ownerRule = ownerRule
self.cssRules = css_parser.css.CSSRuleList()
self._namespaces = _Namespaces(parentStyleSheet=self, log=self._log)
self._variables = CSSVariablesDeclaration()
self._readonly = readonly
# used only during setting cssText by parse*()
self.__encodingOverride = None
self._fetcher = None
def __iter__(self):
"Generator which iterates over cssRules."
for rule in self._cssRules:
yield rule
def __repr__(self):
if self.media:
mediaText = self.media.mediaText
else:
mediaText = None
return "css_parser.css.%s(href=%r, media=%r, title=%r)" % (
self.__class__.__name__,
self.href, mediaText, self.title)
def __str__(self):
if self.media:
mediaText = self.media.mediaText
else:
mediaText = None
return "" % (
self.__class__.__name__, self.encoding, self.href,
mediaText, self.title, self.namespaces.namespaces,
id(self))
def _cleanNamespaces(self):
"Remove all namespace rules with same namespaceURI but last."
rules = self.cssRules
namespaceitems = as_list(self.namespaces.items())
i = 0
while i < len(rules):
rule = rules[i]
if rule.type == rule.NAMESPACE_RULE and \
(rule.prefix, rule.namespaceURI) not in namespaceitems:
self.deleteRule(i)
else:
i += 1
def _getUsedURIs(self):
"Return set of URIs used in the sheet."
useduris = set()
for r1 in self:
if r1.STYLE_RULE == r1.type:
useduris.update(r1.selectorList._getUsedUris())
elif r1.MEDIA_RULE == r1.type:
for r2 in r1:
if r2.type == r2.STYLE_RULE:
useduris.update(r2.selectorList._getUsedUris())
return useduris
def _setCssRules(self, cssRules):
"Set new cssRules and update contained rules refs."
cssRules.append = self.insertRule
cssRules.extend = self.insertRule
cssRules.__delitem__ = self.deleteRule
for rule in cssRules:
rule._parentStyleSheet = self
self._cssRules = cssRules
cssRules = property(lambda self: self._cssRules, _setCssRules,
"All Rules in this style sheet, a "
":class:`~css_parser.css.CSSRuleList`.")
def _getCssText(self):
"Textual representation of the stylesheet (a byte string)."
return css_parser.ser.do_CSSStyleSheet(self)
def _setCssText(self, cssText):
"""Parse `cssText` and overwrites the whole stylesheet.
:param cssText:
a parseable string or a tuple of (cssText, dict-of-namespaces)
:exceptions:
- :exc:`~xml.dom.NamespaceErr`:
If a namespace prefix is found which is not declared.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if the rule is readonly.
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error and
is unparsable.
"""
self._checkReadonly()
cssText, namespaces = self._splitNamespacesOff(cssText)
tokenizer = self._tokenize2(cssText)
def S(expected, seq, token, tokenizer=None):
# @charset must be at absolute beginning of style sheet
# or 0 for py3
return max(1, expected or 0)
def COMMENT(expected, seq, token, tokenizer=None):
"special: sets parent*"
self.insertRule(css_parser.css.CSSComment([token], parentStyleSheet=self))
# or 0 for py3
return max(1, expected or 0)
def charsetrule(expected, seq, token, tokenizer):
# parse and consume tokens in any case
rule = css_parser.css.CSSCharsetRule(parentStyleSheet=self)
rule.cssText = self._tokensupto2(tokenizer, token)
if expected > 0:
self._log.error('CSSStylesheet: CSSCharsetRule only allowed '
'at beginning of stylesheet.',
token, xml.dom.HierarchyRequestErr)
return expected
elif rule.wellformed:
self.insertRule(rule)
return 1
def importrule(expected, seq, token, tokenizer):
# parse and consume tokens in any case
rule = css_parser.css.CSSImportRule(parentStyleSheet=self)
rule.cssText = self._tokensupto2(tokenizer, token)
if expected > 1:
self._log.error('CSSStylesheet: CSSImportRule not allowed '
'here.', token, xml.dom.HierarchyRequestErr)
return expected
elif rule.wellformed:
self.insertRule(rule)
return 1
def namespacerule(expected, seq, token, tokenizer):
# parse and consume tokens in any case
rule = css_parser.css.CSSNamespaceRule(
cssText=self._tokensupto2(tokenizer, token), parentStyleSheet=self)
if expected > 2:
self._log.error('CSSStylesheet: CSSNamespaceRule not allowed '
'here.', token, xml.dom.HierarchyRequestErr)
return expected
elif rule.wellformed:
if rule.prefix not in self.namespaces:
# add new if not same prefix
self.insertRule(rule, _clean=False)
else:
# same prefix => replace namespaceURI
for r in self.cssRules.rulesOfType(rule.NAMESPACE_RULE):
if r.prefix == rule.prefix:
r._replaceNamespaceURI(rule.namespaceURI)
self._namespaces[rule.prefix] = rule.namespaceURI
return 2
def variablesrule(expected, seq, token, tokenizer):
# parse and consume tokens in any case
rule = css_parser.css.CSSVariablesRule(parentStyleSheet=self)
rule.cssText = self._tokensupto2(tokenizer, token)
if expected > 2:
self._log.error('CSSStylesheet: CSSVariablesRule not allowed '
'here.', token, xml.dom.HierarchyRequestErr)
return expected
elif rule.wellformed:
self.insertRule(rule)
self._updateVariables()
return 2
def fontfacerule(expected, seq, token, tokenizer):
# parse and consume tokens in any case
rule = css_parser.css.CSSFontFaceRule(parentStyleSheet=self)
rule.cssText = self._tokensupto2(tokenizer, token)
if rule.wellformed:
self.insertRule(rule)
return 3
def mediarule(expected, seq, token, tokenizer):
# parse and consume tokens in any case
rule = css_parser.css.CSSMediaRule(parentStyleSheet=self)
rule.cssText = self._tokensupto2(tokenizer, token)
if rule.wellformed:
self.insertRule(rule)
return 3
def pagerule(expected, seq, token, tokenizer):
# parse and consume tokens in any case
rule = css_parser.css.CSSPageRule(parentStyleSheet=self)
rule.cssText = self._tokensupto2(tokenizer, token)
if rule.wellformed:
self.insertRule(rule)
return 3
def unknownrule(expected, seq, token, tokenizer):
# parse and consume tokens in any case
if token[1] in css_parser.css.MarginRule.margins:
self._log.error('CSSStylesheet: MarginRule out CSSPageRule.',
token, neverraise=True)
rule = css_parser.css.MarginRule(parentStyleSheet=self)
rule.cssText = self._tokensupto2(tokenizer, token)
else:
self._log.warn('CSSStylesheet: Unknown @rule found.',
token, neverraise=True)
rule = css_parser.css.CSSUnknownRule(parentStyleSheet=self)
rule.cssText = self._tokensupto2(tokenizer, token)
if rule.wellformed:
self.insertRule(rule)
# or 0 for py3
return max(1, expected or 0)
def ruleset(expected, seq, token, tokenizer):
# parse and consume tokens in any case
rule = css_parser.css.CSSStyleRule(parentStyleSheet=self)
rule.cssText = self._tokensupto2(tokenizer, token)
if rule.wellformed:
self.insertRule(rule)
return 3
# save for possible reset
oldCssRules = self.cssRules
oldNamespaces = self._namespaces
self.cssRules = css_parser.css.CSSRuleList()
# simple during parse
self._namespaces = namespaces
self._variables = CSSVariablesDeclaration()
# not used?!
newseq = []
# ['CHARSET', 'IMPORT', ('VAR', NAMESPACE'), ('PAGE', 'MEDIA', ruleset)]
wellformed, expected = self._parse(0, newseq, tokenizer,
{'S': S,
'COMMENT': COMMENT,
'CDO': lambda *ignored: None,
'CDC': lambda *ignored: None,
'CHARSET_SYM': charsetrule,
'FONT_FACE_SYM': fontfacerule,
'IMPORT_SYM': importrule,
'NAMESPACE_SYM': namespacerule,
'PAGE_SYM': pagerule,
'MEDIA_SYM': mediarule,
'VARIABLES_SYM': variablesrule,
'ATKEYWORD': unknownrule
},
default=ruleset)
if wellformed:
# use proper namespace object
self._namespaces = _Namespaces(parentStyleSheet=self, log=self._log)
self._cleanNamespaces()
else:
# reset
self._cssRules = oldCssRules
self._namespaces = oldNamespaces
self._updateVariables()
self._cleanNamespaces()
cssText = property(_getCssText, _setCssText,
"Textual representation of the stylesheet (a byte string)")
def _resolveImport(self, url):
"""Read (encoding, enctype, decodedContent) from `url` for @import
sheets."""
try:
# only available during parsing of a complete sheet
parentEncoding = self.__newEncoding
except AttributeError:
try:
# explicit @charset
parentEncoding = self._cssRules[0].encoding
except (IndexError, AttributeError):
# default not UTF-8 but None!
parentEncoding = None
return _readUrl(url, fetcher=self._fetcher,
overrideEncoding=self.__encodingOverride,
parentEncoding=parentEncoding)
def _setCssTextWithEncodingOverride(self, cssText, encodingOverride=None,
encoding=None):
"""Set `cssText` but use `encodingOverride` to overwrite detected
encoding. This is used by parse and @import during setting of cssText.
If `encoding` is given use this but do not save as `encodingOverride`.
"""
if encodingOverride:
# encoding during resolving of @import
self.__encodingOverride = encodingOverride
if encoding:
# save for nested @import
self.__newEncoding = encoding
self.cssText = cssText
if encodingOverride:
# set encodingOverride explicit again!
self.encoding = self.__encodingOverride
# del?
self.__encodingOverride = None
elif encoding:
# may e.g. be httpEncoding
self.encoding = encoding
try:
del self.__newEncoding
except AttributeError:
pass
def _setFetcher(self, fetcher=None):
"""Set @import URL loader, if None the default is used."""
self._fetcher = fetcher
def _getEncoding(self):
"""Encoding set in :class:`~css_parser.css.CSSCharsetRule` or if ``None``
resulting in default ``utf-8`` encoding being used."""
try:
return self._cssRules[0].encoding
except (IndexError, AttributeError):
return 'utf-8'
def _setEncoding(self, encoding):
"""Set `encoding` of charset rule if present in sheet or insert a new
:class:`~css_parser.css.CSSCharsetRule` with given `encoding`.
If `encoding` is None removes charsetrule if present resulting in
default encoding of utf-8.
"""
try:
rule = self._cssRules[0]
except IndexError:
rule = None
if rule and rule.CHARSET_RULE == rule.type:
if encoding:
rule.encoding = encoding
else:
self.deleteRule(0)
elif encoding:
self.insertRule(css_parser.css.CSSCharsetRule(encoding=encoding), 0)
encoding = property(_getEncoding, _setEncoding,
"(css_parser) Reflect encoding of an @charset rule or 'utf-8' "
"(default) if set to ``None``")
namespaces = property(lambda self: self._namespaces,
doc="All Namespaces used in this CSSStyleSheet.")
def _updateVariables(self):
"""Updates self._variables, called when @import or @variables rules
is added to sheet.
"""
for r in self.cssRules.rulesOfType(CSSRule.IMPORT_RULE):
s = r.styleSheet
if s:
for var in s.variables:
self._variables.setVariable(var, s.variables[var])
# for r in self.cssRules.rulesOfType(CSSRule.IMPORT_RULE):
# for vr in r.styleSheet.cssRules.rulesOfType(CSSRule.VARIABLES_RULE):
# for var in vr.variables:
# self._variables.setVariable(var, vr.variables[var])
for vr in self.cssRules.rulesOfType(CSSRule.VARIABLES_RULE):
for var in vr.variables:
self._variables.setVariable(var, vr.variables[var])
variables = property(lambda self: self._variables,
doc="A :class:`css_parser.css.CSSVariablesDeclaration` "
"containing all available variables in this "
"CSSStyleSheet including the ones defined in "
"imported sheets.")
def add(self, rule):
"""Add `rule` to style sheet at appropriate position.
Same as ``insertRule(rule, inOrder=True)``.
"""
return self.insertRule(rule, index=None, inOrder=True)
def deleteRule(self, index):
"""Delete rule at `index` from the style sheet.
:param index:
The `index` of the rule to be removed from the StyleSheet's rule
list. For an `index` < 0 **no** :exc:`~xml.dom.IndexSizeErr` is
raised but rules for normal Python lists are used. E.g.
``deleteRule(-1)`` removes the last rule in cssRules.
`index` may also be a CSSRule object which will then be removed
from the StyleSheet.
:exceptions:
- :exc:`~xml.dom.IndexSizeErr`:
Raised if the specified index does not correspond to a rule in
the style sheet's rule list.
- :exc:`~xml.dom.NamespaceErr`:
Raised if removing this rule would result in an invalid StyleSheet
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this style sheet is readonly.
"""
self._checkReadonly()
if isinstance(index, CSSRule):
for i, r in enumerate(self.cssRules):
if index == r:
index = i
break
else:
raise xml.dom.IndexSizeErr("CSSStyleSheet: Not a rule in"
" this sheets'a cssRules list: %s"
% index)
try:
rule = self._cssRules[index]
except IndexError:
raise xml.dom.IndexSizeErr(
'CSSStyleSheet: %s is not a valid index in the rulelist of '
'length %i' % (index, self._cssRules.length))
else:
if rule.type == rule.NAMESPACE_RULE:
# check all namespacerules if used
uris = [r.namespaceURI for r in self
if r.type == r.NAMESPACE_RULE]
useduris = self._getUsedURIs()
if rule.namespaceURI in useduris and\
uris.count(rule.namespaceURI) == 1:
raise xml.dom.NoModificationAllowedErr(
'CSSStyleSheet: NamespaceURI defined in this rule is '
'used, cannot remove.')
return
rule._parentStyleSheet = None # detach
del self._cssRules[index] # delete from StyleSheet
def insertRule(self, rule, index=None, inOrder=False, _clean=True):
"""
Used to insert a new rule into the style sheet. The new rule now
becomes part of the cascade.
:param rule:
a parsable DOMString, in css_parser also a
:class:`~css_parser.css.CSSRule` or :class:`~css_parser.css.CSSRuleList`
:param index:
of the rule before the new rule will be inserted.
If the specified `index` is equal to the length of the
StyleSheet's rule collection, the rule will be added to the end
of the style sheet.
If `index` is not given or ``None`` rule will be appended to rule
list.
:param inOrder:
if ``True`` the rule will be put to a proper location while
ignoring `index` and without raising
:exc:`~xml.dom.HierarchyRequestErr`.
The resulting index is returned nevertheless.
:returns: The index within the style sheet's rule collection
:Exceptions:
- :exc:`~xml.dom.HierarchyRequestErr`:
Raised if the rule cannot be inserted at the specified `index`
e.g. if an @import rule is inserted after a standard rule set
or other at-rule.
- :exc:`~xml.dom.IndexSizeErr`:
Raised if the specified `index` is not a valid insertion point.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this style sheet is readonly.
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified rule has a syntax error and is
unparsable.
"""
self._checkReadonly()
# check position
if index is None:
index = len(self._cssRules)
elif index < 0 or index > self._cssRules.length:
raise xml.dom.IndexSizeErr(
'CSSStyleSheet: Invalid index %s for CSSRuleList with a '
'length of %s.' % (index, self._cssRules.length))
return
if isinstance(rule, string_type):
# init a temp sheet which has the same properties as self
tempsheet = CSSStyleSheet(href=self.href,
media=self.media,
title=self.title,
parentStyleSheet=self.parentStyleSheet,
ownerRule=self.ownerRule)
tempsheet._ownerNode = self.ownerNode
tempsheet._fetcher = self._fetcher
# prepend encoding if in this sheet to be able to use it in
# @import rules encoding resolution
# do not add if new rule startswith "@charset" (which is exact!)
if not rule.startswith('@charset') and (self._cssRules and
self._cssRules[0].type == self._cssRules[0].CHARSET_RULE):
# rule 0 is @charset!
newrulescount, newruleindex = 2, 1
rule = self._cssRules[0].cssText + rule
else:
newrulescount, newruleindex = 1, 0
# parse the new rule(s)
tempsheet.cssText = (rule, self._namespaces)
if len(tempsheet.cssRules) != newrulescount or (not isinstance(
tempsheet.cssRules[newruleindex], css_parser.css.CSSRule)):
self._log.error('CSSStyleSheet: Not a CSSRule: %s' % rule)
return
rule = tempsheet.cssRules[newruleindex]
rule._parentStyleSheet = None # done later?
# TODO:
# tempsheet._namespaces = self._namespaces
# variables?
elif isinstance(rule, css_parser.css.CSSRuleList):
# insert all rules
for i, r in enumerate(rule):
self.insertRule(r, index + i)
return index
if not rule.wellformed:
self._log.error('CSSStyleSheet: Invalid rules cannot be added.')
return
# CHECK HIERARCHY
# @charset
if rule.type == rule.CHARSET_RULE:
if inOrder:
index = 0
# always first and only
if (self._cssRules
and self._cssRules[0].type == rule.CHARSET_RULE):
self._cssRules[0].encoding = rule.encoding
else:
self._cssRules.insert(0, rule)
elif index != 0 or (self._cssRules and
self._cssRules[0].type == rule.CHARSET_RULE):
self._log.error(
'CSSStylesheet: @charset only allowed once at the'
' beginning of a stylesheet.',
error=xml.dom.HierarchyRequestErr)
return
else:
self._cssRules.insert(index, rule)
# @unknown or comment
elif rule.type in (rule.UNKNOWN_RULE, rule.COMMENT) and not inOrder:
if index == 0 and self._cssRules and\
self._cssRules[0].type == rule.CHARSET_RULE:
self._log.error(
'CSSStylesheet: @charset must be the first rule.',
error=xml.dom.HierarchyRequestErr)
return
else:
self._cssRules.insert(index, rule)
# @import
elif rule.type == rule.IMPORT_RULE:
if inOrder:
# automatic order
if rule.type in (r.type for r in self):
# find last of this type
for i, r in enumerate(reversed(self._cssRules)):
if r.type == rule.type:
index = len(self._cssRules) - i
break
else:
# find first point to insert
if self._cssRules and\
self._cssRules[0].type in (rule.CHARSET_RULE,
rule.COMMENT):
index = 1
else:
index = 0
else:
# after @charset
if index == 0 and self._cssRules and\
self._cssRules[0].type == rule.CHARSET_RULE:
self._log.error(
'CSSStylesheet: Found @charset at index 0.',
error=xml.dom.HierarchyRequestErr)
return
# before @namespace @variables @page @font-face @media stylerule
for r in self._cssRules[:index]:
if r.type in (r.NAMESPACE_RULE,
r.VARIABLES_RULE,
r.MEDIA_RULE,
r.PAGE_RULE,
r.STYLE_RULE,
r.FONT_FACE_RULE):
self._log.error(
'CSSStylesheet: Cannot insert @import here,'
' found @namespace, @variables, @media, @page or'
' CSSStyleRule before index %s.' %
index,
error=xml.dom.HierarchyRequestErr)
return
self._cssRules.insert(index, rule)
self._updateVariables()
# @namespace
elif rule.type == rule.NAMESPACE_RULE:
if inOrder:
if rule.type in (r.type for r in self):
# find last of this type
for i, r in enumerate(reversed(self._cssRules)):
if r.type == rule.type:
index = len(self._cssRules) - i
break
else:
# find first point to insert
for i, r in enumerate(self._cssRules):
if r.type in (r.VARIABLES_RULE, r.MEDIA_RULE,
r.PAGE_RULE, r.STYLE_RULE,
r.FONT_FACE_RULE, r.UNKNOWN_RULE,
r.COMMENT):
index = i # before these
break
else:
# after @charset and @import
for r in self._cssRules[index:]:
if r.type in (r.CHARSET_RULE, r.IMPORT_RULE):
self._log.error(
'CSSStylesheet: Cannot insert @namespace here,'
' found @charset or @import after index %s.' %
index,
error=xml.dom.HierarchyRequestErr)
return
# before @variables @media @page @font-face and stylerule
for r in self._cssRules[:index]:
if r.type in (r.VARIABLES_RULE,
r.MEDIA_RULE,
r.PAGE_RULE,
r.STYLE_RULE,
r.FONT_FACE_RULE):
self._log.error(
'CSSStylesheet: Cannot insert @namespace here,'
' found @variables, @media, @page or CSSStyleRule'
' before index %s.' %
index,
error=xml.dom.HierarchyRequestErr)
return
if not (rule.prefix in self.namespaces and
self.namespaces[rule.prefix] == rule.namespaceURI):
# no doublettes
self._cssRules.insert(index, rule)
if _clean:
self._cleanNamespaces()
# @variables
elif rule.type == rule.VARIABLES_RULE:
if inOrder:
if rule.type in (r.type for r in self):
# find last of this type
for i, r in enumerate(reversed(self._cssRules)):
if r.type == rule.type:
index = len(self._cssRules) - i
break
else:
# find first point to insert
for i, r in enumerate(self._cssRules):
if r.type in (r.MEDIA_RULE,
r.PAGE_RULE,
r.STYLE_RULE,
r.FONT_FACE_RULE,
r.UNKNOWN_RULE,
r.COMMENT):
index = i # before these
break
else:
# after @charset @import @namespace
for r in self._cssRules[index:]:
if r.type in (r.CHARSET_RULE,
r.IMPORT_RULE,
r.NAMESPACE_RULE):
self._log.error(
'CSSStylesheet: Cannot insert @variables here,'
' found @charset, @import or @namespace after'
' index %s.' %
index,
error=xml.dom.HierarchyRequestErr)
return
# before @media @page @font-face and stylerule
for r in self._cssRules[:index]:
if r.type in (r.MEDIA_RULE,
r.PAGE_RULE,
r.STYLE_RULE,
r.FONT_FACE_RULE):
self._log.error(
'CSSStylesheet: Cannot insert @variables here,'
' found @media, @page or CSSStyleRule'
' before index %s.' %
index,
error=xml.dom.HierarchyRequestErr)
return
self._cssRules.insert(index, rule)
self._updateVariables()
# all other where order is not important
else:
if inOrder:
# simply add to end as no specific order
self._cssRules.append(rule)
index = len(self._cssRules) - 1
else:
for r in self._cssRules[index:]:
if r.type in (r.CHARSET_RULE,
r.IMPORT_RULE,
r.NAMESPACE_RULE):
self._log.error(
'CSSStylesheet: Cannot insert rule here, found '
'@charset, @import or @namespace before index %s.'
% index, error=xml.dom.HierarchyRequestErr)
return
self._cssRules.insert(index, rule)
# post settings
rule._parentStyleSheet = self
if rule.IMPORT_RULE == rule.type and not rule.hrefFound:
# try loading the imported sheet which has new relative href now
rule.href = rule.href
return index
ownerRule = property(lambda self: self._ownerRule,
doc='A ref to an @import rule if it is imported, '
'else ``None``.')
def _getValid(self):
"""Check if each contained rule is valid."""
for rule in self.cssRules:
# Not all rules can be checked for validity
if hasattr(rule, 'valid') and not rule.valid:
return False
return True
valid = property(_getValid,
doc='``True`` if all contained rules are valid')
@Deprecated('Use ``css_parser.setSerializer(serializer)`` instead.')
def setSerializer(self, cssserializer):
"""Set the css_parser global Serializer used for all output."""
if isinstance(cssserializer, css_parser.CSSSerializer):
css_parser.ser = cssserializer
else:
raise ValueError('Serializer must be an instance of '
'css_parser.CSSSerializer.')
@Deprecated('Set pref in ``css_parser.ser.prefs`` instead.')
def setSerializerPref(self, pref, value):
"""Set a Preference of CSSSerializer used for output.
See :class:`css_parser.serialize.Preferences` for possible
preferences to be set.
"""
css_parser.ser.prefs.__setattr__(pref, value)
css-parser-1.0.4/src/css_parser/css/cssunknownrule.py 0000644 0001750 0001750 00000021042 13407702010 023236 0 ustar kovid kovid 0000000 0000000 """CSSUnknownRule implements DOM Level 2 CSS CSSUnknownRule."""
from __future__ import unicode_literals, division, absolute_import, print_function
__all__ = ['CSSUnknownRule']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
from . import cssrule
import css_parser
import xml.dom
class CSSUnknownRule(cssrule.CSSRule):
"""
Represents an at-rule not supported by this user agent, so in
effect all other at-rules not defined in css_parser.
Format::
@xxx until ';' or block {...}
"""
def __init__(self, cssText='', parentRule=None,
parentStyleSheet=None, readonly=False):
"""
:param cssText:
of type string
"""
super(CSSUnknownRule, self).__init__(parentRule=parentRule,
parentStyleSheet=parentStyleSheet)
self._atkeyword = None
if cssText:
self.cssText = cssText
self._readonly = readonly
def __repr__(self):
return "css_parser.css.%s(cssText=%r)" % (
self.__class__.__name__,
self.cssText)
def __str__(self):
return "" % (
self.__class__.__name__,
self.cssText,
id(self))
def _getCssText(self):
"""Return serialized property cssText."""
return css_parser.ser.do_CSSUnknownRule(self)
def _setCssText(self, cssText):
"""
:exceptions:
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error and
is unparsable.
- :exc:`~xml.dom.InvalidModificationErr`:
Raised if the specified CSS string value represents a different
type of rule than the current one.
- :exc:`~xml.dom.HierarchyRequestErr`:
Raised if the rule cannot be inserted at this point in the
style sheet.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if the rule is readonly.
"""
super(CSSUnknownRule, self)._setCssText(cssText)
tokenizer = self._tokenize2(cssText)
attoken = self._nexttoken(tokenizer, None)
if not attoken or self._type(attoken) != self._prods.ATKEYWORD:
self._log.error('CSSUnknownRule: No CSSUnknownRule found: %s' %
self._valuestr(cssText),
error=xml.dom.InvalidModificationErr)
else:
# for closures: must be a mutable
new = {'nesting': [], # {} [] or ()
'wellformed': True
}
def CHAR(expected, seq, token, tokenizer=None):
type_, val, line, col = token
if expected != 'EOF':
if val in '{[(':
new['nesting'].append(val)
elif val in '}])':
opening = {'}': '{', ']': '[', ')': '('}[val]
try:
if new['nesting'][-1] == opening:
new['nesting'].pop()
else:
raise IndexError()
except IndexError:
new['wellformed'] = False
self._log.error('CSSUnknownRule: Wrong nesting of '
'{, [ or (.', token=token)
if val in '};' and not new['nesting']:
expected = 'EOF'
seq.append(val, type_, line=line, col=col)
return expected
else:
new['wellformed'] = False
self._log.error('CSSUnknownRule: Expected end of rule.',
token=token)
return expected
def FUNCTION(expected, seq, token, tokenizer=None):
# handled as opening (
type_, val, line, col = token
val = self._tokenvalue(token)
if expected != 'EOF':
new['nesting'].append('(')
seq.append(val, type_, line=line, col=col)
return expected
else:
new['wellformed'] = False
self._log.error('CSSUnknownRule: Expected end of rule.',
token=token)
return expected
def EOF(expected, seq, token, tokenizer=None):
"close all blocks and return 'EOF'"
for x in reversed(new['nesting']):
closing = {'{': '}', '[': ']', '(': ')'}[x]
seq.append(closing, closing)
new['nesting'] = []
return 'EOF'
def INVALID(expected, seq, token, tokenizer=None):
# makes rule invalid
self._log.error('CSSUnknownRule: Bad syntax.',
token=token, error=xml.dom.SyntaxErr)
new['wellformed'] = False
return expected
def STRING(expected, seq, token, tokenizer=None):
type_, val, line, col = token
val = self._stringtokenvalue(token)
if expected != 'EOF':
seq.append(val, type_, line=line, col=col)
return expected
else:
new['wellformed'] = False
self._log.error('CSSUnknownRule: Expected end of rule.',
token=token)
return expected
def URI(expected, seq, token, tokenizer=None):
type_, val, line, col = token
val = self._uritokenvalue(token)
if expected != 'EOF':
seq.append(val, type_, line=line, col=col)
return expected
else:
new['wellformed'] = False
self._log.error('CSSUnknownRule: Expected end of rule.',
token=token)
return expected
def default(expected, seq, token, tokenizer=None):
type_, val, line, col = token
if expected != 'EOF':
seq.append(val, type_, line=line, col=col)
return expected
else:
new['wellformed'] = False
self._log.error('CSSUnknownRule: Expected end of rule.',
token=token)
return expected
# unknown : ATKEYWORD S* ... ; | }
newseq = self._tempSeq()
wellformed, expected = self._parse(expected=None,
seq=newseq, tokenizer=tokenizer,
productions={'CHAR': CHAR,
'EOF': EOF,
'FUNCTION': FUNCTION,
'INVALID': INVALID,
'STRING': STRING,
'URI': URI,
'S': default # overwrite default default!
},
default=default,
new=new)
# wellformed set by parse
wellformed = wellformed and new['wellformed']
# post conditions
if expected != 'EOF':
wellformed = False
self._log.error('CSSUnknownRule: No ending ";" or "}" found: '
'%r' % self._valuestr(cssText))
elif new['nesting']:
wellformed = False
self._log.error('CSSUnknownRule: Unclosed "{", "[" or "(": %r'
% self._valuestr(cssText))
# set all
if wellformed:
self.atkeyword = self._tokenvalue(attoken)
self._setSeq(newseq)
cssText = property(fget=_getCssText, fset=_setCssText,
doc="(DOM) The parsable textual representation.")
type = property(lambda self: self.UNKNOWN_RULE,
doc="The type of this rule, as defined by a CSSRule "
"type constant.")
wellformed = property(lambda self: bool(self.atkeyword))
css-parser-1.0.4/src/css_parser/css/cssvalue.py 0000644 0001750 0001750 00000146673 13407702010 022005 0 ustar kovid kovid 0000000 0000000 from __future__ import unicode_literals, division, absolute_import, print_function
import xml.dom
import re
import math
import css_parser.helper
import css_parser
from css_parser.prodparser import Choice, Sequence, PreDef, Prod, ProdParser
"""CSSValue related classes
- CSSValue implements DOM Level 2 CSS CSSValue
- CSSPrimitiveValue implements DOM Level 2 CSS CSSPrimitiveValue
- CSSValueList implements DOM Level 2 CSS CSSValueList
"""
__all__ = ['CSSValue', 'CSSPrimitiveValue', 'CSSValueList', 'RGBColor',
'CSSVariable']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
import sys
if sys.version_info[0] >= 3:
text_type = str
string_type = str
else:
text_type = unicode
string_type = basestring
class CSSValue(css_parser.util._NewBase):
"""The CSSValue interface represents a simple or a complex value.
A CSSValue object only occurs in a context of a CSS property.
"""
# The value is inherited and the cssText contains "inherit".
CSS_INHERIT = 0
# The value is a CSSPrimitiveValue.
CSS_PRIMITIVE_VALUE = 1
# The value is a CSSValueList.
CSS_VALUE_LIST = 2
# The value is a custom value.
CSS_CUSTOM = 3
# The value is a CSSVariable.
CSS_VARIABLE = 4
_typestrings = {0: 'CSS_INHERIT',
1: 'CSS_PRIMITIVE_VALUE',
2: 'CSS_VALUE_LIST',
3: 'CSS_CUSTOM',
4: 'CSS_VARIABLE'}
def __init__(self, cssText=None, parent=None, readonly=False):
"""
:param cssText:
the parsable cssText of the value
:param readonly:
defaults to False
"""
super(CSSValue, self).__init__()
self._cssValueType = None
self.wellformed = False
self.parent = parent
if cssText is not None: # may be 0
if isinstance(cssText, int):
cssText = text_type(cssText) # if it is an integer
elif isinstance(cssText, float):
cssText = '%f' % cssText # if it is a floating point number
self.cssText = cssText
self._readonly = readonly
def __repr__(self):
return "css_parser.css.%s(%r)" % (
self.__class__.__name__, self.cssText)
def __str__(self):
return "" % (self.__class__.__name__,
self.cssValueTypeString,
self.cssText,
id(self))
def _setCssText(self, cssText):
"""
Format::
unary_operator
: '-' | '+'
;
operator
: '/' S* | ',' S* | /* empty */
;
expr
: term [ operator term ]*
;
term
: unary_operator?
[ NUMBER S* | PERCENTAGE S* | LENGTH S* | EMS S* | EXS S* |
ANGLE S* | TIME S* | FREQ S* ]
| STRING S* | IDENT S* | URI S* | hexcolor | function
| UNICODE-RANGE S*
;
function
: FUNCTION S* expr ')' S*
;
/*
* There is a constraint on the color that it must
* have either 3 or 6 hex-digits (i.e., [0-9a-fA-F])
* after the "#"; e.g., "#000" is OK, but "#abcd" is not.
*/
hexcolor
: HASH S*
;
:exceptions:
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error
(according to the attached property) or is unparsable.
- :exc:`~xml.dom.InvalidModificationErr`:
TODO: Raised if the specified CSS string value represents a
different type of values than the values allowed by the CSS
property.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this value is readonly.
"""
self._checkReadonly()
# used as operator is , / or S
nextSor = ',/'
term = Choice(Sequence(PreDef.unary(),
Choice(PreDef.number(nextSor=nextSor),
PreDef.percentage(nextSor=nextSor),
PreDef.dimension(nextSor=nextSor))),
PreDef.string(nextSor=nextSor),
PreDef.ident(nextSor=nextSor),
PreDef.uri(nextSor=nextSor),
PreDef.hexcolor(nextSor=nextSor),
PreDef.unicode_range(nextSor=nextSor),
# special case IE only expression
Prod(name='expression',
match=lambda t, v: t == self._prods.FUNCTION and (
css_parser.helper.normalize(v) in (
'expression(',
'alpha(',
'blur(',
'chroma(',
'dropshadow(',
'fliph(',
'flipv(',
'glow(',
'gray(',
'invert(',
'mask(',
'shadow(',
'wave(',
'xray(') or
v.startswith('progid:DXImageTransform.Microsoft.')
),
nextSor=nextSor,
toSeq=lambda t, tokens: (ExpressionValue._functionName,
ExpressionValue(
css_parser.helper.pushtoken(t, tokens),
parent=self)
)
),
# CSS Variable var(
PreDef.variable(nextSor=nextSor,
toSeq=lambda t, tokens: ('CSSVariable',
CSSVariable(
css_parser.helper.pushtoken(t, tokens),
parent=self)
)
),
# calc(
PreDef.calc(nextSor=nextSor,
toSeq=lambda t, tokens: (CalcValue._functionName,
CalcValue(
css_parser.helper.pushtoken(t, tokens),
parent=self)
)
),
# TODO:
# # rgb/rgba(
# Prod(name='RGBColor',
# match=lambda t, v: t == self._prods.FUNCTION and (
# css_parser.helper.normalize(v) in (u'rgb(',
# u'rgba('
# )
# ),
# nextSor=nextSor,
# toSeq=lambda t, tokens: (RGBColor._functionName,
# RGBColor(
# css_parser.helper.pushtoken(t, tokens),
# parent=self)
# )
# ),
# other functions like rgb( etc
PreDef.function(nextSor=nextSor,
toSeq=lambda t, tokens: ('FUNCTION',
CSSFunction(
css_parser.helper.pushtoken(t, tokens),
parent=self)
)
)
)
operator = Choice(PreDef.S(),
PreDef.char('comma', ',',
toSeq=lambda t, tokens: ('operator', t[1])),
PreDef.char('slash', '/',
toSeq=lambda t, tokens: ('operator', t[1])),
optional=True)
# CSSValue PRODUCTIONS
valueprods = Sequence(term,
Sequence(operator, # mayEnd this Sequence if whitespace
# TODO: only when setting via other class
# used by variabledeclaration currently
PreDef.char('END', ';',
stopAndKeep=True,
optional=True),
term,
minmax=lambda: (0, None)))
# parse
wellformed, seq, store, notused = ProdParser().parse(cssText,
'CSSValue',
valueprods,
keepS=True)
if wellformed:
# - count actual values and set firstvalue which is used later on
# - combine comma separated list, e.g. font-family to a single item
# - remove S which should be an operator but is no needed
count, firstvalue = 0, ()
newseq = self._tempSeq()
i, end = 0, len(seq)
while i < end:
item = seq[i]
if item.type == self._prods.S:
pass
elif (item.value, item.type) == (',', 'operator'):
# , separared counts as a single STRING for now
# URI or STRING value might be a single CHAR too!
newseq.appendItem(item)
count -= 1
if firstvalue:
# list of IDENTs is handled as STRING!
if firstvalue[1] == self._prods.IDENT:
firstvalue = firstvalue[0], 'STRING'
elif item.value == '/':
# / separated items count as one
newseq.appendItem(item)
elif item.value == '-' or item.value == '+':
# combine +- and following number or other
i += 1
try:
next = seq[i]
except IndexError:
firstvalue = () # raised later
break
newval = item.value + next.value
newseq.append(newval, next.type,
item.line, item.col)
if not firstvalue:
firstvalue = (newval, next.type)
count += 1
elif item.type != css_parser.css.CSSComment:
newseq.appendItem(item)
if not firstvalue:
firstvalue = (item.value, item.type)
count += 1
else:
newseq.appendItem(item)
i += 1
if not firstvalue:
self._log.error(
'CSSValue: Unknown syntax or no value: %r.' %
self._valuestr(cssText))
else:
# ok and set
self._setSeq(newseq)
self.wellformed = wellformed
if hasattr(self, '_value'):
# only in case of CSSPrimitiveValue, else remove!
del self._value
if count == 1:
# inherit, primitive or variable
# Under Pythn 2.x this was basestring but ...
if isinstance(firstvalue[0], string_type) and\
'inherit' == css_parser.helper.normalize(firstvalue[0]):
self.__class__ = CSSValue
self._cssValueType = CSSValue.CSS_INHERIT
elif 'CSSVariable' == firstvalue[1]:
self.__class__ = CSSVariable
self._value = firstvalue
# TODO: remove major hack!
self._name = firstvalue[0]._name
else:
self.__class__ = CSSPrimitiveValue
self._value = firstvalue
elif count > 1:
# valuelist
self.__class__ = CSSValueList
# change items in list to specific type (primitive etc)
newseq = self._tempSeq()
commalist = []
nexttocommalist = False
def itemValue(item):
"Reserialized simple item.value"
if self._prods.STRING == item.type:
return css_parser.helper.string(item.value)
elif self._prods.URI == item.type:
return css_parser.helper.uri(item.value)
elif self._prods.FUNCTION == item.type or\
'CSSVariable' == item.type:
return item.value.cssText
else:
return item.value
def saveifcommalist(commalist, newseq):
"""
saves items in commalist to seq and items
if anything in there
"""
if commalist:
newseq.replace(-1,
CSSPrimitiveValue(cssText=''.join(
commalist)),
CSSPrimitiveValue,
newseq[-1].line,
newseq[-1].col)
del commalist[:]
for i, item in enumerate(self._seq):
if issubclass(type(item.value), CSSValue):
# set parent of CSSValueList items to the lists
# parent
item.value.parent = self.parent
if item.type in (self._prods.DIMENSION,
self._prods.FUNCTION,
self._prods.HASH,
self._prods.IDENT,
self._prods.NUMBER,
self._prods.PERCENTAGE,
self._prods.STRING,
self._prods.URI,
self._prods.UNICODE_RANGE,
'CSSVariable'):
if nexttocommalist:
# wait until complete
commalist.append(itemValue(item))
else:
saveifcommalist(commalist, newseq)
# append new item
if hasattr(item.value, 'cssText'):
newseq.append(item.value,
item.value.__class__,
item.line, item.col)
else:
newseq.append(CSSPrimitiveValue(
itemValue(item)),
CSSPrimitiveValue,
item.line, item.col)
nexttocommalist = False
elif ',' == item.value:
if not commalist:
# save last item to commalist
commalist.append(itemValue(self._seq[i - 1]))
commalist.append(',')
nexttocommalist = True
else:
if nexttocommalist:
commalist.append(item.value.cssText)
else:
newseq.appendItem(item)
saveifcommalist(commalist, newseq)
self._setSeq(newseq)
else:
# should not happen...
self.__class__ = CSSValue
self._cssValueType = CSSValue.CSS_CUSTOM
cssText = property(lambda self: css_parser.ser.do_css_CSSValue(self),
_setCssText,
doc="A string representation of the current value.")
cssValueType = property(lambda self: self._cssValueType,
doc="A (readonly) code defining the type of the value.")
cssValueTypeString = property(
lambda self: CSSValue._typestrings.get(self.cssValueType, None),
doc="(readonly) Name of cssValueType.")
class CSSPrimitiveValue(CSSValue):
"""Represents a single CSS Value. May be used to determine the value of a
specific style property currently set in a block or to set a specific
style property explicitly within the block. Might be obtained from the
getPropertyCSSValue method of CSSStyleDeclaration.
Conversions are allowed between absolute values (from millimeters to
centimeters, from degrees to radians, and so on) but not between
relative values. (For example, a pixel value cannot be converted to a
centimeter value.) Percentage values can't be converted since they are
relative to the parent value (or another property value). There is one
exception for color percentage values: since a color percentage value
is relative to the range 0-255, a color percentage value can be
converted to a number; (see also the RGBColor interface).
"""
# constant: type of this CSSValue class
cssValueType = CSSValue.CSS_PRIMITIVE_VALUE
__types = css_parser.cssproductions.CSSProductions
# An integer indicating which type of unit applies to the value.
CSS_UNKNOWN = 0 # only obtainable via cssText
CSS_NUMBER = 1
CSS_PERCENTAGE = 2
CSS_EMS = 3
CSS_EXS = 4
CSS_PX = 5
CSS_CM = 6
CSS_MM = 7
CSS_IN = 8
CSS_PT = 9
CSS_PC = 10
CSS_DEG = 11
CSS_RAD = 12
CSS_GRAD = 13
CSS_MS = 14
CSS_S = 15
CSS_HZ = 16
CSS_KHZ = 17
CSS_DIMENSION = 18
CSS_STRING = 19
CSS_URI = 20
CSS_IDENT = 21
CSS_ATTR = 22
CSS_COUNTER = 23
CSS_RECT = 24
CSS_RGBCOLOR = 25
# NOT OFFICIAL:
CSS_RGBACOLOR = 26
CSS_UNICODE_RANGE = 27
_floattypes = (CSS_NUMBER, CSS_PERCENTAGE, CSS_EMS, CSS_EXS,
CSS_PX, CSS_CM, CSS_MM, CSS_IN, CSS_PT, CSS_PC,
CSS_DEG, CSS_RAD, CSS_GRAD, CSS_MS, CSS_S,
CSS_HZ, CSS_KHZ, CSS_DIMENSION)
_stringtypes = (CSS_ATTR, CSS_IDENT, CSS_STRING, CSS_URI)
_countertypes = (CSS_COUNTER,)
_recttypes = (CSS_RECT,)
_rbgtypes = (CSS_RGBCOLOR, CSS_RGBACOLOR)
_lengthtypes = (CSS_NUMBER, CSS_EMS, CSS_EXS,
CSS_PX, CSS_CM, CSS_MM, CSS_IN, CSS_PT, CSS_PC)
# oldtype: newType: converterfunc
_converter = {
# cm <-> mm <-> in, 1 inch is equal to 2.54 centimeters.
# pt <-> pc, the points used by CSS 2.1 are equal to 1/72nd of an inch.
# pc: picas - 1 pica is equal to 12 points
(CSS_CM, CSS_MM): lambda x: x * 10,
(CSS_MM, CSS_CM): lambda x: x / 10,
(CSS_PT, CSS_PC): lambda x: x * 12,
(CSS_PC, CSS_PT): lambda x: x / 12,
(CSS_CM, CSS_IN): lambda x: x / 2.54,
(CSS_IN, CSS_CM): lambda x: x * 2.54,
(CSS_MM, CSS_IN): lambda x: x / 25.4,
(CSS_IN, CSS_MM): lambda x: x * 25.4,
(CSS_IN, CSS_PT): lambda x: x / 72,
(CSS_PT, CSS_IN): lambda x: x * 72,
(CSS_CM, CSS_PT): lambda x: x / 2.54 / 72,
(CSS_PT, CSS_CM): lambda x: x * 72 * 2.54,
(CSS_MM, CSS_PT): lambda x: x / 25.4 / 72,
(CSS_PT, CSS_MM): lambda x: x * 72 * 25.4,
(CSS_IN, CSS_PC): lambda x: x / 72 / 12,
(CSS_PC, CSS_IN): lambda x: x * 12 * 72,
(CSS_CM, CSS_PC): lambda x: x / 2.54 / 72 / 12,
(CSS_PC, CSS_CM): lambda x: x * 12 * 72 * 2.54,
(CSS_MM, CSS_PC): lambda x: x / 25.4 / 72 / 12,
(CSS_PC, CSS_MM): lambda x: x * 12 * 72 * 25.4,
# hz <-> khz
(CSS_KHZ, CSS_HZ): lambda x: x * 1000,
(CSS_HZ, CSS_KHZ): lambda x: x / 1000,
# s <-> ms
(CSS_S, CSS_MS): lambda x: x * 1000,
(CSS_MS, CSS_S): lambda x: x / 1000,
(CSS_RAD, CSS_DEG): lambda x: math.degrees(x),
(CSS_DEG, CSS_RAD): lambda x: math.radians(x),
# TODO: convert grad <-> deg or rad
# (CSS_RAD, CSS_GRAD): lambda x: math.degrees(x),
# (CSS_DEG, CSS_GRAD): lambda x: math.radians(x),
# (CSS_GRAD, CSS_RAD): lambda x: math.radians(x),
# (CSS_GRAD, CSS_DEG): lambda x: math.radians(x)
}
def __init__(self, cssText=None, parent=None, readonly=False):
"""See CSSPrimitiveValue.__init__()"""
super(CSSPrimitiveValue, self).__init__(cssText=cssText,
parent=parent,
readonly=readonly)
def __str__(self):
return ""\
% (self.__class__.__name__,
self.primitiveTypeString,
self.cssText,
id(self))
_unitnames = ['CSS_UNKNOWN',
'CSS_NUMBER', 'CSS_PERCENTAGE',
'CSS_EMS', 'CSS_EXS',
'CSS_PX',
'CSS_CM', 'CSS_MM',
'CSS_IN',
'CSS_PT', 'CSS_PC',
'CSS_DEG', 'CSS_RAD', 'CSS_GRAD',
'CSS_MS', 'CSS_S',
'CSS_HZ', 'CSS_KHZ',
'CSS_DIMENSION',
'CSS_STRING', 'CSS_URI', 'CSS_IDENT',
'CSS_ATTR', 'CSS_COUNTER', 'CSS_RECT',
'CSS_RGBCOLOR', 'CSS_RGBACOLOR',
'CSS_UNICODE_RANGE'
]
_reNumDim = re.compile(r'([+-]?\d*\.\d+|[+-]?\d+)(.*)$', re.I | re.U | re.X)
def _unitDIMENSION(value):
"""Check val for dimension name."""
units = {'em': 'CSS_EMS', 'ex': 'CSS_EXS',
'px': 'CSS_PX',
'cm': 'CSS_CM', 'mm': 'CSS_MM',
'in': 'CSS_IN',
'pt': 'CSS_PT', 'pc': 'CSS_PC',
'deg': 'CSS_DEG', 'rad': 'CSS_RAD', 'grad': 'CSS_GRAD',
'ms': 'CSS_MS', 's': 'CSS_S',
'hz': 'CSS_HZ', 'khz': 'CSS_KHZ'
}
val, dim = CSSPrimitiveValue._reNumDim.findall(css_parser.helper.normalize(value))[0]
return units.get(dim, 'CSS_DIMENSION')
def _unitFUNCTION(value):
"""Check val for function name."""
units = {'attr(': 'CSS_ATTR',
'counter(': 'CSS_COUNTER',
'rect(': 'CSS_RECT',
'rgb(': 'CSS_RGBCOLOR',
'rgba(': 'CSS_RGBACOLOR',
}
return units.get(re.findall(r'^(.*?\()',
css_parser.helper.normalize(value.cssText),
re.U)[0],
'CSS_UNKNOWN')
__unitbytype = {
__types.NUMBER: 'CSS_NUMBER',
__types.PERCENTAGE: 'CSS_PERCENTAGE',
__types.STRING: 'CSS_STRING',
__types.UNICODE_RANGE: 'CSS_UNICODE_RANGE',
__types.URI: 'CSS_URI',
__types.IDENT: 'CSS_IDENT',
__types.HASH: 'CSS_RGBCOLOR',
__types.DIMENSION: _unitDIMENSION,
__types.FUNCTION: _unitFUNCTION
}
def __set_primitiveType(self):
"""primitiveType is readonly but is set lazy if accessed"""
# TODO: check unary and font-family STRING a, b, "c"
val, type_ = self._value
# try get by type_
pt = self.__unitbytype.get(type_, 'CSS_UNKNOWN')
if callable(pt):
# multiple options, check value too
pt = pt(val)
self._primitiveType = getattr(self, pt)
def _getPrimitiveType(self):
if not hasattr(self, '_primitivetype'):
self.__set_primitiveType()
return self._primitiveType
primitiveType = property(_getPrimitiveType,
doc="(readonly) The type of the value as defined "
"by the constants in this class.")
def _getPrimitiveTypeString(self):
return self._unitnames[self.primitiveType]
primitiveTypeString = property(_getPrimitiveTypeString,
doc="Name of primitive type of this value.")
def _getCSSPrimitiveTypeString(self, type):
"get TypeString by given type which may be unknown, used by setters"
try:
return self._unitnames[type]
except (IndexError, TypeError):
return '%r (UNKNOWN TYPE)' % type
def _getNumDim(self, value=None):
"Split self._value in numerical and dimension part."
if value is None:
value = css_parser.helper.normalize(self._value[0])
try:
val, dim = CSSPrimitiveValue._reNumDim.findall(value)[0]
except IndexError:
val, dim = value, ''
try:
val = float(val)
if val == int(val):
val = int(val)
except ValueError:
raise xml.dom.InvalidAccessErr(
'CSSPrimitiveValue: No float value %r' % self._value[0])
return val, dim
def getFloatValue(self, unitType=None):
"""(DOM) This method is used to get a float value in a
specified unit. If this CSS value doesn't contain a float value
or can't be converted into the specified unit, a DOMException
is raised.
:param unitType:
to get the float value. The unit code can only be a float unit type
(i.e. CSS_NUMBER, CSS_PERCENTAGE, CSS_EMS, CSS_EXS, CSS_PX, CSS_CM,
CSS_MM, CSS_IN, CSS_PT, CSS_PC, CSS_DEG, CSS_RAD, CSS_GRAD, CSS_MS,
CSS_S, CSS_HZ, CSS_KHZ, CSS_DIMENSION) or None in which case
the current dimension is used.
:returns:
not necessarily a float but some cases just an integer
e.g. if the value is ``1px`` it return ``1`` and **not** ``1.0``
Conversions might return strange values like 1.000000000001
"""
if unitType is not None and unitType not in self._floattypes:
raise xml.dom.InvalidAccessErr(
'unitType Parameter is not a float type')
val, dim = self._getNumDim()
if unitType is not None and self.primitiveType != unitType:
# convert if needed
try:
val = self._converter[self.primitiveType, unitType](val)
except KeyError:
raise xml.dom.InvalidAccessErr(
'CSSPrimitiveValue: Cannot coerce primitiveType %r to %r'
% (self.primitiveTypeString,
self._getCSSPrimitiveTypeString(unitType)))
if val == int(val):
val = int(val)
return val
def setFloatValue(self, unitType, floatValue):
"""(DOM) A method to set the float value with a specified unit.
If the property attached with this value can not accept the
specified unit or the float value, the value will be unchanged and
a DOMException will be raised.
:param unitType:
a unit code as defined above. The unit code can only be a float
unit type
:param floatValue:
the new float value which does not have to be a float value but
may simple be an int e.g. if setting::
setFloatValue(CSS_PX, 1)
:exceptions:
- :exc:`~xml.dom.InvalidAccessErr`:
Raised if the attached property doesn't
support the float value or the unit type.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this property is readonly.
"""
self._checkReadonly()
if unitType not in self._floattypes:
raise xml.dom.InvalidAccessErr(
'CSSPrimitiveValue: unitType %r is not a float type' %
self._getCSSPrimitiveTypeString(unitType))
try:
val = float(floatValue)
except ValueError:
raise xml.dom.InvalidAccessErr(
'CSSPrimitiveValue: floatValue %r is not a float' %
floatValue)
oldval, dim = self._getNumDim()
if self.primitiveType != unitType:
# convert if possible
try:
val = self._converter[unitType, self.primitiveType](val)
except KeyError:
raise xml.dom.InvalidAccessErr(
'CSSPrimitiveValue: Cannot coerce primitiveType %r to %r'
% (self.primitiveTypeString,
self._getCSSPrimitiveTypeString(unitType)))
if val == int(val):
val = int(val)
self.cssText = '%s%s' % (val, dim)
def getStringValue(self):
"""(DOM) This method is used to get the string value. If the
CSS value doesn't contain a string value, a DOMException is raised.
Some properties (like 'font-family' or 'voice-family')
convert a whitespace separated list of idents to a string.
Only the actual value is returned so e.g. all the following return the
actual value ``a``: url(a), attr(a), "a", 'a'
"""
if self.primitiveType not in self._stringtypes:
raise xml.dom.InvalidAccessErr(
'CSSPrimitiveValue %r is not a string type'
% self.primitiveTypeString)
if CSSPrimitiveValue.CSS_ATTR == self.primitiveType:
return self._value[0].cssText[5:-1]
else:
return self._value[0]
def setStringValue(self, stringType, stringValue):
"""(DOM) A method to set the string value with the specified
unit. If the property attached to this value can't accept the
specified unit or the string value, the value will be unchanged and
a DOMException will be raised.
:param stringType:
a string code as defined above. The string code can only be a
string unit type (i.e. CSS_STRING, CSS_URI, CSS_IDENT, and
CSS_ATTR).
:param stringValue:
the new string value
Only the actual value is expected so for (CSS_URI, "a") the
new value will be ``url(a)``. For (CSS_STRING, "'a'")
the new value will be ``"\\'a\\'"`` as the surrounding ``'`` are
not part of the string value
:exceptions:
- :exc:`~xml.dom.InvalidAccessErr`:
Raised if the CSS value doesn't contain a
string value or if the string value can't be converted into
the specified unit.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this property is readonly.
"""
self._checkReadonly()
# self not stringType
if self.primitiveType not in self._stringtypes:
raise xml.dom.InvalidAccessErr(
'CSSPrimitiveValue %r is not a string type'
% self.primitiveTypeString)
# given stringType is no StringType
if stringType not in self._stringtypes:
raise xml.dom.InvalidAccessErr(
'CSSPrimitiveValue: stringType %s is not a string type'
% self._getCSSPrimitiveTypeString(stringType))
if self._primitiveType != stringType:
raise xml.dom.InvalidAccessErr(
'CSSPrimitiveValue: Cannot coerce primitiveType %r to %r'
% (self.primitiveTypeString,
self._getCSSPrimitiveTypeString(stringType)))
if CSSPrimitiveValue.CSS_STRING == self._primitiveType:
self.cssText = css_parser.helper.string(stringValue)
elif CSSPrimitiveValue.CSS_URI == self._primitiveType:
self.cssText = css_parser.helper.uri(stringValue)
elif CSSPrimitiveValue.CSS_ATTR == self._primitiveType:
self.cssText = 'attr(%s)' % stringValue
else:
self.cssText = stringValue
self._primitiveType = stringType
def getCounterValue(self):
"""(DOM) This method is used to get the Counter value. If
this CSS value doesn't contain a counter value, a DOMException
is raised. Modification to the corresponding style property
can be achieved using the Counter interface.
**Not implemented.**
"""
if not self.CSS_COUNTER == self.primitiveType:
raise xml.dom.InvalidAccessErr('Value is not a counter type')
# TODO: use Counter class
raise NotImplementedError()
def getRGBColorValue(self):
"""(DOM) This method is used to get the RGB color. If this
CSS value doesn't contain a RGB color value, a DOMException
is raised. Modification to the corresponding style property
can be achieved using the RGBColor interface.
"""
if self.primitiveType not in self._rbgtypes:
raise xml.dom.InvalidAccessErr('Value is not a RGBColor value')
return RGBColor(self._value[0])
def getRectValue(self):
"""(DOM) This method is used to get the Rect value. If this CSS
value doesn't contain a rect value, a DOMException is raised.
Modification to the corresponding style property can be achieved
using the Rect interface.
**Not implemented.**
"""
if self.primitiveType not in self._recttypes:
raise xml.dom.InvalidAccessErr('value is not a Rect value')
# TODO: use Rect class
raise NotImplementedError()
def _getCssText(self):
"""Overwrites CSSValue."""
return css_parser.ser.do_css_CSSPrimitiveValue(self)
def _setCssText(self, cssText):
"""Use CSSValue."""
return super(CSSPrimitiveValue, self)._setCssText(cssText)
cssText = property(_getCssText, _setCssText,
doc="A string representation of the current value.")
class CSSValueList(CSSValue):
"""The CSSValueList interface provides the abstraction of an ordered
collection of CSS values.
Some properties allow an empty list into their syntax. In that case,
these properties take the none identifier. So, an empty list means
that the property has the value none.
The items in the CSSValueList are accessible via an integral index,
starting from 0.
"""
cssValueType = CSSValue.CSS_VALUE_LIST
def __init__(self, cssText=None, parent=None, readonly=False):
"""Init a new CSSValueList"""
super(CSSValueList, self).__init__(cssText=cssText,
parent=parent,
readonly=readonly)
self._items = []
def __iter__(self):
"CSSValueList is iterable."
for item in self.__items():
yield item.value
def __str__(self):
return "" % (self.__class__.__name__,
self.cssValueTypeString,
self.cssText,
self.length,
id(self))
def __items(self):
return [item for item in self._seq
if isinstance(item.value, CSSValue)]
def item(self, index):
"""(DOM) Retrieve a CSSValue by ordinal `index`. The
order in this collection represents the order of the values in the
CSS style property. If `index` is greater than or equal to the number
of values in the list, this returns ``None``.
"""
try:
return self.__items()[index].value
except IndexError:
return None
length = property(lambda self: len(self.__items()),
doc="(DOM attribute) The number of CSSValues in the "
"list.")
class CSSFunction(CSSPrimitiveValue):
"""A CSS function value like rect() etc."""
_functionName = 'CSSFunction'
primitiveType = CSSPrimitiveValue.CSS_UNKNOWN
def __init__(self, cssText=None, parent=None, readonly=False):
"""
Init a new CSSFunction
:param cssText:
the parsable cssText of the value
:param readonly:
defaults to False
"""
super(CSSFunction, self).__init__(parent=parent)
self._funcType = None
self.valid = False
self.wellformed = False
if cssText is not None:
self.cssText = cssText
self._readonly = readonly
def _productiondefinition(self):
"""Return definition used for parsing."""
types = self._prods # rename!
value = Sequence(PreDef.unary(),
Prod(name='PrimitiveValue',
match=lambda t, v: t in (types.DIMENSION,
types.HASH,
types.IDENT,
types.NUMBER,
types.PERCENTAGE,
types.STRING),
toSeq=lambda t, tokens: (t[0],
CSSPrimitiveValue(t[1]))
)
)
valueOrFunc = Choice(value,
# FUNC is actually not in spec but used in e.g. Prince
PreDef.function(toSeq=lambda t,
tokens: ('FUNCTION',
CSSFunction(
css_parser.helper.pushtoken(t, tokens))
)
)
)
funcProds = Sequence(Prod(name='FUNC',
match=lambda t, v: t == types.FUNCTION,
toSeq=lambda t, tokens: (t[0], css_parser.helper.normalize(t[1]))),
Choice(Sequence(valueOrFunc,
# more values starting with Comma
# should use store where colorType is saved to
# define min and may, closure?
Sequence(PreDef.comma(),
valueOrFunc,
minmax=lambda: (0, None)),
PreDef.funcEnd(stop=True)),
PreDef.funcEnd(stop=True))
)
return funcProds
def _setCssText(self, cssText):
self._checkReadonly()
# store: colorType, parts
wellformed, seq, store, unusedtokens = ProdParser().parse(cssText,
self._functionName,
self._productiondefinition(),
keepS=True)
if wellformed:
# combine +/- and following CSSPrimitiveValue, remove S
newseq = self._tempSeq()
i, end = 0, len(seq)
while i < end:
item = seq[i]
if item.type == self._prods.S:
pass
elif item.value == '+' or item.value == '-':
i += 1
next = seq[i]
newval = next.value
if isinstance(newval, CSSPrimitiveValue):
newval.setFloatValue(newval.primitiveType,
float(item.value + str(newval.getFloatValue())))
newseq.append(newval, next.type,
item.line, item.col)
else:
# expressions only?
newseq.appendItem(item)
newseq.appendItem(next)
else:
newseq.appendItem(item)
i += 1
self.wellformed = True
self._setSeq(newseq)
self._funcType = newseq[0].value
cssText = property(lambda self: css_parser.ser.do_css_FunctionValue(self),
_setCssText)
funcType = property(lambda self: self._funcType)
class RGBColor(CSSFunction):
"""A CSS color like RGB, RGBA or a simple value like `#000` or `red`."""
_functionName = 'Function rgb()'
def __init__(self, cssText=None, parent=None, readonly=False):
"""
Init a new RGBColor
:param cssText:
the parsable cssText of the value
:param readonly:
defaults to False
"""
super(CSSFunction, self).__init__(parent=parent)
self._colorType = None
self.valid = False
self.wellformed = False
if cssText is not None:
try:
# if it is a Function object
cssText = cssText.cssText
except AttributeError:
pass
self.cssText = cssText
self._readonly = readonly
def __repr__(self):
return "css_parser.css.%s(%r)" % (
self.__class__.__name__, self.cssText)
def __str__(self):
return "" % (
self.__class__.__name__,
self.colorType,
self.cssText,
id(self))
def _setCssText(self, cssText):
self._checkReadonly()
types = self._prods # rename!
valueProd = Prod(name='value',
match=lambda t, v: t in (types.NUMBER, types.PERCENTAGE),
toSeq=lambda t, v: (CSSPrimitiveValue, CSSPrimitiveValue(v)),
toStore='parts'
)
# COLOR PRODUCTION
funccolor = Sequence(Prod(name='FUNC',
match=lambda t, v: t == types.FUNCTION and css_parser.helper.normalize(
v) in ('rgb(', 'rgba(', 'hsl(', 'hsla('),
toSeq=lambda t, v: (t, v), # css_parser.helper.normalize(v)),
toStore='colorType'),
PreDef.unary(),
valueProd,
# 2 or 3 more values starting with Comma
Sequence(PreDef.comma(),
PreDef.unary(),
valueProd,
minmax=lambda: (2, 3)),
PreDef.funcEnd()
)
colorprods = Choice(funccolor,
PreDef.hexcolor('colorType'),
Prod(name='named color',
match=lambda t, v: t == types.IDENT,
toStore='colorType'
)
)
# store: colorType, parts
wellformed, seq, store, unusedtokens = ProdParser().parse(cssText,
'RGBColor',
colorprods,
keepS=True,
store={'parts': []})
if wellformed:
self.wellformed = True
if store['colorType'].type == self._prods.HASH:
self._colorType = 'HEX'
elif store['colorType'].type == self._prods.IDENT:
self._colorType = 'Named Color'
else:
self._colorType = store['colorType'].value[:-1]
# self._colorType = css_parser.helper.normalize(store['colorType'].value)[:-1]
self._setSeq(seq)
cssText = property(lambda self: css_parser.ser.do_css_RGBColor(self),
_setCssText)
colorType = property(lambda self: self._colorType)
class CalcValue(CSSFunction):
"""Calc Function"""
_functionName = 'Function calc()'
def _productiondefinition(self):
"""Return defintion used for parsing."""
types = self._prods # rename!
def toSeq(t, tokens):
"Do not normalize function name!"
return t[0], t[1]
funcProds = Sequence(Prod(name='calc',
match=lambda t, v: t == types.FUNCTION,
toSeq=toSeq
),
Sequence(Choice(Prod(name='nested function',
match=lambda t, v: t == self._prods.FUNCTION,
toSeq=lambda t, tokens: (
CSSFunction._functionName,
CSSFunction(
css_parser.helper.pushtoken(t, tokens)))
),
Prod(name='part',
match=lambda t, v: v != ')',
toSeq=lambda t, tokens: (t[0], t[1])),
),
minmax=lambda: (0, None)),
PreDef.funcEnd(stop=True))
return funcProds
def _getCssText(self):
return css_parser.ser.do_css_CalcValue(self)
def _setCssText(self, cssText):
return super(CalcValue, self)._setCssText(cssText)
cssText = property(_getCssText, _setCssText,
doc="A string representation of the current value.")
class ExpressionValue(CSSFunction):
"""Special IE only CSSFunction which may contain *anything*.
Used for expressions and ``alpha(opacity=100)`` currently."""
_functionName = 'Expression (IE only)'
def _productiondefinition(self):
"""Return defintion used for parsing."""
types = self._prods # rename!
def toSeq(t, tokens):
"Do not normalize function name!"
return t[0], t[1]
funcProds = Sequence(Prod(name='expression',
match=lambda t, v: t == types.FUNCTION,
toSeq=toSeq
),
Sequence(Choice(Prod(name='nested function',
match=lambda t, v: t == self._prods.FUNCTION,
toSeq=lambda t, tokens: (
ExpressionValue._functionName,
ExpressionValue(css_parser.helper.pushtoken(t, tokens)))
),
Prod(name='part',
match=lambda t, v: v != ')',
toSeq=lambda t, tokens: (t[0], t[1])),
),
minmax=lambda: (0, None)),
PreDef.funcEnd(stop=True))
return funcProds
def _getCssText(self):
return css_parser.ser.do_css_ExpressionValue(self)
def _setCssText(self, cssText):
# self._log.warn(u'CSSValue: Unoffial and probably invalid MS value used!')
return super(ExpressionValue, self)._setCssText(cssText)
cssText = property(_getCssText, _setCssText,
doc="A string representation of the current value.")
class CSSVariable(CSSValue):
"""The CSSVariable represents a call to CSS Variable."""
def __init__(self, cssText=None, parent=None, readonly=False):
"""Init a new CSSVariable.
:param cssText:
the parsable cssText of the value, e.g. ``var(x)``
:param readonly:
defaults to False
"""
self._name = None
super(CSSVariable, self).__init__(cssText=cssText,
parent=parent,
readonly=readonly)
def __repr__(self):
return "css_parser.css.%s(%r)" % (self.__class__.__name__, self.cssText)
def __str__(self):
return "" % (
self.__class__.__name__,
self.name,
self.value,
id(self))
def _setCssText(self, cssText):
self._checkReadonly()
types = self._prods # rename!
funcProds = Sequence(Prod(name='var',
match=lambda t, v: t == types.FUNCTION
),
PreDef.ident(toStore='ident'),
PreDef.funcEnd(stop=True))
# store: name of variable
store = {'ident': None}
wellformed, seq, store, unusedtokens = ProdParser().parse(cssText,
'CSSVariable',
funcProds,
keepS=True)
if wellformed:
self._name = store['ident'].value
self._setSeq(seq)
self.wellformed = True
cssText = property(lambda self: css_parser.ser.do_css_CSSVariable(self),
_setCssText,
doc="A string representation of the current variable.")
cssValueType = CSSValue.CSS_VARIABLE
# TODO: writable? check if var (value) available?
name = property(lambda self: self._name)
def _getValue(self):
"Find contained sheet and @variables there"
try:
variables = self.parent.parent.parentRule.parentStyleSheet.variables
except AttributeError:
return None
else:
try:
return variables[self.name]
except KeyError:
return None
value = property(_getValue)
css-parser-1.0.4/src/css_parser/css/cssvariablesdeclaration.py 0000644 0001750 0001750 00000027730 13407702010 025037 0 ustar kovid kovid 0000000 0000000 """CSSVariablesDeclaration
http://disruptive-innovations.com/zoo/cssvariables/#mozTocId496530
"""
from __future__ import unicode_literals, division, absolute_import, print_function
__all__ = ['CSSVariablesDeclaration']
__docformat__ = 'restructuredtext'
__version__ = '$Id: cssstyledeclaration.py 1819 2009-08-01 20:52:43Z cthedot $'
from css_parser.prodparser import Prod, Sequence, PreDef, ProdParser
from css_parser.helper import normalize
from .value import PropertyValue
import css_parser
import itertools
def as_list(p):
if isinstance(p, list):
return p
return list(p)
class CSSVariablesDeclaration(css_parser.util._NewBase):
"""The CSSVariablesDeclaration interface represents a single block of
variable declarations.
"""
def __init__(self, cssText='', parentRule=None, readonly=False):
"""
:param cssText:
Shortcut, sets CSSVariablesDeclaration.cssText
:param parentRule:
The CSS rule that contains this declaration block or
None if this CSSVariablesDeclaration is not attached to a CSSRule.
:param readonly:
defaults to False
Format::
variableset
: vardeclaration [ ';' S* vardeclaration ]* S*
;
vardeclaration
: varname ':' S* term
;
varname
: IDENT S*
;
"""
super(CSSVariablesDeclaration, self).__init__()
self._parentRule = parentRule
self._vars = {}
if cssText:
self.cssText = cssText
self._readonly = readonly
def __repr__(self):
return "css_parser.css.%s(cssText=%r)" % (self.__class__.__name__, self.cssText)
def __str__(self):
return "" % (
self.__class__.__name__,
self.length,
id(self))
def __contains__(self, variableName):
"""Check if a variable is in variable declaration block.
:param variableName:
a string
"""
return normalize(variableName) in as_list(self.keys())
def __getitem__(self, variableName):
"""Retrieve the value of variable ``variableName`` from this
declaration.
"""
return self.getVariableValue(variableName)
def __setitem__(self, variableName, value):
self.setVariable(variableName, value)
def __delitem__(self, variableName):
return self.removeVariable(variableName)
def __iter__(self):
"""Iterator of names of set variables."""
for name in as_list(self.keys()):
yield name
def keys(self):
"""Analoguous to standard dict returns variable names which are set in
this declaration."""
return as_list(self._vars.keys())
def _getCssText(self):
"""Return serialized property cssText."""
return css_parser.ser.do_css_CSSVariablesDeclaration(self)
def _setCssText(self, cssText):
"""Setting this attribute will result in the parsing of the new value
and resetting of all the properties in the declaration block
including the removal or addition of properties.
:exceptions:
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this declaration is readonly or a property is readonly.
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error and
is unparsable.
Format::
variableset
: vardeclaration [ ';' S* vardeclaration ]*
;
vardeclaration
: varname ':' S* term
;
varname
: IDENT S*
;
expr
: [ VARCALL | term ] [ operator [ VARCALL | term ] ]*
;
"""
self._checkReadonly()
vardeclaration = Sequence(
PreDef.ident(),
PreDef.char(':', ':', toSeq=False, optional=True),
# PreDef.S(toSeq=False, optional=True),
Prod(name='term', match=lambda t, v: True,
toSeq=lambda t, tokens: ('value',
PropertyValue(itertools.chain([t],
tokens),
parent=self)
)
)
)
prods = Sequence(vardeclaration,
Sequence(PreDef.S(optional=True),
PreDef.char(';', ';', toSeq=False, optional=True),
PreDef.S(optional=True),
vardeclaration,
minmax=lambda: (0, None)),
PreDef.S(optional=True),
PreDef.char(';', ';', toSeq=False, optional=True)
)
# parse
wellformed, seq, store, notused = \
ProdParser().parse(cssText,
'CSSVariableDeclaration',
prods,
emptyOk=True)
if wellformed:
newseq = self._tempSeq()
newvars = {}
# seq contains only name: value pairs plus comments etc
nameitem = None
for item in seq:
if 'IDENT' == item.type:
nameitem = item
elif 'value' == item.type:
nname = normalize(nameitem.value)
if nname in newvars:
# replace var with same name
for i, it in enumerate(newseq):
if normalize(it.value[0]) == nname:
newseq.replace(i,
(nameitem.value, item.value),
'var',
nameitem.line, nameitem.col)
else:
# saved non normalized name for reserialization
newseq.append((nameitem.value, item.value),
'var',
nameitem.line, nameitem.col)
# newseq.append((nameitem.value, item.value),
# 'var',
# nameitem.line, nameitem.col)
newvars[nname] = item.value
else:
newseq.appendItem(item)
self._setSeq(newseq)
self._vars = newvars
self.wellformed = True
cssText = property(_getCssText, _setCssText,
doc="(DOM) A parsable textual representation of the declaration "
"block excluding the surrounding curly braces.")
def _setParentRule(self, parentRule):
self._parentRule = parentRule
parentRule = property(lambda self: self._parentRule, _setParentRule,
doc="(DOM) The CSS rule that contains this"
" declaration block or None if this block"
" is not attached to a CSSRule.")
def getVariableValue(self, variableName):
"""Used to retrieve the value of a variable if it has been explicitly
set within this variable declaration block.
:param variableName:
The name of the variable.
:returns:
the value of the variable if it has been explicitly set in this
variable declaration block. Returns the empty string if the
variable has not been set.
"""
try:
return self._vars[normalize(variableName)].cssText
except KeyError:
return ''
def removeVariable(self, variableName):
"""Used to remove a variable if it has been explicitly set within this
variable declaration block.
:param variableName:
The name of the variable.
:returns:
the value of the variable if it has been explicitly set for this
variable declaration block. Returns the empty string if the
variable has not been set.
:exceptions:
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this declaration is readonly is readonly.
"""
normalname = variableName
try:
r = self._vars[normalname]
except KeyError:
return ''
else:
self.seq._readonly = False
if normalname in self._vars:
for i, x in enumerate(self.seq):
if x.value[0] == variableName:
del self.seq[i]
self.seq._readonly = True
del self._vars[normalname]
return r.cssText
def setVariable(self, variableName, value):
"""Used to set a variable value within this variable declaration block.
:param variableName:
The name of the CSS variable.
:param value:
The new value of the variable, may also be a PropertyValue object.
:exceptions:
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified value has a syntax error and is
unparsable.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this declaration is readonly or the property is
readonly.
"""
self._checkReadonly()
# check name
wellformed, seq, store, unused = \
ProdParser().parse(normalize(variableName),
'variableName',
Sequence(PreDef.ident()))
if not wellformed:
self._log.error('Invalid variableName: %r: %r'
% (variableName, value))
else:
# check value
if isinstance(value, PropertyValue):
v = value
else:
v = PropertyValue(cssText=value, parent=self)
if not v.wellformed:
self._log.error('Invalid variable value: %r: %r'
% (variableName, value))
else:
# update seq
self.seq._readonly = False
variableName = normalize(variableName)
if variableName in self._vars:
for i, x in enumerate(self.seq):
if x.value[0] == variableName:
self.seq.replace(i,
[variableName, v],
x.type,
x.line,
x.col)
break
else:
self.seq.append([variableName, v], 'var')
self.seq._readonly = True
self._vars[variableName] = v
def item(self, index):
"""Used to retrieve the variables that have been explicitly set in
this variable declaration block. The order of the variables
retrieved using this method does not have to be the order in which
they were set. This method can be used to iterate over all variables
in this variable declaration block.
:param index:
of the variable name to retrieve, negative values behave like
negative indexes on Python lists, so -1 is the last element
:returns:
The name of the variable at this ordinal position. The empty
string if no variable exists at this position.
"""
try:
return as_list(self.keys())[index]
except IndexError:
return ''
length = property(lambda self: len(self._vars),
doc="The number of variables that have been explicitly set in this"
" variable declaration block. The range of valid indices is 0"
" to length-1 inclusive.")
css-parser-1.0.4/src/css_parser/css/cssvariablesrule.py 0000644 0001750 0001750 00000017265 13407702010 023523 0 ustar kovid kovid 0000000 0000000 from __future__ import unicode_literals, division, absolute_import, print_function
import xml.dom
import css_parser
from . import cssrule
from .cssvariablesdeclaration import CSSVariablesDeclaration
"""CSSVariables implements (and only partly) experimental
`CSS Variables `_
"""
__all__ = ['CSSVariablesRule']
__docformat__ = 'restructuredtext'
__version__ = '$Id: cssfontfacerule.py 1818 2009-07-30 21:39:00Z cthedot $'
import sys
if sys.version_info[0] >= 3:
string_type = str
else:
string_type = basestring
class CSSVariablesRule(cssrule.CSSRule):
"""
The CSSVariablesRule interface represents a @variables rule within a CSS
style sheet. The @variables rule is used to specify variables.
css_parser uses a :class:`~css_parser.css.CSSVariablesDeclaration` to
represent the variables.
Format::
variables
VARIABLES_SYM S* medium [ COMMA S* medium ]* LBRACE S*
variableset* '}' S*
;
for variableset see :class:`css_parser.css.CSSVariablesDeclaration`
**Media are not implemented. Reason is that css_parser is using CSS
variables in a kind of preprocessing and therefor no media information
is available at this stage. For now do not use media!**
Example::
@variables {
CorporateLogoBGColor: #fe8d12;
}
div.logoContainer {
background-color: var(CorporateLogoBGColor);
}
"""
def __init__(self, mediaText=None, variables=None, parentRule=None,
parentStyleSheet=None, readonly=False):
"""
If readonly allows setting of properties in constructor only.
"""
super(CSSVariablesRule, self).__init__(parentRule=parentRule,
parentStyleSheet=parentStyleSheet)
self._atkeyword = '@variables'
# dummy
self._media = css_parser.stylesheets.MediaList(mediaText, readonly=readonly)
if variables:
self.variables = variables
else:
self.variables = CSSVariablesDeclaration(parentRule=self)
self._readonly = readonly
def __repr__(self):
return "css_parser.css.%s(mediaText=%r, variables=%r)" % (
self.__class__.__name__,
self._media.mediaText,
self.variables.cssText)
def __str__(self):
return "" % (self.__class__.__name__,
self._media.mediaText,
self.variables.cssText,
self.valid,
id(self))
def _getCssText(self):
"""Return serialized property cssText."""
return css_parser.ser.do_CSSVariablesRule(self)
def _setCssText(self, cssText):
"""
:exceptions:
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error and
is unparsable.
- :exc:`~xml.dom.InvalidModificationErr`:
Raised if the specified CSS string value represents a different
type of rule than the current one.
- :exc:`~xml.dom.HierarchyRequestErr`:
Raised if the rule cannot be inserted at this point in the
style sheet.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if the rule is readonly.
Format::
variables
: VARIABLES_SYM S* medium [ COMMA S* medium ]* LBRACE S*
variableset* '}' S*
;
variableset
: LBRACE S* vardeclaration [ ';' S* vardeclaration ]* '}' S*
;
"""
super(CSSVariablesRule, self)._setCssText(cssText)
tokenizer = self._tokenize2(cssText)
attoken = self._nexttoken(tokenizer, None)
if self._type(attoken) != self._prods.VARIABLES_SYM:
self._log.error('CSSVariablesRule: No CSSVariablesRule found: %s' %
self._valuestr(cssText),
error=xml.dom.InvalidModificationErr)
else:
newVariables = CSSVariablesDeclaration(parentRule=self)
ok = True
beforetokens, brace = self._tokensupto2(tokenizer,
blockstartonly=True,
separateEnd=True)
if self._tokenvalue(brace) != '{':
ok = False
self._log.error('CSSVariablesRule: No start { of variable '
'declaration found: %r'
% self._valuestr(cssText), brace)
# parse stuff before { which should be comments and S only
new = {'wellformed': True}
newseq = self._tempSeq() # []
beforewellformed, expected = self._parse(expected=':',
seq=newseq, tokenizer=self._tokenize2(beforetokens),
productions={})
ok = ok and beforewellformed and new['wellformed']
variablestokens, braceorEOFtoken = self._tokensupto2(tokenizer,
blockendonly=True,
separateEnd=True)
val, type_ = self._tokenvalue(braceorEOFtoken), \
self._type(braceorEOFtoken)
if val != '}' and type_ != 'EOF':
ok = False
self._log.error('CSSVariablesRule: No "}" after variables '
'declaration found: %r'
% self._valuestr(cssText))
nonetoken = self._nexttoken(tokenizer)
if nonetoken:
ok = False
self._log.error('CSSVariablesRule: Trailing content found.',
token=nonetoken)
if 'EOF' == type_:
# add again as variables needs it
variablestokens.append(braceorEOFtoken)
# SET but may raise:
newVariables.cssText = variablestokens
if ok:
# contains probably comments only upto {
self._setSeq(newseq)
self.variables = newVariables
cssText = property(_getCssText, _setCssText,
doc="(DOM) The parsable textual representation of this "
"rule.")
media = property(doc="NOT IMPLEMENTED! As css_parser resolves variables "
"during serializing media information is lost.")
def _setVariables(self, variables):
"""
:param variables:
a CSSVariablesDeclaration or string
"""
self._checkReadonly()
# Under Pythoin 2.x this was basestring but ...
if isinstance(variables, string_type):
self._variables = CSSVariablesDeclaration(cssText=variables,
parentRule=self)
else:
variables._parentRule = self
self._variables = variables
variables = property(lambda self: self._variables, _setVariables,
doc="(DOM) The variables of this rule set, a "
":class:`css_parser.css.CSSVariablesDeclaration`.")
type = property(lambda self: self.VARIABLES_RULE,
doc="The type of this rule, as defined by a CSSRule "
"type constant.")
valid = property(lambda self: True, doc='NOT IMPLEMTED REALLY (TODO)')
# constant but needed:
wellformed = property(lambda self: True)
css-parser-1.0.4/src/css_parser/css/marginrule.py 0000644 0001750 0001750 00000016755 13407702010 022322 0 ustar kovid kovid 0000000 0000000 from __future__ import unicode_literals, division, absolute_import, print_function
import xml.dom
import css_parser
from . import cssrule
from .cssstyledeclaration import CSSStyleDeclaration
from css_parser.prodparser import Prod, PreDef, Sequence, Choice, ProdParser
"""MarginRule implements DOM Level 2 CSS MarginRule."""
__all__ = ['MarginRule']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
import sys
if sys.version_info[0] >= 3:
string_type = str
else:
string_type = basestring
class MarginRule(cssrule.CSSRule):
"""
A margin at-rule consists of an ATKEYWORD that identifies the margin box
(e.g. '@top-left') and a block of declarations (said to be in the margin
context).
Format::
margin :
margin_sym S* '{' declaration [ ';' S* declaration? ]* '}' S*
;
margin_sym :
TOPLEFTCORNER_SYM |
TOPLEFT_SYM |
TOPCENTER_SYM |
TOPRIGHT_SYM |
TOPRIGHTCORNER_SYM |
BOTTOMLEFTCORNER_SYM |
BOTTOMLEFT_SYM |
BOTTOMCENTER_SYM |
BOTTOMRIGHT_SYM |
BOTTOMRIGHTCORNER_SYM |
LEFTTOP_SYM |
LEFTMIDDLE_SYM |
LEFTBOTTOM_SYM |
RIGHTTOP_SYM |
RIGHTMIDDLE_SYM |
RIGHTBOTTOM_SYM
;
e.g.::
@top-left {
content: "123";
}
"""
margins = ['@top-left-corner',
'@top-left',
'@top-center',
'@top-right',
'@top-right-corner',
'@bottom-left-corner',
'@bottom-left',
'@bottom-center',
'@bottom-right',
'@bottom-right-corner',
'@left-top',
'@left-middle',
'@left-bottom',
'@right-top',
'@right-middle',
'@right-bottom'
]
def __init__(self, margin=None, style=None, parentRule=None,
parentStyleSheet=None, readonly=False):
"""
:param atkeyword:
The margin area, e.g. '@top-left' for this rule
:param style:
CSSStyleDeclaration for this MarginRule
"""
super(MarginRule, self).__init__(parentRule=parentRule,
parentStyleSheet=parentStyleSheet)
self._atkeyword = self._keyword = None
if margin:
self.margin = margin
if style:
self.style = style
else:
self.style = CSSStyleDeclaration(parentRule=self)
self._readonly = readonly
def _setMargin(self, margin):
"""Check if new keyword fits the rule it is used for."""
n = self._normalize(margin)
if n not in MarginRule.margins:
self._log.error('Invalid margin @keyword for this %s rule: %r' %
(self.margin, margin),
error=xml.dom.InvalidModificationErr)
else:
self._atkeyword = n
self._keyword = margin
margin = property(lambda self: self._atkeyword, _setMargin,
doc="Margin area of parent CSSPageRule. "
"`margin` and `atkeyword` are both normalized "
"@keyword of the @rule.")
atkeyword = margin
def __repr__(self):
return "css_parser.css.%s(margin=%r, style=%r)" % (
self.__class__.__name__, self.margin, self.style.cssText)
def __str__(self):
return "" % (self.__class__.__name__,
self.margin,
self.style.cssText,
id(self))
def _getCssText(self):
"""Return serialized property cssText."""
return css_parser.ser.do_MarginRule(self)
def _setCssText(self, cssText):
"""
:exceptions:
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error and
is unparsable.
- :exc:`~xml.dom.InvalidModificationErr`:
Raised if the specified CSS string value represents a different
type of rule than the current one.
- :exc:`~xml.dom.HierarchyRequestErr`:
Raised if the rule cannot be inserted at this point in the
style sheet.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if the rule is readonly.
"""
super(MarginRule, self)._setCssText(cssText)
# TEMP: all style tokens are saved in store to fill styledeclaration
# TODO: resolve when all generators
styletokens = Prod(name='styletokens',
match=lambda t, v: v != '}',
# toSeq=False,
toStore='styletokens',
storeToken=True
)
prods = Sequence(Prod(name='@ margin',
match=lambda t, v:
t == 'ATKEYWORD' and
self._normalize(v) in MarginRule.margins,
toStore='margin'
# TODO?
# , exception=xml.dom.InvalidModificationErr
),
PreDef.char('OPEN', '{'),
Sequence(Choice(PreDef.unknownrule(toStore='@'),
styletokens),
minmax=lambda: (0, None)
),
PreDef.char('CLOSE', '}', stopAndKeep=True)
)
# parse
ok, seq, store, unused = ProdParser().parse(cssText,
'MarginRule',
prods)
if ok:
# TODO: use seq for serializing instead of fixed stuff?
self._setSeq(seq)
if 'margin' in store:
# may raise:
self.margin = store['margin'].value
else:
self._log.error('No margin @keyword for this %s rule' %
self.margin,
error=xml.dom.InvalidModificationErr)
# new empty style
self.style = CSSStyleDeclaration(parentRule=self)
if 'styletokens' in store:
# may raise:
self.style.cssText = store['styletokens']
cssText = property(fget=_getCssText, fset=_setCssText,
doc="(DOM) The parsable textual representation.")
def _setStyle(self, style):
"""
:param style: A string or CSSStyleDeclaration which replaces the
current style object.
"""
self._checkReadonly()
if isinstance(style, string_type):
self._style = CSSStyleDeclaration(cssText=style, parentRule=self)
else:
style._parentRule = self
self._style = style
style = property(lambda self: self._style, _setStyle,
doc="(DOM) The declaration-block of this rule set.")
type = property(lambda self: self.MARGIN_RULE,
doc="The type of this rule, as defined by a CSSRule "
"type constant.")
wellformed = property(lambda self: bool(self.atkeyword))
css-parser-1.0.4/src/css_parser/css/property.py 0000644 0001750 0001750 00000044753 13407702010 022040 0 ustar kovid kovid 0000000 0000000 from __future__ import unicode_literals, division, absolute_import, print_function
import css_parser
from .value import PropertyValue
from css_parser.helper import Deprecated
"""Property is a single CSS property in a CSSStyleDeclaration."""
__all__ = ['Property']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
import sys
if sys.version_info[0] >= 3:
string_type = str
else:
string_type = basestring
class Property(css_parser.util.Base):
"""A CSS property in a StyleDeclaration of a CSSStyleRule (css_parser).
Format::
property = name
: IDENT S*
;
expr = value
: term [ operator term ]*
;
term
: unary_operator?
[ NUMBER S* | PERCENTAGE S* | LENGTH S* | EMS S* | EXS S* |
ANGLE S* | TIME S* | FREQ S* | function ]
| STRING S* | IDENT S* | URI S* | hexcolor
;
function
: FUNCTION S* expr ')' S*
;
/*
* There is a constraint on the color that it must
* have either 3 or 6 hex-digits (i.e., [0-9a-fA-F])
* after the "#"; e.g., "#000" is OK, but "#abcd" is not.
*/
hexcolor
: HASH S*
;
prio
: IMPORTANT_SYM S*
;
"""
def __init__(self, name=None, value=None, priority='',
_mediaQuery=False, parent=None):
"""
:param name:
a property name string (will be normalized)
:param value:
a property value string
:param priority:
an optional priority string which currently must be u'',
u'!important' or u'important'
:param _mediaQuery:
if ``True`` value is optional (used by MediaQuery)
:param parent:
the parent object, normally a
:class:`css_parser.css.CSSStyleDeclaration`
"""
super(Property, self).__init__()
self.seqs = [[], None, []]
self.wellformed = False
self._mediaQuery = _mediaQuery
self.parent = parent
self.__nametoken = None
self._name = ''
self._literalname = ''
self.seqs[1] = PropertyValue(parent=self)
if name:
self.name = name
self.propertyValue = value
self._priority = ''
self._literalpriority = ''
if priority:
self.priority = priority
def __repr__(self):
return "css_parser.css.%s(name=%r, value=%r, priority=%r)" % (
self.__class__.__name__,
self.literalname,
self.propertyValue.cssText,
self.priority)
def __str__(self):
return "<%s.%s object name=%r value=%r priority=%r valid=%r at 0x%x>" \
% (self.__class__.__module__,
self.__class__.__name__,
self.name,
self.propertyValue.cssText,
self.priority,
self.valid,
id(self))
def _isValidating(self):
"""Return True if validation is enabled."""
try:
return self.parent.validating
except AttributeError:
# default (no parent)
return True
def _getCssText(self):
"""Return serialized property cssText."""
return css_parser.ser.do_Property(self)
def _setCssText(self, cssText):
"""
:exceptions:
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error and
is unparsable.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if the rule is readonly.
"""
# check and prepare tokenlists for setting
tokenizer = self._tokenize2(cssText)
nametokens = self._tokensupto2(tokenizer, propertynameendonly=True)
if nametokens:
wellformed = True
valuetokens = self._tokensupto2(tokenizer,
propertyvalueendonly=True)
prioritytokens = self._tokensupto2(tokenizer,
propertypriorityendonly=True)
if self._mediaQuery and not valuetokens:
# MediaQuery may consist of name only
self.name = nametokens
self.propertyValue = None
self.priority = None
return
# remove colon from nametokens
colontoken = nametokens.pop()
if self._tokenvalue(colontoken) != ':':
wellformed = False
self._log.error('Property: No ":" after name found: %s' %
self._valuestr(cssText), colontoken)
elif not nametokens:
wellformed = False
self._log.error('Property: No property name found: %s' %
self._valuestr(cssText), colontoken)
if valuetokens:
if self._tokenvalue(valuetokens[-1]) == '!':
# priority given, move "!" to prioritytokens
prioritytokens.insert(0, valuetokens.pop(-1))
else:
wellformed = False
self._log.error('Property: No property value found: %s' %
self._valuestr(cssText), colontoken)
if wellformed:
self.wellformed = True
self.name = nametokens
self.propertyValue = valuetokens
self.priority = prioritytokens
# also invalid values are set!
if self._isValidating():
self.validate()
else:
self._log.error('Property: No property name found: %s' %
self._valuestr(cssText))
cssText = property(fget=_getCssText, fset=_setCssText,
doc="A parsable textual representation.")
def _setName(self, name):
"""
:exceptions:
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified name has a syntax error and is
unparsable.
"""
# for closures: must be a mutable
new = {'literalname': None,
'wellformed': True}
def _ident(expected, seq, token, tokenizer=None):
# name
if 'name' == expected:
new['literalname'] = self._tokenvalue(token).lower()
seq.append(new['literalname'])
return 'EOF'
else:
new['wellformed'] = False
self._log.error('Property: Unexpected ident.', token)
return expected
newseq = []
wellformed, expected = self._parse(expected='name',
seq=newseq,
tokenizer=self._tokenize2(name),
productions={'IDENT': _ident})
wellformed = wellformed and new['wellformed']
# post conditions
# define a token for error logging
if isinstance(name, list):
token = name[0]
self.__nametoken = token
else:
token = None
if not new['literalname']:
wellformed = False
self._log.error('Property: No name found: %s' %
self._valuestr(name), token=token)
if wellformed:
self.wellformed = True
self._literalname = new['literalname']
self._name = self._normalize(self._literalname)
self.seqs[0] = newseq
# validate
if self._isValidating() and self._name not in css_parser.profile.knownNames:
# self.valid = False
self._log.warn('Property: Unknown Property name.',
token=token, neverraise=True)
else:
pass
# self.valid = True
# if self.propertyValue:
# self.propertyValue._propertyName = self._name
# #self.valid = self.propertyValue.valid
else:
self.wellformed = False
name = property(lambda self: self._name, _setName,
doc="Name of this property.")
literalname = property(lambda self: self._literalname,
doc="Readonly literal (not normalized) name "
"of this property")
def _setPropertyValue(self, cssText):
"""
See css.PropertyValue
:exceptions:
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error
(according to the attached property) or is unparsable.
- :exc:`~xml.dom.InvalidModificationErr`:
TODO: Raised if the specified CSS string value represents a different
type of values than the values allowed by the CSS property.
"""
if self._mediaQuery and not cssText:
self.seqs[1] = PropertyValue(parent=self)
else:
self.seqs[1].cssText = cssText
self.wellformed = self.wellformed and self.seqs[1].wellformed
propertyValue = property(lambda self: self.seqs[1],
_setPropertyValue,
doc="(css_parser) PropertyValue object of property")
def _getValue(self):
if self.propertyValue:
# value without comments
return self.propertyValue.value
else:
return ''
def _setValue(self, value):
self._setPropertyValue(value)
value = property(_getValue, _setValue,
doc="The textual value of this Properties propertyValue.")
def _setPriority(self, priority):
"""
priority
a string, currently either u'', u'!important' or u'important'
Format::
prio
: IMPORTANT_SYM S*
;
"!"{w}"important" {return IMPORTANT_SYM;}
:exceptions:
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified priority has a syntax error and is
unparsable.
In this case a priority not equal to None, "" or "!{w}important".
As CSSOM defines CSSStyleDeclaration.getPropertyPriority resulting
in u'important' this value is also allowed to set a Properties
priority
"""
if self._mediaQuery:
self._priority = ''
self._literalpriority = ''
if priority:
self._log.error('Property: No priority in a MediaQuery - '
'ignored.')
return
if isinstance(priority, string_type) and 'important' == self._normalize(priority):
priority = '!%s' % priority
# for closures: must be a mutable
new = {'literalpriority': '',
'wellformed': True}
def _char(expected, seq, token, tokenizer=None):
# "!"
val = self._tokenvalue(token)
if '!' == expected == val:
seq.append(val)
return 'important'
else:
new['wellformed'] = False
self._log.error('Property: Unexpected char.', token)
return expected
def _ident(expected, seq, token, tokenizer=None):
# "important"
val = self._tokenvalue(token)
if 'important' == expected:
new['literalpriority'] = val
seq.append(val)
return 'EOF'
else:
new['wellformed'] = False
self._log.error('Property: Unexpected ident.', token)
return expected
newseq = []
wellformed, expected = self._parse(expected='!',
seq=newseq,
tokenizer=self._tokenize2(priority),
productions={'CHAR': _char,
'IDENT': _ident})
wellformed = wellformed and new['wellformed']
# post conditions
if priority and not new['literalpriority']:
wellformed = False
self._log.info('Property: Invalid priority: %s' %
self._valuestr(priority))
if wellformed:
self.wellformed = self.wellformed and wellformed
self._literalpriority = new['literalpriority']
self._priority = self._normalize(self.literalpriority)
self.seqs[2] = newseq
# validate priority
if self._priority not in ('', 'important'):
self._log.error('Property: No CSS priority value: %s' %
self._priority)
priority = property(lambda self: self._priority, _setPriority,
doc="Priority of this property.")
literalpriority = property(lambda self: self._literalpriority,
doc="Readonly literal (not normalized) priority of this property")
def _setParent(self, parent):
self._parent = parent
parent = property(lambda self: self._parent, _setParent,
doc="The Parent Node (normally a CSSStyledeclaration) of this "
"Property")
def validate(self):
"""Validate value against `profiles` which are checked dynamically.
properties in e.g. @font-face rules are checked against
``css_parser.profile.CSS3_FONT_FACE`` only.
For each of the following cases a message is reported:
- INVALID (so the property is known but not valid)
``ERROR Property: Invalid value for "{PROFILE-1[/PROFILE-2...]"
property: ...``
- VALID but not in given profiles or defaultProfiles
``WARNING Property: Not valid for profile "{PROFILE-X}" but valid
"{PROFILE-Y}" property: ...``
- VALID in current profile
``DEBUG Found valid "{PROFILE-1[/PROFILE-2...]" property...``
- UNKNOWN property
``WARNING Unknown Property name...`` is issued
so for example::
css_parser.log.setLevel(logging.DEBUG)
parser = css_parser.CSSParser()
s = parser.parseString('''body {
unknown-property: x;
color: 4;
color: rgba(1,2,3,4);
color: red
}''')
# Log output:
WARNING Property: Unknown Property name. [2:9: unknown-property]
ERROR Property: Invalid value for "CSS Color Module Level 3/CSS Level 2.1" property: 4 [3:9: color]
DEBUG Property: Found valid "CSS Color Module Level 3" value: rgba(1, 2, 3, 4) [4:9: color]
DEBUG Property: Found valid "CSS Level 2.1" value: red [5:9: color]
and when setting an explicit default profile::
css_parser.profile.defaultProfiles = css_parser.profile.CSS_LEVEL_2
s = parser.parseString('''body {
unknown-property: x;
color: 4;
color: rgba(1,2,3,4);
color: red
}''')
# Log output:
WARNING Property: Unknown Property name. [2:9: unknown-property]
ERROR Property: Invalid value for "CSS Color Module Level 3/CSS Level 2.1" property: 4 [3:9: color]
WARNING Property: Not valid for profile "CSS Level 2.1" but valid
"CSS Color Module Level 3" value: rgba(1, 2, 3, 4) [4:9: color]
DEBUG Property: Found valid "CSS Level 2.1" value: red [5:9: color]
"""
valid = False
profiles = None
try:
# if @font-face use that profile
rule = self.parent.parentRule
except AttributeError:
pass
else:
if rule is not None:
if rule.type == rule.FONT_FACE_RULE:
profiles = [css_parser.profile.CSS3_FONT_FACE]
# TODO: same for @page
if self.name and self.value:
pass
# TODO
# cv = self.propertyValue
# if cv.cssValueType == cv.CSS_VARIABLE and not cv.value:
# # TODO: false alarms too!
# css_parser.log.warn(u'No value for variable "%s" found, keeping '
# u'variable.' % cv.name, neverraise=True)
if self.name in css_parser.profile.knownNames:
# add valid, matching, validprofiles...
valid, matching, validprofiles = \
css_parser.profile.validateWithProfile(self.name, self.value, profiles)
if not valid:
self._log.error('Property: Invalid value for '
'"%s" property: %s'
% ('/'.join(validprofiles), self.value),
token=self.__nametoken,
neverraise=True)
# TODO: remove logic to profiles!
elif valid and not matching: # (profiles and profiles not in validprofiles):
if not profiles:
notvalidprofiles = '/'.join(css_parser.profile.defaultProfiles)
else:
notvalidprofiles = profiles
self._log.warn('Property: Not valid for profile "%s" '
'but valid "%s" value: %s '
% (notvalidprofiles, '/'.join(validprofiles),
self.value),
token=self.__nametoken,
neverraise=True)
valid = False
elif valid:
self._log.debug('Property: Found valid "%s" value: %s'
% ('/'.join(validprofiles), self.value),
token=self.__nametoken,
neverraise=True)
if self._priority not in ('', 'important'):
valid = False
return valid
valid = property(validate, doc="Check if value of this property is valid "
"in the properties context.")
@Deprecated('Use ``property.propertyValue`` instead.')
def _getCSSValue(self):
return self.propertyValue
@Deprecated('Use ``property.propertyValue`` instead.')
def _setCSSValue(self, cssText):
self._setPropertyValue(cssText)
cssValue = property(_getCSSValue, _setCSSValue,
doc="(DEPRECATED) Use ``property.propertyValue`` instead.")
css-parser-1.0.4/src/css_parser/css/selector.py 0000644 0001750 0001750 00000100031 13407702010 021752 0 ustar kovid kovid 0000000 0000000 """Selector is a single Selector of a CSSStyleRule SelectorList.
Partly implements http://www.w3.org/TR/css3-selectors/.
TODO
- .contains(selector)
- .isSubselector(selector)
"""
from __future__ import unicode_literals, division, absolute_import, print_function
__all__ = ['Selector']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
from css_parser.helper import Deprecated
from css_parser.util import _SimpleNamespaces
import css_parser
import xml.dom
def as_list(p):
if isinstance(p, list):
return p
class Selector(css_parser.util.Base2):
"""
(css_parser) a single selector in a :class:`~css_parser.css.SelectorList`
of a :class:`~css_parser.css.CSSStyleRule`.
Format::
# implemented in SelectorList
selectors_group
: selector [ COMMA S* selector ]*
;
selector
: simple_selector_sequence [ combinator simple_selector_sequence ]*
;
combinator
/* combinators can be surrounded by white space */
: PLUS S* | GREATER S* | TILDE S* | S+
;
simple_selector_sequence
: [ type_selector | universal ]
[ HASH | class | attrib | pseudo | negation ]*
| [ HASH | class | attrib | pseudo | negation ]+
;
type_selector
: [ namespace_prefix ]? element_name
;
namespace_prefix
: [ IDENT | '*' ]? '|'
;
element_name
: IDENT
;
universal
: [ namespace_prefix ]? '*'
;
class
: '.' IDENT
;
attrib
: '[' S* [ namespace_prefix ]? IDENT S*
[ [ PREFIXMATCH |
SUFFIXMATCH |
SUBSTRINGMATCH |
'=' |
INCLUDES |
DASHMATCH ] S* [ IDENT | STRING ] S*
]? ']'
;
pseudo
/* '::' starts a pseudo-element, ':' a pseudo-class */
/* Exceptions: :first-line, :first-letter, :before and :after. */
/* Note that pseudo-elements are restricted to one per selector and */
/* occur only in the last simple_selector_sequence. */
: ':' ':'? [ IDENT | functional_pseudo ]
;
functional_pseudo
: FUNCTION S* expression ')'
;
expression
/* In CSS3, the expressions are identifiers, strings, */
/* or of the form "an+b" */
: [ [ PLUS | '-' | DIMENSION | NUMBER | STRING | IDENT ] S* ]+
;
negation
: NOT S* negation_arg S* ')'
;
negation_arg
: type_selector | universal | HASH | class | attrib | pseudo
;
"""
def __init__(self, selectorText=None, parent=None,
readonly=False):
"""
:Parameters:
selectorText
initial value of this selector
parent
a SelectorList
readonly
default to False
"""
super(Selector, self).__init__()
self.__namespaces = _SimpleNamespaces(log=self._log)
self._element = None
self._parent = parent
self._specificity = (0, 0, 0, 0)
if selectorText:
self.selectorText = selectorText
self._readonly = readonly
def __repr__(self):
if self.__getNamespaces():
st = (self.selectorText, self._getUsedNamespaces())
else:
st = self.selectorText
return "css_parser.css.%s(selectorText=%r)" % (self.__class__.__name__, st)
def __str__(self):
return "" % (self.__class__.__name__,
self.selectorText,
self.specificity,
self._getUsedNamespaces(),
id(self))
def _getUsedUris(self):
"Return list of actually used URIs in this Selector."
uris = set()
for item in self.seq:
type_, val = item.type, item.value
if type_.endswith('-selector') or type_ == 'universal' and \
isinstance(val, tuple) and val[0] not in (None, '*'):
uris.add(val[0])
return uris
def _getUsedNamespaces(self):
"Return actually used namespaces only."
useduris = self._getUsedUris()
namespaces = _SimpleNamespaces(log=self._log)
for p, uri in as_list(self._namespaces.items()):
if uri in useduris:
namespaces[p] = uri
return namespaces
def __getNamespaces(self):
"Use own namespaces if not attached to a sheet, else the sheet's ones."
try:
return self._parent.parentRule.parentStyleSheet.namespaces
except AttributeError:
return self.__namespaces
_namespaces = property(__getNamespaces,
doc="If this Selector is attached to a "
"CSSStyleSheet the namespaces of that sheet "
"are mirrored here. While the Selector (or "
"parent SelectorList or parentRule(s) of that "
"are not attached a own dict of {prefix: "
"namespaceURI} is used.")
element = property(lambda self: self._element,
doc="Effective element target of this selector.")
parent = property(lambda self: self._parent,
doc="(DOM) The SelectorList that contains this Selector "
"or None if this Selector is not attached to a "
"SelectorList.")
def _getSelectorText(self):
"""Return serialized format."""
return css_parser.ser.do_css_Selector(self)
def _setSelectorText(self, selectorText):
"""
:param selectorText:
parsable string or a tuple of (selectorText, dict-of-namespaces).
Given namespaces are ignored if this object is attached to a
CSSStyleSheet!
:exceptions:
- :exc:`~xml.dom.NamespaceErr`:
Raised if the specified selector uses an unknown namespace
prefix.
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error
and is unparsable.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this rule is readonly.
"""
self._checkReadonly()
# might be (selectorText, namespaces)
selectorText, namespaces = self._splitNamespacesOff(selectorText)
try:
# uses parent stylesheets namespaces if available,
# otherwise given ones
namespaces = self.parent.parentRule.parentStyleSheet.namespaces
except AttributeError:
pass
tokenizer = self._tokenize2(selectorText)
if not tokenizer:
self._log.error('Selector: No selectorText given.')
else:
# prepare tokenlist:
# "*" -> type "universal"
# "*"|IDENT + "|" -> combined to "namespace_prefix"
# "|" -> type "namespace_prefix"
# "." + IDENT -> combined to "class"
# ":" + IDENT, ":" + FUNCTION -> pseudo-class
# FUNCTION "not(" -> negation
# "::" + IDENT, "::" + FUNCTION -> pseudo-element
tokens = []
for t in tokenizer:
typ, val, lin, col = t
if val == ':' and tokens and\
self._tokenvalue(tokens[-1]) == ':':
# combine ":" and ":"
tokens[-1] = (typ, '::', lin, col)
elif typ == 'IDENT' and tokens\
and self._tokenvalue(tokens[-1]) == '.':
# class: combine to .IDENT
tokens[-1] = ('class', '.'+val, lin, col)
elif typ == 'IDENT' and tokens and \
self._tokenvalue(tokens[-1]).startswith(':') and\
not self._tokenvalue(tokens[-1]).endswith('('):
# pseudo-X: combine to :IDENT or ::IDENT but not ":a(" + "b"
if self._tokenvalue(tokens[-1]).startswith('::'):
t = 'pseudo-element'
else:
t = 'pseudo-class'
tokens[-1] = (t, self._tokenvalue(tokens[-1])+val, lin, col)
elif typ == 'FUNCTION' and val == 'not(' and tokens and \
':' == self._tokenvalue(tokens[-1]):
tokens[-1] = ('negation', ':' + val, lin, tokens[-1][3])
elif typ == 'FUNCTION' and tokens\
and self._tokenvalue(tokens[-1]).startswith(':'):
# pseudo-X: combine to :FUNCTION( or ::FUNCTION(
if self._tokenvalue(tokens[-1]).startswith('::'):
t = 'pseudo-element'
else:
t = 'pseudo-class'
tokens[-1] = (t, self._tokenvalue(tokens[-1])+val, lin, col)
elif val == '*' and tokens and\
self._type(tokens[-1]) == 'namespace_prefix' and\
self._tokenvalue(tokens[-1]).endswith('|'):
# combine prefix|*
tokens[-1] = ('universal', self._tokenvalue(tokens[-1])+val,
lin, col)
elif val == '*':
# universal: "*"
tokens.append(('universal', val, lin, col))
elif val == '|' and tokens and\
self._type(tokens[-1]) in (self._prods.IDENT, 'universal')\
and self._tokenvalue(tokens[-1]).find('|') == -1:
# namespace_prefix: "IDENT|" or "*|"
tokens[-1] = ('namespace_prefix',
self._tokenvalue(tokens[-1])+'|', lin, col)
elif val == '|':
# namespace_prefix: "|"
tokens.append(('namespace_prefix', val, lin, col))
else:
tokens.append(t)
tokenizer = iter(tokens)
# for closures: must be a mutable
new = {'context': [''], # stack of: 'attrib', 'negation', 'pseudo'
'element': None,
'_PREFIX': None,
'specificity': [0, 0, 0, 0], # mutable, finally a tuple!
'wellformed': True
}
# used for equality checks and setting of a space combinator
S = ' '
def append(seq, val, typ=None, token=None):
"""
appends to seq
namespace_prefix, IDENT will be combined to a tuple
(prefix, name) where prefix might be None, the empty string
or a prefix.
Saved are also:
- specificity definition: style, id, class/att, type
- element: the element this Selector is for
"""
context = new['context'][-1]
if token:
line, col = token[2], token[3]
else:
line, col = None, None
if typ == '_PREFIX':
# SPECIAL TYPE: save prefix for combination with next
new['_PREFIX'] = val[:-1]
# handle next time
return
if new['_PREFIX'] is not None:
# as saved from before and reset to None
prefix, new['_PREFIX'] = new['_PREFIX'], None
elif typ == 'universal' and '|' in val:
# val == *|* or prefix|*
prefix, val = val.split('|')
else:
prefix = None
# namespace
if (typ.endswith('-selector') or typ == 'universal') and not (
'attribute-selector' == typ and not prefix):
# att **IS NOT** in default ns
if prefix == '*':
# *|name: in ANY_NS
namespaceURI = css_parser._ANYNS
elif prefix is None:
# e or *: default namespace with prefix u''
# or local-name()
namespaceURI = namespaces.get('', None)
elif prefix == '':
# |name or |*: in no (or the empty) namespace
namespaceURI = ''
else:
# explicit namespace prefix
# does not raise KeyError, see _SimpleNamespaces
namespaceURI = namespaces[prefix]
if namespaceURI is None:
new['wellformed'] = False
self._log.error('Selector: No namespaceURI found '
'for prefix %r' % prefix,
token=token,
error=xml.dom.NamespaceErr)
return
# val is now (namespaceprefix, name) tuple
val = (namespaceURI, val)
# specificity
if not context or context == 'negation':
if 'id' == typ:
new['specificity'][1] += 1
elif 'class' == typ or '[' == val:
new['specificity'][2] += 1
elif typ in ('type-selector', 'negation-type-selector',
'pseudo-element'):
new['specificity'][3] += 1
if not context and typ in ('type-selector', 'universal'):
# define element
new['element'] = val
seq.append(val, typ, line=line, col=col)
# expected constants
simple_selector_sequence = 'type_selector universal HASH class ' \
'attrib pseudo negation '
simple_selector_sequence2 = 'HASH class attrib pseudo negation '
element_name = 'element_name'
negation_arg = 'type_selector universal HASH class attrib pseudo'
negationend = ')'
attname = 'prefix attribute'
attname2 = 'attribute'
attcombinator = 'combinator ]' # optional
attvalue = 'value' # optional
attend = ']'
expressionstart = 'PLUS - DIMENSION NUMBER STRING IDENT'
expression = expressionstart + ' )'
combinator = ' combinator'
def _COMMENT(expected, seq, token, tokenizer=None):
"special implementation for comment token"
append(seq, css_parser.css.CSSComment([token]), 'COMMENT',
token=token)
return expected
def _S(expected, seq, token, tokenizer=None):
# S
context = new['context'][-1]
if context.startswith('pseudo-'):
if seq and seq[-1].value not in '+-':
# e.g. x:func(a + b)
append(seq, S, 'S', token=token)
return expected
elif context != 'attrib' and 'combinator' in expected:
append(seq, S, 'descendant', token=token)
return simple_selector_sequence + combinator
else:
return expected
def _universal(expected, seq, token, tokenizer=None):
# *|* or prefix|*
context = new['context'][-1]
val = self._tokenvalue(token)
if 'universal' in expected:
append(seq, val, 'universal', token=token)
if 'negation' == context:
return negationend
else:
return simple_selector_sequence2 + combinator
else:
new['wellformed'] = False
self._log.error(
'Selector: Unexpected universal.', token=token)
return expected
def _namespace_prefix(expected, seq, token, tokenizer=None):
# prefix| => element_name
# or prefix| => attribute_name if attrib
context = new['context'][-1]
val = self._tokenvalue(token)
if 'attrib' == context and 'prefix' in expected:
# [PREFIX|att]
append(seq, val, '_PREFIX', token=token)
return attname2
elif 'type_selector' in expected:
# PREFIX|*
append(seq, val, '_PREFIX', token=token)
return element_name
else:
new['wellformed'] = False
self._log.error(
'Selector: Unexpected namespace prefix.', token=token)
return expected
def _pseudo(expected, seq, token, tokenizer=None):
# pseudo-class or pseudo-element :a ::a :a( ::a(
"""
/* '::' starts a pseudo-element, ':' a pseudo-class */
/* Exceptions: :first-line, :first-letter, :before and
:after. */
/* Note that pseudo-elements are restricted to one per selector
and */
/* occur only in the last simple_selector_sequence. */
"""
context = new['context'][-1]
val, typ = self._tokenvalue(token, normalize=True),\
self._type(token)
if 'pseudo' in expected:
if val in (':first-line',
':first-letter',
':before',
':after'):
# always pseudo-element ???
typ = 'pseudo-element'
append(seq, val, typ, token=token)
if val.endswith('('):
# function
# "pseudo-" "class" or "element"
new['context'].append(typ)
return expressionstart
elif 'negation' == context:
return negationend
elif 'pseudo-element' == typ:
# only one per element, check at ) also!
return combinator
else:
return simple_selector_sequence2 + combinator
else:
new['wellformed'] = False
self._log.error(
'Selector: Unexpected start of pseudo.', token=token)
return expected
def _expression(expected, seq, token, tokenizer=None):
# [ [ PLUS | '-' | DIMENSION | NUMBER | STRING | IDENT ] S* ]+
context = new['context'][-1]
val, typ = self._tokenvalue(token), self._type(token)
if context.startswith('pseudo-'):
append(seq, val, typ, token=token)
return expression
else:
new['wellformed'] = False
self._log.error(
'Selector: Unexpected %s.' % typ, token=token)
return expected
def _attcombinator(expected, seq, token, tokenizer=None):
# context: attrib
# PREFIXMATCH | SUFFIXMATCH | SUBSTRINGMATCH | INCLUDES |
# DASHMATCH
context = new['context'][-1]
val, typ = self._tokenvalue(token), self._type(token)
if 'attrib' == context and 'combinator' in expected:
# combinator in attrib
append(seq, val, typ.lower(), token=token)
return attvalue
else:
new['wellformed'] = False
self._log.error(
'Selector: Unexpected %s.' % typ, token=token)
return expected
def _string(expected, seq, token, tokenizer=None):
# identifier
context = new['context'][-1]
typ, val = self._type(token), self._stringtokenvalue(token)
# context: attrib
if 'attrib' == context and 'value' in expected:
# attrib: [...=VALUE]
append(seq, val, typ, token=token)
return attend
# context: pseudo
elif context.startswith('pseudo-'):
# :func(...)
append(seq, val, typ, token=token)
return expression
else:
new['wellformed'] = False
self._log.error(
'Selector: Unexpected STRING.', token=token)
return expected
def _ident(expected, seq, token, tokenizer=None):
# identifier
context = new['context'][-1]
val, typ = self._tokenvalue(token), self._type(token)
# context: attrib
if 'attrib' == context and 'attribute' in expected:
# attrib: [...|ATT...]
append(seq, val, 'attribute-selector', token=token)
return attcombinator
elif 'attrib' == context and 'value' in expected:
# attrib: [...=VALUE]
append(seq, val, 'attribute-value', token=token)
return attend
# context: negation
elif 'negation' == context:
# negation: (prefix|IDENT)
append(seq, val, 'negation-type-selector', token=token)
return negationend
# context: pseudo
elif context.startswith('pseudo-'):
# :func(...)
append(seq, val, typ, token=token)
return expression
elif 'type_selector' in expected or element_name == expected:
# element name after ns or complete type_selector
append(seq, val, 'type-selector', token=token)
return simple_selector_sequence2 + combinator
else:
new['wellformed'] = False
self._log.error('Selector: Unexpected IDENT.', token=token)
return expected
def _class(expected, seq, token, tokenizer=None):
# .IDENT
context = new['context'][-1]
val = self._tokenvalue(token)
if 'class' in expected:
append(seq, val, 'class', token=token)
if 'negation' == context:
return negationend
else:
return simple_selector_sequence2 + combinator
else:
new['wellformed'] = False
self._log.error('Selector: Unexpected class.', token=token)
return expected
def _hash(expected, seq, token, tokenizer=None):
# #IDENT
context = new['context'][-1]
val = self._tokenvalue(token)
if 'HASH' in expected:
append(seq, val, 'id', token=token)
if 'negation' == context:
return negationend
else:
return simple_selector_sequence2 + combinator
else:
new['wellformed'] = False
self._log.error('Selector: Unexpected HASH.', token=token)
return expected
def _char(expected, seq, token, tokenizer=None):
# + > ~ ) [ ] + -
context = new['context'][-1]
val = self._tokenvalue(token)
# context: attrib
if ']' == val and 'attrib' == context and ']' in expected:
# end of attrib
append(seq, val, 'attribute-end', token=token)
context = new['context'].pop() # attrib is done
context = new['context'][-1]
if 'negation' == context:
return negationend
else:
return simple_selector_sequence2 + combinator
elif '=' == val and 'attrib' == context\
and 'combinator' in expected:
# combinator in attrib
append(seq, val, 'equals', token=token)
return attvalue
# context: negation
elif ')' == val and 'negation' == context and ')' in expected:
# not(negation_arg)"
append(seq, val, 'negation-end', token=token)
new['context'].pop() # negation is done
context = new['context'][-1]
return simple_selector_sequence + combinator
# context: pseudo (at least one expression)
elif val in '+-' and context.startswith('pseudo-'):
# :func(+ -)"
_names = {'+': 'plus', '-': 'minus'}
if val == '+' and seq and seq[-1].value == S:
seq.replace(-1, val, _names[val])
else:
append(seq, val, _names[val],
token=token)
return expression
elif ')' == val and context.startswith('pseudo-') and\
expression == expected:
# :func(expression)"
append(seq, val, 'function-end', token=token)
new['context'].pop() # pseudo is done
if 'pseudo-element' == context:
return combinator
else:
return simple_selector_sequence + combinator
# context: ROOT
elif '[' == val and 'attrib' in expected:
# start of [attrib]
append(seq, val, 'attribute-start', token=token)
new['context'].append('attrib')
return attname
elif val in '+>~' and 'combinator' in expected:
# no other combinator except S may be following
_names = {
'>': 'child',
'+': 'adjacent-sibling',
'~': 'following-sibling'}
if seq and seq[-1].value == S:
seq.replace(-1, val, _names[val])
else:
append(seq, val, _names[val], token=token)
return simple_selector_sequence
elif ',' == val:
# not a selectorlist
new['wellformed'] = False
self._log.error(
'Selector: Single selector only.',
error=xml.dom.InvalidModificationErr,
token=token)
return expected
else:
new['wellformed'] = False
self._log.error(
'Selector: Unexpected CHAR.', token=token)
return expected
def _negation(expected, seq, token, tokenizer=None):
# not(
val = self._tokenvalue(token, normalize=True)
if 'negation' in expected:
new['context'].append('negation')
append(seq, val, 'negation-start', token=token)
return negation_arg
else:
new['wellformed'] = False
self._log.error(
'Selector: Unexpected negation.', token=token)
return expected
def _atkeyword(expected, seq, token, tokenizer=None):
"invalidates selector"
new['wellformed'] = False
self._log.error(
'Selector: Unexpected ATKEYWORD.', token=token)
return expected
# expected: only|not or mediatype, mediatype, feature, and
newseq = self._tempSeq()
wellformed, expected = self._parse(
expected=simple_selector_sequence,
seq=newseq, tokenizer=tokenizer,
productions={'CHAR': _char,
'class': _class,
'HASH': _hash,
'STRING': _string,
'IDENT': _ident,
'namespace_prefix': _namespace_prefix,
'negation': _negation,
'pseudo-class': _pseudo,
'pseudo-element': _pseudo,
'universal': _universal,
# pseudo
'NUMBER': _expression,
'DIMENSION': _expression,
# attribute
'PREFIXMATCH': _attcombinator,
'SUFFIXMATCH': _attcombinator,
'SUBSTRINGMATCH': _attcombinator,
'DASHMATCH': _attcombinator,
'INCLUDES': _attcombinator,
'S': _S,
'COMMENT': _COMMENT,
'ATKEYWORD': _atkeyword})
wellformed = wellformed and new['wellformed']
# post condition
if len(new['context']) > 1 or not newseq:
wellformed = False
self._log.error('Selector: Invalid or incomplete selector: %s'
% self._valuestr(selectorText))
if expected == 'element_name':
wellformed = False
self._log.error('Selector: No element name found: %s'
% self._valuestr(selectorText))
if expected == simple_selector_sequence and newseq:
wellformed = False
self._log.error('Selector: Cannot end with combinator: %s'
% self._valuestr(selectorText))
if newseq and hasattr(newseq[-1].value, 'strip') \
and newseq[-1].value.strip() == '':
del newseq[-1]
# set
if wellformed:
self.__namespaces = namespaces
self._element = new['element']
self._specificity = tuple(new['specificity'])
self._setSeq(newseq)
# filter that only used ones are kept
self.__namespaces = self._getUsedNamespaces()
selectorText = property(_getSelectorText, _setSelectorText,
doc="(DOM) The parsable textual representation of "
"the selector.")
specificity = property(lambda self: self._specificity,
doc="""Specificity of this selector (READONLY).
Tuple of (a, b, c, d) where:
a
presence of style in document, always 0 if not used on a
document
b
number of ID selectors
c
number of .class selectors
d
number of Element (type) selectors""")
wellformed = property(lambda self: bool(len(self.seq)))
@Deprecated('Use property parent instead')
def _getParentList(self):
return self.parent
parentList = property(_getParentList,
doc="DEPRECATED, see property parent instead")
css-parser-1.0.4/src/css_parser/css/selectorlist.py 0000644 0001750 0001750 00000021014 13407702010 022651 0 ustar kovid kovid 0000000 0000000 """SelectorList is a list of CSS Selector objects.
TODO
- remove duplicate Selectors. -> CSSOM canonicalize
- ??? CSS2 gives a special meaning to the comma (,) in selectors.
However, since it is not known if the comma may acquire other
meanings in future versions of CSS, the whole statement should be
ignored if there is an error anywhere in the selector, even though
the rest of the selector may look reasonable in CSS2.
Illegal example(s):
For example, since the "&" is not a valid token in a CSS2 selector,
a CSS2 user agent must ignore the whole second line, and not set
the color of H3 to red:
"""
from __future__ import unicode_literals, division, absolute_import, print_function
__all__ = ['SelectorList']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
from .selector import Selector
import css_parser
class SelectorList(css_parser.util.Base, css_parser.util.ListSeq):
"""A list of :class:`~css_parser.css.Selector` objects
of a :class:`~css_parser.css.CSSStyleRule`."""
def __init__(self, selectorText=None, parentRule=None,
readonly=False):
"""
:Parameters:
selectorText
parsable list of Selectors
parentRule
the parent CSSRule if available
"""
super(SelectorList, self).__init__()
self._parentRule = parentRule
if selectorText:
self.selectorText = selectorText
self._readonly = readonly
def __repr__(self):
if self._namespaces:
st = (self.selectorText, self._namespaces)
else:
st = self.selectorText
return "css_parser.css.%s(selectorText=%r)" % (self.__class__.__name__, st)
def __str__(self):
return "" % (self.__class__.__name__,
self.selectorText,
self._namespaces,
id(self))
def __setitem__(self, index, newSelector):
"""Overwrite ListSeq.__setitem__
Any duplicate Selectors are **not** removed.
"""
newSelector = self.__prepareset(newSelector)
if newSelector:
self.seq[index] = newSelector
def __prepareset(self, newSelector, namespaces=None):
"Used by appendSelector and __setitem__"
if not namespaces:
namespaces = {}
self._checkReadonly()
if not isinstance(newSelector, Selector):
newSelector = Selector((newSelector, namespaces),
parent=self)
if newSelector.wellformed:
newSelector._parent = self # maybe set twice but must be!
return newSelector
def __getNamespaces(self):
"""Use children namespaces if not attached to a sheet, else the sheet's
ones.
"""
try:
return self.parentRule.parentStyleSheet.namespaces
except AttributeError:
namespaces = {}
for selector in self.seq:
namespaces.update(selector._namespaces)
return namespaces
def _getUsedUris(self):
"Used by CSSStyleSheet to check if @namespace rules are needed"
uris = set()
for s in self:
uris.update(s._getUsedUris())
return uris
_namespaces = property(__getNamespaces, doc="""If this SelectorList is
attached to a CSSStyleSheet the namespaces of that sheet are mirrored
here. While the SelectorList (or parentRule(s) are
not attached the namespaces of all children Selectors are used.""")
def append(self, newSelector):
"Same as :meth:`appendSelector`."
self.appendSelector(newSelector)
def appendSelector(self, newSelector):
"""
Append `newSelector` to this list (a string will be converted to a
:class:`~css_parser.css.Selector`).
:param newSelector:
comma-separated list of selectors (as a single string) or a tuple of
`(newSelector, dict-of-namespaces)`
:returns: New :class:`~css_parser.css.Selector` or ``None`` if
`newSelector` is not wellformed.
:exceptions:
- :exc:`~xml.dom.NamespaceErr`:
Raised if the specified selector uses an unknown namespace
prefix.
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error
and is unparsable.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this rule is readonly.
"""
self._checkReadonly()
# might be (selectorText, namespaces)
newSelector, namespaces = self._splitNamespacesOff(newSelector)
try:
# use parent's only if available
namespaces = self.parentRule.parentStyleSheet.namespaces
except AttributeError:
# use already present namespaces plus new given ones
_namespaces = self._namespaces
_namespaces.update(namespaces)
namespaces = _namespaces
newSelector = self.__prepareset(newSelector, namespaces)
if newSelector:
seq = self.seq[:]
del self.seq[:]
for s in seq:
if s.selectorText != newSelector.selectorText:
self.seq.append(s)
self.seq.append(newSelector)
return newSelector
def _getSelectorText(self):
"Return serialized format."
return css_parser.ser.do_css_SelectorList(self)
def _setSelectorText(self, selectorText):
"""
:param selectorText:
comma-separated list of selectors or a tuple of
(selectorText, dict-of-namespaces)
:exceptions:
- :exc:`~xml.dom.NamespaceErr`:
Raised if the specified selector uses an unknown namespace
prefix.
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error
and is unparsable.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this rule is readonly.
"""
self._checkReadonly()
# might be (selectorText, namespaces)
selectorText, namespaces = self._splitNamespacesOff(selectorText)
try:
# use parent's only if available
namespaces = self.parentRule.parentStyleSheet.namespaces
except AttributeError:
pass
wellformed = True
tokenizer = self._tokenize2(selectorText)
newseq = []
expected = True
while True:
# find all upto and including next ",", EOF or nothing
selectortokens = self._tokensupto2(tokenizer, listseponly=True)
if selectortokens:
if self._tokenvalue(selectortokens[-1]) == ',':
expected = selectortokens.pop()
else:
expected = None
selector = Selector((selectortokens, namespaces),
parent=self)
if selector.wellformed:
newseq.append(selector)
else:
wellformed = False
self._log.error('SelectorList: Invalid Selector: %s' %
self._valuestr(selectortokens))
else:
break
# post condition
if ',' == expected:
wellformed = False
self._log.error('SelectorList: Cannot end with ",": %r' %
self._valuestr(selectorText))
elif expected:
wellformed = False
self._log.error('SelectorList: Unknown Syntax: %r' %
self._valuestr(selectorText))
if wellformed:
self.seq = newseq
selectorText = property(_getSelectorText, _setSelectorText,
doc="(css_parser) The textual representation of the "
"selector for a rule set.")
length = property(lambda self: len(self),
doc="The number of :class:`~css_parser.css.Selector` "
"objects in the list.")
parentRule = property(lambda self: self._parentRule,
doc="(DOM) The CSS rule that contains this "
"SelectorList or ``None`` if this SelectorList "
"is not attached to a CSSRule.")
wellformed = property(lambda self: bool(len(self.seq)))
css-parser-1.0.4/src/css_parser/css/value.py 0000644 0001750 0001750 00000113501 13413036502 021257 0 ustar kovid kovid 0000000 0000000 from __future__ import unicode_literals, division, absolute_import, print_function
import re
import colorsys
from css_parser.helper import normalize, pushtoken
import css_parser
from css_parser.prodparser import Choice, PreDef, Sequence, ProdParser, Prod
"""Value related classes.
DOM Level 2 CSS CSSValue, CSSPrimitiveValue and CSSValueList are **no longer**
supported and are replaced by these new classes.
"""
__all__ = ['PropertyValue',
'Value',
'ColorValue',
'DimensionValue',
'URIValue',
'CSSFunction',
'CSSCalc',
'CSSVariable',
'MSValue'
]
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
import sys
if sys.version_info[0] >= 3:
text_type = str
from urllib.parse import urljoin as urllib_urljoin
else:
text_type = unicode
from urlparse import urljoin as urllib_urljoin
def as_list(p):
if isinstance(p, list):
return p
return list(p)
class PropertyValue(css_parser.util._NewBase):
"""
An unstructured list like holder for all values defined for a
:class:`~css_parser.css.Property`. Contains :class:`~css_parser.css.Value`
or subclass objects. Currently there is no access to the combinators of
the defined values which might simply be space or comma or slash.
You may:
- iterate over all contained Value objects (not the separators like ``,``,
``/`` or `` `` though!)
- get a Value item by index or use ``PropertyValue[index]``
- find out the number of values defined (unstructured)
"""
def __init__(self, cssText=None, parent=None, readonly=False):
"""
:param cssText:
the parsable cssText of the value
:param readonly:
defaults to False
"""
super(PropertyValue, self).__init__()
self.parent = parent
self.wellformed = False
if cssText is not None: # may be 0
if isinstance(cssText, (int, float)):
cssText = text_type(cssText) # if it is a number
self.cssText = cssText
self._readonly = readonly
def __len__(self):
return len(as_list(self.__items()))
def __getitem__(self, index):
try:
return as_list(self.__items())[index]
except IndexError:
return None
def __iter__(self):
"Generator which iterates over values."
for item in self.__items():
yield item
def __repr__(self):
return "css_parser.css.%s(%r)" % (self.__class__.__name__, self.cssText)
def __str__(self):
return "" % (self.__class__.__name__,
self.length, self.cssText, id(self))
def __items(self, seq=None):
"a generator of Value obects only, no , / or ' '"
if seq is None:
seq = self.seq
return (x.value for x in seq if isinstance(x.value, Value))
def _setCssText(self, cssText):
if isinstance(cssText, (int, float)):
cssText = text_type(cssText) # if it is a number
"""
Format::
unary_operator
: '-' | '+'
;
operator
: '/' S* | ',' S* | /* empty */
;
expr
: term [ operator term ]*
;
term
: unary_operator?
[ NUMBER S* | PERCENTAGE S* | LENGTH S* | EMS S* | EXS S* |
ANGLE S* | TIME S* | FREQ S* ]
| STRING S* | IDENT S* | URI S* | hexcolor | function
| UNICODE-RANGE S*
;
function
: FUNCTION S* expr ')' S*
;
/*
* There is a constraint on the color that it must
* have either 3 or 6 hex-digits (i.e., [0-9a-fA-F])
* after the "#"; e.g., "#000" is OK, but "#abcd" is not.
*/
hexcolor
: HASH S*
;
:exceptions:
- :exc:`~xml.dom.SyntaxErr`:
Raised if the specified CSS string value has a syntax error
(according to the attached property) or is unparsable.
- :exc:`~xml.dom.InvalidModificationErr`:
TODO: Raised if the specified CSS string value represents a
different type of values than the values allowed by the CSS
property.
- :exc:`~xml.dom.NoModificationAllowedErr`:
Raised if this value is readonly.
"""
self._checkReadonly()
# used as operator is , / or S
nextSor = ',/'
term = Choice(_ColorProd(self, nextSor),
_DimensionProd(self, nextSor),
_URIProd(self, nextSor),
_ValueProd(self, nextSor),
# _Rect(self, nextSor),
# all other functions
_CSSVariableProd(self, nextSor),
_MSValueProd(self, nextSor),
_CalcValueProd(self, nextSor),
_CSSFunctionProd(self, nextSor)
)
operator = Choice(PreDef.S(toSeq=False),
PreDef.char('comma', ',',
toSeq=lambda t, tokens: ('operator', t[1]),
optional=True
),
PreDef.char('slash', '/',
toSeq=lambda t, tokens: ('operator', t[1]),
optional=True),
optional=True)
prods = Sequence(term,
Sequence( # mayEnd this Sequence if whitespace
operator,
# TODO: only when setting via other class
# used by variabledeclaration currently
PreDef.char('END', ';',
stopAndKeep=True,
optional=True),
# TODO: } and !important ends too!
term,
minmax=lambda: (0, None)))
# parse
ok, seq, store, unused = ProdParser().parse(cssText,
'PropertyValue',
prods)
# must be at least one value!
ok = ok and len(as_list(self.__items(seq))) > 0
for item in seq:
if hasattr(item.value, 'wellformed') and not item.value.wellformed:
ok = False
break
self.wellformed = ok
if ok:
self._setSeq(seq)
else:
self._log.error('PropertyValue: Unknown syntax or no value: %s' %
self._valuestr(cssText))
cssText = property(lambda self: css_parser.ser.do_css_PropertyValue(self),
_setCssText,
doc="A string representation of the current value.")
def item(self, index):
"""
The value at position `index`. Alternatively simple use
``PropertyValue[index]``.
:param index:
the parsable cssText of the value
:exceptions:
- :exc:`~IndexError`:
Raised if index if out of bounds
"""
return self[index]
length = property(lambda self: len(self),
doc="Number of values set.")
value = property(lambda self: css_parser.ser.do_css_PropertyValue(self, valuesOnly=True),
doc="A string representation of the current value "
"without any comments used for validation.")
class Value(css_parser.util._NewBase):
"""
Represents a single CSS value. For now simple values of
IDENT, STRING, or UNICODE-RANGE values are represented directly
as Value objects. Other values like e.g. FUNCTIONs are represented by
subclasses with an extended API.
"""
IDENT = 'IDENT'
STRING = 'STRING'
UNICODE_RANGE = 'UNICODE-RANGE'
URI = 'URI'
DIMENSION = 'DIMENSION'
NUMBER = 'NUMBER'
PERCENTAGE = 'PERCENTAGE'
COLOR_VALUE = 'COLOR_VALUE'
HASH = 'HASH'
FUNCTION = 'FUNCTION'
CALC = 'CALC'
VARIABLE = 'VARIABLE'
_type = None
_value = ''
def __init__(self, cssText=None, parent=None, readonly=False):
super(Value, self).__init__()
self.parent = parent
self.wellformed = False
if cssText:
self.cssText = cssText
def __repr__(self):
return "css_parser.css.%s(%r)" % (self.__class__.__name__, self.cssText)
def __str__(self):
return ""\
% (self.__class__.__name__,
self.type, self.value, self.cssText,
id(self))
def _setCssText(self, cssText):
self._checkReadonly()
prods = Choice(PreDef.hexcolor(stop=True),
PreDef.ident(stop=True),
PreDef.string(stop=True),
PreDef.unicode_range(stop=True),
)
ok, seq, store, unused = ProdParser().parse(cssText, 'Value', prods)
self.wellformed = ok
if ok:
# only 1 value anyway!
self._type = seq[0].type
self._value = seq[0].value
self._setSeq(seq)
cssText = property(lambda self: css_parser.ser.do_css_Value(self),
_setCssText,
doc='String value of this value.')
type = property(lambda self: self._type, # _setType,
doc="Type of this value, for now the production type "
"like e.g. `DIMENSION` or `STRING`. All types are "
"defined as constants in :class:`~css_parser.css.Value`.")
def _setValue(self, value):
# TODO: check!
self._value = value
value = property(lambda self: self._value, _setValue,
doc="Actual value if possible: An int or float or else "
" a string")
class ColorValue(Value):
"""
A color value like rgb(), rgba(), hsl(), hsla() or #rgb, #rrggbb
TODO: Color Keywords
"""
from .colors import COLORS
type = Value.COLOR_VALUE
# hexcolor, FUNCTION?
_colorType = None
_red = 0
_green = 0
_blue = 0
_alpha = 0
def __str__(self):
return ""\
% (self.__class__.__name__,
self.type, self.value,
self.colorType, self.red, self.green, self.blue, self.alpha,
id(self))
def _setCssText(self, cssText):
self._checkReadonly()
types = self._prods # rename!
component = Choice(PreDef.unary(toSeq=lambda t, tokens: (t[0],
DimensionValue(pushtoken(t, tokens),
parent=self)
)),
PreDef.number(toSeq=lambda t, tokens: (t[0],
DimensionValue(pushtoken(t, tokens),
parent=self)
)),
PreDef.percentage(toSeq=lambda t, tokens: (t[0],
DimensionValue(pushtoken(t, tokens),
parent=self)
))
)
noalp = Sequence(Prod(name='FUNCTION',
match=lambda t, v: t == types.FUNCTION and
v.lower() in ('rgb(', 'hsl('),
toSeq=lambda t, tokens: (t[0], normalize(t[1]))),
component,
Sequence(PreDef.comma(optional=True),
component,
minmax=lambda: (2, 2)
),
PreDef.funcEnd(stop=True)
)
witha = Sequence(Prod(name='FUNCTION',
match=lambda t, v: t == types.FUNCTION and
v.lower() in ('rgba(', 'hsla('),
toSeq=lambda t, tokens: (t[0],
normalize(t[1]))
),
component,
Sequence(PreDef.comma(optional=True),
component,
minmax=lambda: (3, 3)
),
PreDef.funcEnd(stop=True)
)
namedcolor = Prod(name='Named Color',
match=lambda t, v: t == 'IDENT' and (
normalize(v) in as_list(self.COLORS.keys())
),
stop=True)
prods = Choice(PreDef.hexcolor(stop=True),
namedcolor,
noalp,
witha)
ok, seq, store, unused = ProdParser().parse(cssText,
self.type,
prods)
self.wellformed = ok
if ok:
t, v = seq[0].type, seq[0].value
if 'IDENT' == t:
rgba = self.COLORS[normalize(v)]
if 'HASH' == t:
if len(v) == 4:
# HASH #rgb
rgba = (int(2*v[1], 16),
int(2*v[2], 16),
int(2*v[3], 16),
1.0)
else:
# HASH #rrggbb
rgba = (int(v[1:3], 16),
int(v[3:5], 16),
int(v[5:7], 16),
1.0)
elif 'FUNCTION' == t:
functiontype, raw, check = None, [], ''
HSL = False
for item in seq:
try:
type_ = item.value.type
except AttributeError:
# type of function, e.g. rgb(
if item.type == 'FUNCTION':
functiontype = item.value
HSL = functiontype in ('hsl(', 'hsla(')
continue
# save components
if type_ == Value.NUMBER:
raw.append(item.value.value)
check += 'N'
elif type_ == Value.PERCENTAGE:
if HSL:
# save as percentage fraction
raw.append(item.value.value / 100.0)
else:
# save as real value of percentage of 255
raw.append(int(255 * item.value.value / 100))
check += 'P'
if HSL:
# convert to rgb
# h is 360 based (circle)
h, s, l_ = raw[0] / 360.0, raw[1], raw[2]
# ORDER h l_ s !!!
r, g, b = colorsys.hls_to_rgb(h, l_, s)
# back to 255 based
rgba = [int(round(r*255)),
int(round(g*255)),
int(round(b*255))]
if len(raw) > 3:
rgba.append(raw[3])
else:
# rgb, rgba
rgba = raw
if len(rgba) < 4:
rgba.append(1.0)
# validate
checks = {'rgb(': ('NNN', 'PPP'),
'rgba(': ('NNNN', 'PPPN'),
'hsl(': ('NPP',),
'hsla(': ('NPPN',)
}
if check not in checks[functiontype]:
self._log.error('ColorValue has invalid %s) parameters: '
'%s (N=Number, P=Percentage)' %
(functiontype, check))
self._colorType = t
self._red, self._green, self._blue, self._alpha = tuple(rgba)
self._setSeq(seq)
cssText = property(lambda self: css_parser.ser.do_css_ColorValue(self),
_setCssText,
doc="String value of this value.")
value = property(lambda self: css_parser.ser.do_css_CSSFunction(self, True),
doc='Same as cssText but without comments.')
type = property(lambda self: Value.COLOR_VALUE,
doc="Type is fixed to Value.COLOR_VALUE.")
def _getName(self):
for n, v in as_list(self.COLORS.items()):
if v == (self.red, self.green, self.blue, self.alpha):
return n
colorType = property(lambda self: self._colorType,
doc="IDENT (red), HASH (#f00) or FUNCTION (rgb(255, 0, 0).")
name = property(_getName,
doc='Name of the color if known (in ColorValue.COLORS) '
'else None')
red = property(lambda self: self._red,
doc='red part as integer between 0 and 255')
green = property(lambda self: self._green,
doc='green part as integer between 0 and 255')
blue = property(lambda self: self._blue,
doc='blue part as integer between 0 and 255')
alpha = property(lambda self: self._alpha,
doc='alpha part as float between 0.0 and 1.0')
class DimensionValue(Value):
"""
A numerical value with an optional dimension like e.g. "px" or "%".
Covers DIMENSION, PERCENTAGE or NUMBER values.
"""
__reUnNumDim = re.compile(r'^([+-]?)(\d*\.\d+|\d+)(.*)$', re.I | re.U | re.X)
_dimension = None
_sign = None
def __str__(self):
return ""\
% (self.__class__.__name__,
self.type, self.value, self.dimension, self.cssText,
id(self))
def _setCssText(self, cssText):
self._checkReadonly()
prods = Sequence( # PreDef.unary(),
Choice(PreDef.dimension(stop=True),
PreDef.number(stop=True),
PreDef.percentage(stop=True)
)
)
ok, seq, store, unused = ProdParser().parse(cssText,
'DimensionValue',
prods)
self.wellformed = ok
if ok:
item = seq[0]
sign, v, d = self.__reUnNumDim.findall(
normalize(item.value))[0]
if '.' in v:
val = float(sign + v)
else:
val = int(sign + v)
dim = None
if d:
dim = d
self._sign = sign
self._value = val
self._dimension = dim
self._type = item.type
self._setSeq(seq)
cssText = property(lambda self: css_parser.ser.do_css_Value(self),
_setCssText,
doc="String value of this value including dimension.")
dimension = property(lambda self: self._dimension, # _setValue,
doc="Dimension if a DIMENSION or PERCENTAGE value, "
"else None")
class URIValue(Value):
"""
An URI value like ``url(example.png)``.
"""
_type = Value.URI
_uri = Value._value
def __str__(self):
return ""\
% (self.__class__.__name__,
self.type, self.value, self.uri, self.cssText,
id(self))
def _setCssText(self, cssText):
self._checkReadonly()
prods = Sequence(PreDef.uri(stop=True))
ok, seq, store, unused = ProdParser().parse(cssText, 'URIValue', prods)
self.wellformed = ok
if ok:
# only 1 value only anyway
self._type = seq[0].type
self._value = seq[0].value
self._setSeq(seq)
cssText = property(lambda self: css_parser.ser.do_css_Value(self),
_setCssText,
doc='String value of this value.')
def _setUri(self, uri):
# TODO: check?
self._value = uri
uri = property(lambda self: self._value, _setUri,
doc="Actual URL without delimiters or the empty string")
def absoluteUri(self):
"""Actual URL, made absolute if possible, else same as `uri`."""
# Ancestry: PropertyValue, Property, CSSStyleDeclaration, CSSStyleRule,
# CSSStyleSheet
try:
# TODO: better way?
styleSheet = self.parent.parent.parent.parentRule.parentStyleSheet
except AttributeError:
return self.uri
else:
return urllib_urljoin(styleSheet.href, self.uri)
absoluteUri = property(absoluteUri, doc=absoluteUri.__doc__)
class CSSFunction(Value):
"""
A function value.
"""
_functionName = 'Function'
def _productions(self):
"""Return definition used for parsing."""
types = self._prods # rename!
itemProd = Choice(_ColorProd(self),
_DimensionProd(self),
_URIProd(self),
_ValueProd(self),
_CalcValueProd(self),
_CSSVariableProd(self),
_CSSFunctionProd(self)
)
funcProds = Sequence(Prod(name='FUNCTION',
match=lambda t, v: t == types.FUNCTION,
toSeq=lambda t, tokens: (t[0],
normalize(t[1]))),
Choice(Sequence(itemProd,
Sequence(PreDef.comma(optional=True),
itemProd,
minmax=lambda: (0, None)),
PreDef.funcEnd(stop=True)),
PreDef.funcEnd(stop=True))
)
return funcProds
def _setCssText(self, cssText):
self._checkReadonly()
ok, seq, store, unused = ProdParser().parse(cssText,
self.type,
self._productions())
self.wellformed = ok
if ok:
self._setSeq(seq)
cssText = property(lambda self: css_parser.ser.do_css_CSSFunction(self),
_setCssText,
doc="String value of this value.")
value = property(lambda self: css_parser.ser.do_css_CSSFunction(self, True),
doc='Same as cssText but without comments.')
type = property(lambda self: Value.FUNCTION,
doc="Type is fixed to Value.FUNCTION.")
class MSValue(CSSFunction):
"""An IE specific Microsoft only function value which is much looser
in what is syntactically allowed."""
_functionName = 'MSValue'
def _productions(self):
"""Return definition used for parsing."""
types = self._prods # rename!
func = Prod(name='MSValue-Sub',
match=lambda t, v: t == self._prods.FUNCTION,
toSeq=lambda t, tokens: (MSValue._functionName,
MSValue(pushtoken(t,
tokens
),
parent=self
)
)
)
funcProds = Sequence(Prod(name='FUNCTION',
match=lambda t, v: t == types.FUNCTION,
toSeq=lambda t, tokens: (t[0], t[1])
),
Sequence(Choice(_ColorProd(self),
_DimensionProd(self),
_URIProd(self),
_ValueProd(self),
_MSValueProd(self),
# _CalcValueProd(self),
_CSSVariableProd(self),
func,
# _CSSFunctionProd(self),
Prod(name='MSValuePart',
match=lambda t, v: v != ')',
toSeq=lambda t, tokens: (t[0], t[1])
)
),
minmax=lambda: (0, None)
),
PreDef.funcEnd(stop=True)
)
return funcProds
def _setCssText(self, cssText):
super(MSValue, self)._setCssText(cssText)
cssText = property(lambda self: css_parser.ser.do_css_MSValue(self),
_setCssText,
doc="String value of this value.")
class CSSCalc(CSSFunction):
"""The CSSCalc function represents a CSS calc() function.
No further API is provided. For multiplication and division no check
if one operand is a NUMBER is made.
"""
_functionName = 'CSSCalc'
def __str__(self):
return "" % (
self.__class__.__name__, id(self))
def _setCssText(self, cssText):
self._checkReadonly()
types = self._prods # rename!
_operator = Choice(Prod(name='Operator */',
match=lambda t, v: v in '*/',
toSeq=lambda t, tokens: (t[0], t[1])
),
Sequence(
PreDef.S(),
Choice(
Sequence(
Prod(name='Operator */',
match=lambda t, v: v in '*/',
toSeq=lambda t, tokens: (t[0], t[1])
),
PreDef.S(optional=True)
),
Sequence(
Prod(name='Operator +-',
match=lambda t, v: v in '+-',
toSeq=lambda t, tokens: (t[0], t[1])
),
PreDef.S()
),
PreDef.funcEnd(stop=True, mayEnd=True)
)
)
)
def _operant(): return Choice(_DimensionProd(self),
_CSSVariableProd(self))
prods = Sequence(Prod(name='CALC',
match=lambda t, v: t == types.FUNCTION and
normalize(v) == 'calc('
),
PreDef.S(optional=True),
_operant(),
Sequence(_operator,
_operant(),
minmax=lambda: (0, None)
),
PreDef.funcEnd(stop=True)
)
# store: name of variable
ok, seq, store, unused = ProdParser().parse(cssText,
'CSSCalc',
prods,
checkS=True)
self.wellformed = ok
if ok:
self._setSeq(seq)
cssText = property(lambda self: css_parser.ser.do_css_CSSCalc(self),
_setCssText, doc="String representation of calc function.")
type = property(lambda self: Value.CALC,
doc="Type is fixed to Value.CALC.")
class CSSVariable(CSSFunction):
"""The CSSVariable represents a CSS variables like ``var(varname)``.
A variable has a (nonnormalized!) `name` and a `value` which is
tried to be resolved from any available CSSVariablesRule definition.
"""
_functionName = 'CSSVariable'
_name = None
_fallback = None
def __str__(self):
return "" % (
self.__class__.__name__, self.name, self.value, id(self))
def _setCssText(self, cssText):
self._checkReadonly()
types = self._prods # rename!
prods = Sequence(Prod(name='var',
match=lambda t, v: t == types.FUNCTION and
normalize(v) == 'var('
),
PreDef.ident(toStore='ident'),
Sequence(PreDef.comma(),
Choice(_ColorProd(self, toStore='fallback'),
_DimensionProd(self, toStore='fallback'),
_URIProd(self, toStore='fallback'),
_ValueProd(self, toStore='fallback'),
_CalcValueProd(self, toStore='fallback'),
_CSSVariableProd(self, toStore='fallback'),
_CSSFunctionProd(self, toStore='fallback')
),
minmax=lambda: (0, 1)
),
PreDef.funcEnd(stop=True))
# store: name of variable
store = {'ident': None, 'fallback': None}
ok, seq, store, unused = ProdParser().parse(cssText,
'CSSVariable',
prods)
self.wellformed = ok
if ok:
self._name = store['ident'].value
try:
self._fallback = store['fallback'].value
except KeyError:
self._fallback = None
self._setSeq(seq)
cssText = property(lambda self: css_parser.ser.do_css_CSSVariable(self),
_setCssText, doc="String representation of variable.")
# TODO: writable? check if var (value) available?
name = property(lambda self: self._name,
doc="The name identifier of this variable referring to "
"a value in a "
":class:`css_parser.css.CSSVariablesDeclaration`.")
fallback = property(lambda self: self._fallback,
doc="The fallback Value of this variable")
type = property(lambda self: Value.VARIABLE,
doc="Type is fixed to Value.VARIABLE.")
def _getValue(self):
"Find contained sheet and @variables there"
rel = self
while True:
# find node which has parentRule to get to StyleSheet
if hasattr(rel, 'parent'):
rel = rel.parent
else:
break
try:
variables = rel.parentRule.parentStyleSheet.variables
except AttributeError:
return None
else:
try:
return variables[self.name]
except KeyError:
return None
value = property(_getValue,
doc='The resolved actual value or None.')
# helper for productions
def _ValueProd(parent, nextSor=False, toStore=None):
return Prod(name='Value',
match=lambda t, v: t in ('IDENT', 'STRING', 'UNICODE-RANGE'),
nextSor=nextSor,
toStore=toStore,
toSeq=lambda t, tokens: ('Value', Value(pushtoken(t, tokens),
parent=parent)
)
)
def _DimensionProd(parent, nextSor=False, toStore=None):
return Prod(name='Dimension',
match=lambda t, v: t in ('DIMENSION',
'NUMBER',
'PERCENTAGE'),
nextSor=nextSor,
toStore=toStore,
toSeq=lambda t, tokens: ('DIMENSION', DimensionValue(
pushtoken(t,
tokens),
parent=parent)
)
)
def _URIProd(parent, nextSor=False, toStore=None):
return Prod(name='URIValue',
match=lambda t, v: t == 'URI',
toStore=toStore,
nextSor=nextSor,
toSeq=lambda t, tokens: ('URIValue', URIValue(
pushtoken(t,
tokens),
parent=parent)
)
)
reHexcolor = re.compile(r'^\#(?:[0-9abcdefABCDEF]{3}|[0-9abcdefABCDEF]{6})$')
def _ColorProd(parent, nextSor=False, toStore=None):
return Prod(name='ColorValue',
match=lambda t, v:
(t == 'HASH' and
reHexcolor.match(v)
) or
(t == 'FUNCTION' and
normalize(v) in ('rgb(',
'rgba(',
'hsl(',
'hsla(')
) or
(t == 'IDENT' and
normalize(v) in as_list(ColorValue.COLORS.keys())
),
nextSor=nextSor,
toStore=toStore,
toSeq=lambda t, tokens: ('ColorValue', ColorValue(
pushtoken(t,
tokens),
parent=parent)
)
)
def _CSSFunctionProd(parent, nextSor=False, toStore=None):
return PreDef.function(nextSor=nextSor,
toStore=toStore,
toSeq=lambda t, tokens: (CSSFunction._functionName,
CSSFunction(
pushtoken(t, tokens),
parent=parent)
)
)
def _CalcValueProd(parent, nextSor=False, toStore=None):
return Prod(name=CSSCalc._functionName,
match=lambda t, v: t == PreDef.types.FUNCTION and
normalize(v) == 'calc(',
toStore=toStore,
toSeq=lambda t, tokens: (CSSCalc._functionName,
CSSCalc(
pushtoken(t, tokens),
parent=parent)
),
nextSor=nextSor)
def _CSSVariableProd(parent, nextSor=False, toStore=None):
return PreDef.variable(nextSor=nextSor,
toStore=toStore,
toSeq=lambda t, tokens: (CSSVariable._functionName,
CSSVariable(
pushtoken(t, tokens),
parent=parent)
)
)
def _MSValueProd(parent, nextSor=False):
return Prod(name=MSValue._functionName,
match=lambda t, v: ( # t == self._prods.FUNCTION and (
normalize(v) in ('expression(',
'alpha(',
'blur(',
'chroma(',
'dropshadow(',
'fliph(',
'flipv(',
'glow(',
'gray(',
'invert(',
'mask(',
'shadow(',
'wave(',
'xray(') or
v.startswith('progid:DXImageTransform.Microsoft.')
),
nextSor=nextSor,
toSeq=lambda t, tokens: (MSValue._functionName,
MSValue(pushtoken(t,
tokens
),
parent=parent
)
)
)
def MediaQueryValueProd(parent):
return Choice(_ColorProd(parent),
_DimensionProd(parent),
_ValueProd(parent),
PreDef.ratio(),
)
css-parser-1.0.4/src/css_parser/cssproductions.py 0000644 0001750 0001750 00000011471 13407702010 022435 0 ustar kovid kovid 0000000 0000000 """productions for css_parser based on a mix of CSS 2.1 and CSS 3 Syntax
productions
- http://www.w3.org/TR/css3-syntax
- http://www.w3.org/TR/css3-syntax/#grammar0
open issues
- numbers contain "-" if present
- HASH: #aaa is, #000 is not anymore,
CSS2.1: 'nmchar': r'[_a-z0-9-]|{nonascii}|{escape}',
CSS3: 'nmchar': r'[_a-z-]|{nonascii}|{escape}',
"""
from __future__ import unicode_literals, division, absolute_import, print_function
__all__ = ['CSSProductions', 'MACROS', 'PRODUCTIONS']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
# a complete list of css3 macros
MACROS = {
'nonascii': r'[^\0-\177]',
'unicode': r'\\[0-9A-Fa-f]{1,6}(?:{nl}|{s})?',
# 'escape': r'{unicode}|\\[ -~\200-\777]',
'escape': r'{unicode}|\\[^\n\r\f0-9a-f]',
'nmstart': r'[_a-zA-Z]|{nonascii}|{escape}',
'nmchar': r'[-_a-zA-Z0-9]|{nonascii}|{escape}',
'string1': r'"([^\n\r\f\\"]|\\{nl}|{escape})*"',
'string2': r"'([^\n\r\f\\']|\\{nl}|{escape})*'",
'invalid1': r'\"([^\n\r\f\\"]|\\{nl}|{escape})*',
'invalid2': r"\'([^\n\r\f\\']|\\{nl}|{escape})*",
'comment': r'\/\*[^*]*\*+([^/][^*]*\*+)*\/',
'ident': r'[-]?{nmstart}{nmchar}*',
'name': r'{nmchar}+',
# TODO???
'num': r'[+-]?[0-9]*\.[0-9]+|[+-]?[0-9]+', # r'[-]?\d+|[-]?\d*\.\d+',
'string': r'{string1}|{string2}',
# from CSS2.1
'invalid': r'{invalid1}|{invalid2}',
'url': r'[\x09\x21\x23-\x26\x28\x2a-\x7E]|{nonascii}|{escape}',
's': r'\t|\r|\n|\f|\x20',
'w': r'{s}*',
'nl': r'\n|\r\n|\r|\f',
'A': r'A|a|\\0{0,4}(?:41|61)(?:\r\n|[ \t\r\n\f])?',
'B': r'B|b|\\0{0,4}(?:42|62)(?:\r\n|[ \t\r\n\f])?',
'C': r'C|c|\\0{0,4}(?:43|63)(?:\r\n|[ \t\r\n\f])?',
'D': r'D|d|\\0{0,4}(?:44|64)(?:\r\n|[ \t\r\n\f])?',
'E': r'E|e|\\0{0,4}(?:45|65)(?:\r\n|[ \t\r\n\f])?',
'F': r'F|f|\\0{0,4}(?:46|66)(?:\r\n|[ \t\r\n\f])?',
'G': r'G|g|\\0{0,4}(?:47|67)(?:\r\n|[ \t\r\n\f])?|\\G|\\g',
'H': r'H|h|\\0{0,4}(?:48|68)(?:\r\n|[ \t\r\n\f])?|\\H|\\h',
'I': r'I|i|\\0{0,4}(?:49|69)(?:\r\n|[ \t\r\n\f])?|\\I|\\i',
'K': r'K|k|\\0{0,4}(?:4b|6b)(?:\r\n|[ \t\r\n\f])?|\\K|\\k',
'L': r'L|l|\\0{0,4}(?:4c|6c)(?:\r\n|[ \t\r\n\f])?|\\L|\\l',
'M': r'M|m|\\0{0,4}(?:4d|6d)(?:\r\n|[ \t\r\n\f])?|\\M|\\m',
'N': r'N|n|\\0{0,4}(?:4e|6e)(?:\r\n|[ \t\r\n\f])?|\\N|\\n',
'O': r'O|o|\\0{0,4}(?:4f|6f)(?:\r\n|[ \t\r\n\f])?|\\O|\\o',
'P': r'P|p|\\0{0,4}(?:50|70)(?:\r\n|[ \t\r\n\f])?|\\P|\\p',
'R': r'R|r|\\0{0,4}(?:52|72)(?:\r\n|[ \t\r\n\f])?|\\R|\\r',
'S': r'S|s|\\0{0,4}(?:53|73)(?:\r\n|[ \t\r\n\f])?|\\S|\\s',
'T': r'T|t|\\0{0,4}(?:54|74)(?:\r\n|[ \t\r\n\f])?|\\T|\\t',
'U': r'U|u|\\0{0,4}(?:55|75)(?:\r\n|[ \t\r\n\f])?|\\U|\\u',
'V': r'V|v|\\0{0,4}(?:56|76)(?:\r\n|[ \t\r\n\f])?|\\V|\\v',
'X': r'X|x|\\0{0,4}(?:58|78)(?:\r\n|[ \t\r\n\f])?|\\X|\\x',
'Z': r'Z|z|\\0{0,4}(?:5a|7a)(?:\r\n|[ \t\r\n\f])?|\\Z|\\z',
}
# The following productions are the complete list of tokens
# used by css_parser, a mix of CSS3 and some CSS2.1 productions.
# The productions are **ordered**:
PRODUCTIONS = [
# UTF8_BOM or UTF8_BOM_SIG will only be checked at beginning of CSS
('BOM', '\xfe\xff|\xef\xbb\xbf'),
('S', r'{s}+'), # 1st in list of general productions
('URI', r'{U}{R}{L}\({w}({string}|{url}*){w}\)'),
('RATIO', r'[^\(]{w}[0-9]+{w}\/{w}[0-9]+'),
('UNICODE-RANGE', r'{U}\+[0-9A-Fa-f?]{1,6}(\-[0-9A-Fa-f]{1,6})?'),
('IDENT', r'{ident}'),
('FUNCTION', r'{ident}\('),
('DIMENSION', r'{num}{ident}'),
('PERCENTAGE', r'{num}\%'),
('NUMBER', r'{num}'),
('HASH', r'\#{name}'),
('COMMENT', r'{comment}'), # r'\/\*[^*]*\*+([^/][^*]*\*+)*\/'),
('STRING', r'{string}'),
('INVALID', r'{invalid}'), # from CSS2.1
('ATKEYWORD', r'@{ident}'), # other keywords are done in the tokenizer
('INCLUDES', r'\~\='),
('DASHMATCH', r'\|\='),
('PREFIXMATCH', r'\^\='),
('SUFFIXMATCH', r'\$\='),
('SUBSTRINGMATCH', r'\*\='),
('CDO', r'\<\!\-\-'),
('CDC', r'\-\-\>'),
('CHAR', r'[^"\']') # MUST always be last
# valid ony at start so not checked everytime
# ('CHARSET_SYM', r'@charset '), # from Errata includes ending space!
# checked specially if fullsheet is parsed
]
class CSSProductions(object):
"""
most attributes are set later
"""
EOF = True
# removed from productions as they simply are ATKEYWORD until
# tokenizing
CHARSET_SYM = 'CHARSET_SYM'
FONT_FACE_SYM = 'FONT_FACE_SYM'
MEDIA_SYM = 'MEDIA_SYM'
IMPORT_SYM = 'IMPORT_SYM'
NAMESPACE_SYM = 'NAMESPACE_SYM'
PAGE_SYM = 'PAGE_SYM'
VARIABLES_SYM = 'VARIABLES_SYM'
for i, t in enumerate(PRODUCTIONS):
setattr(CSSProductions, t[0].replace('-', '_'), t[0])
# may be enabled by settings.set
_DXImageTransform = ('FUNCTION',
r'progid\:DXImageTransform\.Microsoft\..+\('
)
css-parser-1.0.4/src/css_parser/encutils/ 0000755 0001750 0001750 00000000000 13413156014 020627 5 ustar kovid kovid 0000000 0000000 css-parser-1.0.4/src/css_parser/encutils/__init__.py 0000644 0001750 0001750 00000054724 13407702010 022750 0 ustar kovid kovid 0000000 0000000 #!/usr/bin/env python
# -*- coding: utf-8 -*-
"""encutils - encoding detection collection for Python
:Version: 0.9.8
:Author: Christof Hoeke, see http://cthedot.de/encutils/
:Contributor: Robert Siemer, Fredrik Hedman ported to python3
:Copyright: 2005-2012: Christof Hoeke
:License: encutils has a dual-license, please choose whatever you prefer:
* encutils is published under the
`LGPL 3 or later `__
* encutils is published under the
`Creative Commons License `__.
encutils is free software: you can redistribute it and/or modify
it under the terms of the GNU Lesser General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
encutils 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 Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License
along with encutils. If not, see .
A collection of helper functions to detect encodings of text files (like HTML, XHTML, XML, CSS, etc.)
retrieved via HTTP, file or string.
:func:`getEncodingInfo` is probably the main function of interest which uses
other supplied functions itself and gathers all information together and
supplies an :class:`EncodingInfo` object.
example::
>>> import encutils
>>> info = encutils.getEncodingInfo(url='http://cthedot.de/encutils/')
>>> str(info)
utf-8
>>> repr(info) # doctest:+ELLIPSIS
>>> info.logtext
HTTP media_type: text/html
HTTP encoding: utf-8
Encoding (probably): utf-8 (Mismatch: False)
references
XML
RFC 3023 (http://www.ietf.org/rfc/rfc3023.txt)
easier explained in
- http://feedparser.org/docs/advanced.html
- http://www.xml.com/pub/a/2004/07/21/dive.html
HTML
http://www.w3.org/TR/REC-html40/charset.html#h-5.2.2
TODO
- parse @charset of HTML elements?
- check for more texttypes if only text given
"""
from __future__ import unicode_literals, division, absolute_import, print_function
__all__ = [
'buildlog', 'encodingByMediaType', 'getHTTPInfo', 'getMetaInfo',
'detectXMLEncoding', 'getEncodingInfo', 'tryEncodings', 'EncodingInfo'
]
__docformat__ = 'restructuredtext'
__author__ = 'Christof Hoeke, Robert Siemer, Fredrik Hedman'
__version__ = '$Id$'
import sys
import cgi
import re
import types
PY2x = sys.version_info < (3, 0)
if PY2x:
from StringIO import StringIO as io_StringIO
from HTMLParser import HTMLParser as htmlparser_HTMLParser
from HTMLParser import HTMLParseError as htmlparser_HTMLParseError
from urllib2 import urlopen as urllib_urlopen
else:
from io import StringIO as io_StringIO
from html.parser import HTMLParser as htmlparser_HTMLParser
from urllib.request import urlopen as urllib_urlopen
VERSION = '0.9.8'
class _MetaHTMLParser(htmlparser_HTMLParser):
"""Parse given data for ."""
content_type = None
def handle_starttag(self, tag, attrs):
if tag == 'meta' and not self.content_type:
atts = dict([(a.lower(), v.lower()) for a, v in attrs])
if atts.get('http-equiv', '').strip() == 'content-type':
self.content_type = atts.get('content')
# application/xml, application/xml-dtd, application/xml-external-parsed-entity, or a subtype like application/rss+xml.
_XML_APPLICATION_TYPE = 0
# text/xml, text/xml-external-parsed-entity, or a subtype like text/AnythingAtAll+xml
_XML_TEXT_TYPE = 1
# text/html
_HTML_TEXT_TYPE = 2
# any other of text/* like text/plain, ...
_TEXT_TYPE = 3
# any text/* like which defaults to UTF-8 encoding, for now only text/css
_TEXT_UTF8 = 5
# types not fitting in above types
_OTHER_TYPE = 4
class EncodingInfo(object):
"""
All encoding related information, returned by :func:`getEncodingInfo`.
Attributes filled:
- ``encoding``: The guessed encoding
Encoding is the explicit or implicit encoding or None and
always lowercase.
- from HTTP response
* ``http_encoding``
* ``http_media_type``
- from HTML element
* ``meta_encoding``
* ``meta_media_type``
- from XML declaration
* ``xml_encoding``
- ``mismatch``: True if mismatch between XML declaration and HTTP
header.
Mismatch is True if any mismatches between HTTP header, XML
declaration or textcontent (meta) are found. More detailed
mismatch reports are written to the optional log or ``logtext``
Mismatches are not necessarily errors as preferences are defined.
For details see the specifications.
- ``logtext``: if no log was given log reports are given here
"""
def __init__(self):
"""Initialize all possible properties to ``None``, see class
description
"""
self.encoding = self.mismatch = self.logtext =\
self.http_encoding = self.http_media_type =\
self.meta_encoding = self.meta_media_type =\
self.xml_encoding = None
def __str__(self):
"""Output the guessed encoding itself or the empty string."""
if self.encoding:
return self.encoding
else:
return ''
def __repr__(self):
return "<%s.%s object encoding=%r mismatch=%s at 0x%x>" % (
self.__class__.__module__, self.__class__.__name__, self.encoding,
self.mismatch, id(self))
def buildlog(logname='encutils',
level='INFO',
stream=sys.stderr,
filename=None,
filemode="w",
format='%(levelname)s\t%(message)s'):
"""Helper to build a basic log
- if `filename` is given returns a log logging to `filename` with
mode `filemode`
- else uses a log streaming to `stream` which defaults to
`sys.stderr`
- `level` defines the level of the log
- `format` defines the formatter format of the log
:returns:
a log with the name `logname`
"""
import logging
log = logging.getLogger(logname)
if filename:
hdlr = logging.FileHandler(filename, filemode)
else:
hdlr = logging.StreamHandler(stream)
formatter = logging.Formatter(format)
hdlr.setFormatter(formatter)
log.addHandler(hdlr)
log.setLevel(logging.__dict__.get(level, logging.INFO))
return log
def _getTextTypeByMediaType(media_type, log=None):
"""
:returns:
type as defined by constants in this class
"""
if not media_type:
return _OTHER_TYPE
xml_application_types = [
r'application/.*?\+xml', 'application/xml', 'application/xml-dtd',
'application/xml-external-parsed-entity'
]
xml_text_types = [
r'text\/.*?\+xml', 'text/xml', 'text/xml-external-parsed-entity'
]
media_type = media_type.strip().lower()
if media_type in xml_application_types or\
re.match(xml_application_types[0], media_type, re.I | re.S | re.X):
return _XML_APPLICATION_TYPE
elif media_type in xml_text_types or\
re.match(xml_text_types[0], media_type, re.I | re.S | re.X):
return _XML_TEXT_TYPE
elif media_type == 'text/html':
return _HTML_TEXT_TYPE
elif media_type == 'text/css':
return _TEXT_UTF8
elif media_type.startswith('text/'):
return _TEXT_TYPE
else:
return _OTHER_TYPE
def _getTextType(text, log=None):
"""Check if given text is XML (**naive test!**)
used if no content-type given
"""
if text[:30].find('`` element if available in `text`.
XHTML format::
"""
p = _MetaHTMLParser()
try:
p.feed(text)
except htmlparser_HTMLParseError:
pass
if p.content_type:
media_type, params = cgi.parse_header(p.content_type)
encoding = params.get('charset') # defaults to None
if encoding:
encoding = encoding.lower()
if log:
log.info('HTML META media_type: %s', media_type)
log.info('HTML META encoding: %s', encoding)
else:
media_type = encoding = None
return media_type, encoding
def detectXMLEncoding(fp, log=None, includeDefault=True):
"""Attempt to detect the character encoding of the xml file
given by a file object `fp`. `fp` must not be a codec wrapped file
object! `fp` may be a string or unicode string though.
Based on a recipe by Lars Tiede:
http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/363841
which itself is based on Paul Prescotts recipe:
http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/52257
:returns:
- if detection of the BOM succeeds, the codec name of the
corresponding unicode charset is returned
- if BOM detection fails, the xml declaration is searched for
the encoding attribute and its value returned. the "<"
character has to be the very first in the file then (it's xml
standard after all).
- if BOM and xml declaration fail, utf-8 is returned according
to XML 1.0.
"""
if PY2x and isinstance(fp, types.StringTypes):
fp = io_StringIO(fp)
elif isinstance(fp, str):
fp = io_StringIO(fp)
# detection using BOM
# the BOMs we know, by their pattern
bomDict = { # bytepattern: name
(0x00, 0x00, 0xFE, 0xFF): "utf_32_be",
(0xFF, 0xFE, 0x00, 0x00): "utf_32_le",
(0xFE, 0xFF, None, None): "utf_16_be",
(0xFF, 0xFE, None, None): "utf_16_le",
(0xEF, 0xBB, 0xBF, None): "utf-8",
}
# go to beginning of file and get the first 4 bytes
oldFP = fp.tell()
fp.seek(0)
(byte1, byte2, byte3, byte4) = tuple(map(ord, fp.read(4)))
# try bom detection using 4 bytes, 3 bytes, or 2 bytes
bomDetection = bomDict.get((byte1, byte2, byte3, byte4))
if not bomDetection:
bomDetection = bomDict.get((byte1, byte2, byte3, None))
if not bomDetection:
bomDetection = bomDict.get((byte1, byte2, None, None))
# if BOM detected, we're done :-)
if bomDetection:
if log:
log.info('XML BOM encoding: %s' % bomDetection)
fp.seek(oldFP)
return bomDetection
# still here? BOM detection failed.
# now that BOM detection has failed we assume one byte character
# encoding behaving ASCII
# search xml declaration for encoding attribute
# assume xml declaration fits into the first 2 KB (*cough*)
fp.seek(0)
buffer = fp.read(2048)
# set up regular expression
xmlDeclPattern = r"""
^<\?xml # w/o BOM, xmldecl starts with # what's matched in the brackets will be named encstr
[^"']+ # every character not delimiter (not overly exact!)
) # closes the brackets pair for the named group
["'] # attribute end delimiter
.*? # some chars optionally (standalone decl or whitespace)
\?> # xmldecl end
"""
xmlDeclRE = re.compile(xmlDeclPattern, re.VERBOSE)
# search and extract encoding string
match = xmlDeclRE.search(buffer)
fp.seek(oldFP)
if match:
enc = match.group("encstr").lower()
if log:
log.info('XML encoding="%s"' % enc)
return enc
else:
if includeDefault:
if log:
log.info('XML encoding default utf-8')
return 'utf-8'
else:
return None
def tryEncodings(text, log=None):
"""If installed uses chardet http://chardet.feedparser.org/ to detect
encoding, else tries different encodings on `text` and returns the one
that does not raise an exception which is not very advanced or may
be totally wrong. The tried encoding are in order 'ascii', 'iso-8859-1',
'windows-1252' (which probably will never happen as 'iso-8859-1' can decode
these strings too) and lastly 'utf-8'.
:param text:
a byte string
:returns:
Working encoding or ``None`` if no encoding does work at all.
The returned encoding might nevertheless be not the one intended by
the author as it is only checked if the text might be encoded in
that encoding. Some texts might be working in "iso-8859-1" *and*
"windows-1252" *and* "ascii" *and* "utf-8" and ...
"""
try:
import chardet
encoding = chardet.detect(text)["encoding"]
except ImportError:
msg = 'Using simplified encoding detection, you might want to install chardet.'
if log:
log.warning(msg)
else:
print(msg)
encodings = (
'ascii',
'iso-8859-1',
# 'windows-1252', # test later
'utf-8')
encoding = None
for e in encodings:
try:
text.decode(e)
except UnicodeDecodeError:
pass
else:
if 'iso-8859-1' == e:
try:
if '€' in text.decode('windows-1252'):
return 'windows-1252'
except UnicodeDecodeError:
pass
return e
return encoding
def getEncodingInfo(response=None, text='', log=None, url=None):
"""Find all encoding related information in given `text`.
Information in headers of supplied HTTPResponse, possible XML
declaration and X/HTML ```` elements are used.
:param response:
HTTP response object, e.g. via ``urllib.urlopen('url')``
:param text:
a byte string to guess encoding for. XML prolog with
encoding pseudo attribute or HTML meta element will be used to detect
the encoding
:param url:
When given fetches document at `url` and all needed information.
No `reponse` or `text` parameters are needed in this case.
:param log:
an optional logging logger to which messages may go, if
no log given all log messages are available from resulting
``EncodingInfo``
:returns:
instance of :class:`EncodingInfo`.
How the resulting encoding is retrieved:
XML
RFC 3023 states if media type given in the Content-Type HTTP header is
application/xml, application/xml-dtd,
application/xml-external-parsed-entity, or any one of the subtypes of
application/xml such as application/atom+xml or application/rss+xml
etc then the character encoding is determined in this order:
1. the encoding given in the charset parameter of the Content-Type HTTP
header, or
2. the encoding given in the encoding attribute of the XML declaration
within the document, or
3. utf-8.
Mismatch possibilities:
- HTTP + XMLdecla
- HTTP + HTMLmeta
application/xhtml+xml ?
XMLdecla + HTMLmeta
If the media type given in the Content-Type HTTP header is text/xml,
text/xml-external-parsed-entity, or a subtype like text/Anything+xml,
the encoding attribute of the XML declaration is ignored completely
and the character encoding is determined in the order:
1. the encoding given in the charset parameter of the Content-Type HTTP
header, or
2. ascii.
No mismatch possible.
If no media type is given the XML encoding pseuso attribute is used
if present.
No mismatch possible.
HTML
For HTML served as text/html:
http://www.w3.org/TR/REC-html40/charset.html#h-5.2.2
1. An HTTP "charset" parameter in a "Content-Type" field.
(maybe defaults to ISO-8859-1, but should not assume this)
2. A META declaration with "http-equiv" set to "Content-Type" and a
value set for "charset".
3. The charset attribute set on an element that designates an external
resource. (NOT IMPLEMENTED HERE YET)
Mismatch possibilities:
- HTTP + HTMLmeta
TEXT
For most text/* types the encoding will be reported as iso-8859-1.
Exceptions are XML formats send as text/* mime type (see above) and
text/css which has a default encoding of UTF-8.
"""
if url:
# may cause IOError which is raised
response = urllib_urlopen(url)
if text is None:
# read text from response only if not explicitly given
try:
text = response.read()
except IOError:
pass
if text is None:
# text must be a string (not None)
text = ''
encinfo = EncodingInfo()
logstream = io_StringIO()
if not log:
log = buildlog(stream=logstream, format='%(message)s')
# HTTP
if response:
encinfo.http_media_type, encinfo.http_encoding = getHTTPInfo(
response, log)
texttype = _getTextTypeByMediaType(encinfo.http_media_type, log)
else:
# check if maybe XML or (TODO:) HTML
texttype = _getTextType(text, log)
# XML only served as application/xml ! #(also XHTML served as text/html)
if texttype == _XML_APPLICATION_TYPE: # or texttype == _XML_TEXT_TYPE:
try:
encinfo.xml_encoding = detectXMLEncoding(text, log)
except (AttributeError, ValueError):
encinfo.xml_encoding = None
# XML (also XHTML served as text/html)
if texttype == _HTML_TEXT_TYPE:
try:
encinfo.xml_encoding = detectXMLEncoding(
text, log, includeDefault=False)
except (AttributeError, ValueError):
encinfo.xml_encoding = None
# HTML
if texttype == _HTML_TEXT_TYPE or texttype == _TEXT_TYPE:
encinfo.meta_media_type, encinfo.meta_encoding = getMetaInfo(text, log)
# guess
# 1. HTTP charset?
encinfo.encoding = encinfo.http_encoding
encinfo.mismatch = False
# 2. media_type?
# XML application/...
if texttype == _XML_APPLICATION_TYPE:
if not encinfo.encoding:
encinfo.encoding = encinfo.xml_encoding
# xml_encoding has default of utf-8
# text/html
elif texttype == _HTML_TEXT_TYPE:
if not encinfo.encoding:
encinfo.encoding = encinfo.meta_encoding
if not encinfo.encoding:
encinfo.encoding = encodingByMediaType(encinfo.http_media_type)
if not encinfo.encoding:
encinfo.encoding = tryEncodings(text)
# text/... + xml or text/*
elif texttype == _XML_TEXT_TYPE or texttype == _TEXT_TYPE:
if not encinfo.encoding:
encinfo.encoding = encodingByMediaType(encinfo.http_media_type)
elif texttype == _TEXT_UTF8:
if not encinfo.encoding:
encinfo.encoding = encodingByMediaType(encinfo.http_media_type)
# possible mismatches, checks if present at all and then if equal
# HTTP + XML
if encinfo.http_encoding and encinfo.xml_encoding and\
encinfo.http_encoding != encinfo.xml_encoding:
encinfo.mismatch = True
log.warning('"%s" (HTTP) != "%s" (XML) encoding mismatch' %
(encinfo.http_encoding, encinfo.xml_encoding))
# HTTP + Meta
if encinfo.http_encoding and encinfo.meta_encoding and\
encinfo.http_encoding != encinfo.meta_encoding:
encinfo.mismatch = True
log.warning('"%s" (HTTP) != "%s" (HTML ) encoding mismatch' %
(encinfo.http_encoding, encinfo.meta_encoding))
# XML + Meta
if encinfo.xml_encoding and encinfo.meta_encoding and\
encinfo.xml_encoding != encinfo.meta_encoding:
encinfo.mismatch = True
log.warning('"%s" (XML) != "%s" (HTML ) encoding mismatch' %
(encinfo.xml_encoding, encinfo.meta_encoding))
log.info('Encoding (probably): %s (Mismatch: %s)', encinfo.encoding,
encinfo.mismatch)
encinfo.logtext = logstream.getvalue()
return encinfo
if __name__ == '__main__':
import pydoc
pydoc.help(__name__)
css-parser-1.0.4/src/css_parser/errorhandler.py 0000644 0001750 0001750 00000010172 13407702010 022037 0 ustar kovid kovid 0000000 0000000 #!/usr/bin/env python
from __future__ import unicode_literals, division, absolute_import, print_function
import xml.dom
import logging
"""css_parser ErrorHandler
ErrorHandler
used as log with usual levels (debug, info, warn, error)
if instanciated with ``raiseExceptions=True`` raises exeptions instead
of logging
log
defaults to instance of ErrorHandler for any kind of log message from
lexerm, parser etc.
- raiseExceptions = [False, True]
- setloglevel(loglevel)
"""
__all__ = ['ErrorHandler']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
import sys
if sys.version_info[0] >= 3:
from urllib.error import HTTPError as urllib_HTTPError
from urllib.error import URLError as urllib_URLError
else:
from urllib2 import HTTPError as urllib_HTTPError
from urllib2 import URLError as urllib_URLError
class _ErrorHandler(object):
"""
handles all errors and log messages
"""
def __init__(self, log, defaultloglevel=logging.INFO,
raiseExceptions=True):
"""
inits log if none given
log
for parse messages, default logs to sys.stderr
defaultloglevel
if none give this is logging.DEBUG
raiseExceptions
- True: Errors will be raised e.g. during building
- False: Errors will be written to the log, this is the
default behaviour when parsing
"""
# may be disabled during setting of known valid items
self.enabled = True
if log:
self._log = log
else:
import sys
self._log = logging.getLogger('CSSUTILS')
hdlr = logging.StreamHandler(sys.stderr)
formatter = logging.Formatter('%(levelname)s\t%(message)s')
hdlr.setFormatter(formatter)
self._log.addHandler(hdlr)
self._log.setLevel(defaultloglevel)
self.raiseExceptions = raiseExceptions
def __getattr__(self, name):
"use self._log items"
calls = ('debug', 'info', 'warn', 'error', 'critical', 'fatal')
other = ('setLevel', 'getEffectiveLevel', 'addHandler', 'removeHandler')
if name in calls:
if name == 'warn':
name = 'warning'
self._logcall = getattr(self._log, name)
return self.__handle
elif name in other:
return getattr(self._log, name)
else:
raise AttributeError(
'(errorhandler) No Attribute %r found' % name)
def __handle(self, msg='', token=None, error=xml.dom.SyntaxErr,
neverraise=False, args=None):
"""
handles all calls
logs or raises exception
"""
if self.enabled:
if error is None:
error = xml.dom.SyntaxErr
line, col = None, None
if token:
if isinstance(token, tuple):
value, line, col = token[1], token[2], token[3]
else:
value, line, col = token.value, token.line, token.col
msg = '%s [%s:%s: %s]' % (
msg, line, col, value)
if error and self.raiseExceptions and not neverraise:
if isinstance(error, urllib_HTTPError) or isinstance(error, urllib_URLError):
raise
elif issubclass(error, xml.dom.DOMException):
error.line = line
error.col = col
raise error(msg)
else:
self._logcall(msg)
def setLog(self, log):
"""set log of errorhandler's log"""
self._log = log
class ErrorHandler(_ErrorHandler):
"Singleton, see _ErrorHandler"
instance = None
def __init__(self,
log=None, defaultloglevel=logging.INFO, raiseExceptions=True):
if ErrorHandler.instance is None:
ErrorHandler.instance = _ErrorHandler(log=log,
defaultloglevel=defaultloglevel,
raiseExceptions=raiseExceptions)
self.__dict__ = ErrorHandler.instance.__dict__
css-parser-1.0.4/src/css_parser/helper.py 0000644 0001750 0001750 00000007662 13407702010 020641 0 ustar kovid kovid 0000000 0000000 """css_parser helper TEST
"""
from __future__ import unicode_literals, division, absolute_import, print_function
__docformat__ = 'restructuredtext'
__version__ = '$Id: errorhandler.py 1234 2008-05-22 20:26:12Z cthedot $'
import os
import re
import sys
if sys.version_info[0] >= 3:
from urllib.request import pathname2url as urllib_pathname2url
else:
from urllib import pathname2url as urllib_pathname2url
class Deprecated(object):
"""This is a decorator which can be used to mark functions
as deprecated. It will result in a warning being emitted
when the function is used.
It accepts a single paramter ``msg`` which is shown with the warning.
It should contain information which function or method to use instead.
"""
def __init__(self, msg):
self.msg = msg
def __call__(self, func):
def newFunc(*args, **kwargs):
import warnings
warnings.warn("Call to deprecated method %r. %s" %
(func.__name__, self.msg),
category=DeprecationWarning,
stacklevel=2)
return func(*args, **kwargs)
newFunc.__name__ = func.__name__
newFunc.__doc__ = func.__doc__
newFunc.__dict__.update(func.__dict__)
return newFunc
# simple escapes, all non unicodes
_simpleescapes = re.compile(r'(\\[^0-9a-fA-F])').sub
def normalize(x):
r"""
normalizes x, namely:
- remove any \ before non unicode sequences (0-9a-zA-Z) so for
x=="c\olor\" return "color" (unicode escape sequences should have
been resolved by the tokenizer already)
- lowercase
"""
if x:
def removeescape(matchobj):
return matchobj.group(0)[1:]
x = _simpleescapes(removeescape, x)
return x.lower()
else:
return x
def path2url(path):
"""Return file URL of `path`"""
return 'file:' + urllib_pathname2url(os.path.abspath(path))
def pushtoken(token, tokens):
"""Return new generator starting with token followed by all tokens in
``tokens``"""
# TODO: may use itertools.chain?
yield token
for t in tokens:
yield t
def string(value):
"""
Serialize value with quotes e.g.::
``a \'string`` => ``'a \'string'``
"""
# \n = 0xa, \r = 0xd, \f = 0xc
value = value.replace('\n', '\\a ').replace(
'\r', '\\d ').replace(
'\f', '\\c ').replace(
'"', '\\"')
if value.endswith('\\'):
value = value[:-1] + '\\\\'
return '"%s"' % value
def stringvalue(string):
"""
Retrieve actual value of string without quotes. Escaped
quotes inside the value are resolved, e.g.::
``'a \'string'`` => ``a 'string``
"""
return string.replace('\\'+string[0], string[0])[1:-1]
_match_forbidden_in_uri = re.compile(r'''.*?[\(\)\s\;,'"]''', re.U).match
def uri(value):
"""
Serialize value by adding ``url()`` and with quotes if needed e.g.::
``"`` => ``url("\"")``
"""
if _match_forbidden_in_uri(value):
value = string(value)
return 'url(%s)' % value
def urivalue(uri):
"""
Return actual content without surrounding "url(" and ")"
and removed surrounding quotes too including contained
escapes of quotes, e.g.::
``url("\"")`` => ``"``
"""
uri = uri[uri.find('(')+1:-1].strip()
if uri and (uri[0] in '\'"') and (uri[0] == uri[-1]):
return stringvalue(uri)
else:
return uri
# def normalnumber(num):
# """
# Return normalized number as string.
# """
# sign = ''
# if num.startswith('-'):
# sign = '-'
# num = num[1:]
# elif num.startswith('+'):
# num = num[1:]
#
# if float(num) == 0.0:
# return '0'
# else:
# if num.find('.') == -1:
# return sign + str(int(num))
# else:
# a, b = num.split('.')
# if not a:
# a = '0'
# return '%s%s.%s' % (sign, int(a), b)
css-parser-1.0.4/src/css_parser/parse.py 0000644 0001750 0001750 00000021143 13407703013 020466 0 ustar kovid kovid 0000000 0000000 #!/usr/bin/env python
"""A validating CSSParser"""
from __future__ import unicode_literals, division, absolute_import, print_function
__all__ = ['CSSParser']
__docformat__ = 'restructuredtext'
__version__ = '$Id$'
from .helper import path2url
import codecs
import css_parser
import sys
from . import tokenize2
from css_parser import css
if sys.version_info < (2, 6):
bytes = str
class CSSParser(object):
"""Parse a CSS StyleSheet from URL, string or file and return a DOM Level 2
CSS StyleSheet object.
Usage::
parser = CSSParser()
# optionally
parser.setFetcher(fetcher)
sheet = parser.parseFile('test1.css', 'ascii')
print sheet.cssText
"""
def __init__(self, log=None, loglevel=None, raiseExceptions=None,
fetcher=None, parseComments=True,
validate=True):
"""
:param log:
logging object
:param loglevel:
logging loglevel
:param raiseExceptions:
if log should simply log (default) or raise errors during
parsing. Later while working with the resulting sheets
the setting used in css_parser.log.raiseExeptions is used
:param fetcher:
see ``setFetcher(fetcher)``
:param parseComments:
if comments should be added to CSS DOM or simply omitted
:param validate:
if parsing should validate, may be overwritten in parse methods
"""
if log is not None:
css_parser.log.setLog(log)
if loglevel is not None:
css_parser.log.setLevel(loglevel)
# remember global setting
self.__globalRaising = css_parser.log.raiseExceptions
if raiseExceptions:
self.__parseRaising = raiseExceptions
else:
# DEFAULT during parse
self.__parseRaising = False
self.__tokenizer = tokenize2.Tokenizer(doComments=parseComments)
self.setFetcher(fetcher)
self._validate = validate
def __parseSetting(self, parse):
"""during parse exceptions may be handled differently depending on
init parameter ``raiseExceptions``
"""
if parse:
css_parser.log.raiseExceptions = self.__parseRaising
else:
css_parser.log.raiseExceptions = self.__globalRaising
def parseStyle(self, cssText, encoding='utf-8', validate=None):
"""Parse given `cssText` which is assumed to be the content of
a HTML style attribute.
:param cssText:
CSS string to parse
:param encoding:
It will be used to decode `cssText` if given as a (byte)
string.
:param validate:
If given defines if validation is used. Uses CSSParser settings as
fallback
:returns:
:class:`~css_parser.css.CSSStyleDeclaration`
"""
self.__parseSetting(True)
if isinstance(cssText, bytes):
# TODO: use codecs.getdecoder('css') here?
cssText = cssText.decode(encoding)
if validate is None:
validate = self._validate
style = css.CSSStyleDeclaration(cssText, validating=validate)
self.__parseSetting(False)
return style
def parseString(self, cssText, encoding=None, href=None, media=None,
title=None,
validate=None):
"""Parse `cssText` as :class:`~css_parser.css.CSSStyleSheet`.
Errors may be raised (e.g. UnicodeDecodeError).
:param cssText:
CSS string to parse
:param encoding:
If ``None`` the encoding will be read from BOM or an @charset
rule or defaults to UTF-8.
If given overrides any found encoding including the ones for
imported sheets.
It also will be used to decode `cssText` if given as a (byte)
string.
:param href:
The ``href`` attribute to assign to the parsed style sheet.
Used to resolve other urls in the parsed sheet like @import hrefs.
:param media:
The ``media`` attribute to assign to the parsed style sheet
(may be a MediaList, list or a string).
:param title:
The ``title`` attribute to assign to the parsed style sheet.
:param validate:
If given defines if validation is used. Uses CSSParser settings as
fallback
:returns:
:class:`~css_parser.css.CSSStyleSheet`.
"""
self.__parseSetting(True)
# TODO: py3 needs bytes here!
if isinstance(cssText, bytes):
cssText = codecs.getdecoder('css')(cssText, encoding=encoding)[0]
if validate is None:
validate = self._validate
sheet = css_parser.css.CSSStyleSheet(
href=href,
media=css_parser.stylesheets.MediaList(media),
title=title,
validating=validate)
sheet._setFetcher(self.__fetcher)
# tokenizing this ways closes open constructs and adds EOF
sheet._setCssTextWithEncodingOverride(self.__tokenizer.tokenize(cssText,
fullsheet=True),
encodingOverride=encoding)
self.__parseSetting(False)
return sheet
def parseFile(self, filename, encoding=None,
href=None, media=None, title=None,
validate=None):
"""Retrieve content from `filename` and parse it. Errors may be raised
(e.g. IOError).
:param filename:
of the CSS file to parse, if no `href` is given filename is
converted to a (file:) URL and set as ``href`` of resulting
stylesheet.
If `href` is given it is set as ``sheet.href``. Either way
``sheet.href`` is used to resolve e.g. stylesheet imports via
@import rules.
:param encoding:
Value ``None`` defaults to encoding detection via BOM or an
@charset rule.
Other values override detected encoding for the sheet at
`filename` including any imported sheets.
:returns:
:class:`~css_parser.css.CSSStyleSheet`.
"""
if not href:
href = path2url(filename)
f = open(filename, 'rb')
css = f.read()
f.close()
return self.parseString(css,
encoding=encoding, # read returns a str
href=href, media=media, title=title,
validate=validate)
def parseUrl(self, href, encoding=None, media=None, title=None,
validate=None):
"""Retrieve content from URL `href` and parse it. Errors may be raised
(e.g. URLError).
:param href:
URL of the CSS file to parse, will also be set as ``href`` of
resulting stylesheet
:param encoding:
Value ``None`` defaults to encoding detection via HTTP, BOM or an
@charset rule.
A value overrides detected encoding for the sheet at ``href``
including any imported sheets.
:returns:
:class:`~css_parser.css.CSSStyleSheet`.
"""
encoding, enctype, text = css_parser.util._readUrl(
href,
fetcher=self.__fetcher,
overrideEncoding=encoding)
if enctype == 5:
# do not use if defaulting to UTF-8
encoding = None
if text is not None:
return self.parseString(text, encoding=encoding,
href=href, media=media, title=title,
validate=validate)
def setFetcher(self, fetcher=None):
"""Replace the default URL fetch function with a custom one.
:param fetcher:
A function which gets a single parameter
``url``
the URL to read
and must return ``(encoding, content)`` where ``encoding`` is the
HTTP charset normally given via the Content-Type header (which may
simply omit the charset in which case ``encoding`` would be
``None``) and ``content`` being the string (or unicode) content.
The Mimetype should be 'text/css' but this has to be checked by the
fetcher itself (the default fetcher emits a warning if encountering
a different mimetype).
Calling ``setFetcher`` with ``fetcher=None`` resets css_parser
to use its default function.
"""
self.__fetcher = fetcher
css-parser-1.0.4/src/css_parser/prodparser.py 0000644 0001750 0001750 00000067254 13407702706 021562 0 ustar kovid kovid 0000000 0000000 # -*- coding: utf-8 -*-
"""Productions parser used by css and stylesheets classes to parse
test into a css_parser.util.Seq and at the same time retrieving
additional specific css_parser.util.Item objects for later use.
TODO:
- ProdsParser
- handle EOF or STOP?
- handle unknown @rules
- handle S: maybe save to Seq? parameterized?
- store['_raw']: always?
- Sequence:
- opt first(), naive impl for now
"""
from __future__ import unicode_literals, division, absolute_import, print_function
__all__ = ['ProdParser', 'Sequence', 'Choice', 'Prod', 'PreDef']
__docformat__ = 'restructuredtext'
__version__ = '$Id: parse.py 1418 2008-08-09 19:27:50Z cthedot $'
from .helper import pushtoken
import css_parser
import itertools
import re
import sys
import types
if sys.version_info[0] >= 3:
text_type = str
string_type = str
else:
text_type = unicode
string_type = basestring
class ParseError(Exception):
"""Base Exception class for ProdParser (used internally)."""
pass
class Done(ParseError):
"""Raised if Sequence or Choice is finished and no more Prods left."""
pass
class Exhausted(ParseError):
"""Raised if Sequence or Choice is finished but token is given."""
pass
class Missing(ParseError):
"""Raised if Sequence or Choice is not finished but no matching token given."""
pass
class NoMatch(ParseError):
"""Raised if nothing in Sequence or Choice does match."""
pass
class Choice(object):
"""A Choice of productions (Sequence or single Prod)."""
def __init__(self, *prods, **options):
"""
*prods
Prod or Sequence objects
options:
optional=False
"""
self._prods = prods
try:
self.optional = options['optional']
except KeyError:
for p in self._prods:
if p.optional:
self.optional = True
break
else:
self.optional = False
self.reset()
def reset(self):
"""Start Choice from zero"""
self._exhausted = False
def matches(self, token):
"""Check if token matches"""
for prod in self._prods:
if prod.matches(token):
return True
return False
def nextProd(self, token):
"""
Return:
- next matching Prod or Sequence
- ``None`` if any Prod or Sequence is optional and no token matched
- raise ParseError if nothing matches and all are mandatory
- raise Exhausted if choice already done
``token`` may be None but this occurs when no tokens left."""
# print u'TEST for %s in %s' % (token, self)
if not self._exhausted:
optional = False
for p in self._prods:
if p.matches(token):
self._exhausted = True
p.reset()
# print u'FOUND for %s: %s' % (token, p);#print
return p
elif p.optional:
optional = True
else:
if not optional:
# None matched but also None is optional
raise NoMatch('No match for %s in %s' % (token, self))
# raise ParseError(u'No match in %s for %s' % (self, token))
elif token:
raise Exhausted('Extra token')
def __repr__(self):
return "" % (
self.__class__.__name__, self.__str__(), self.optional, id(self))
def __str__(self):
return 'Choice(%s)' % ', '.join([str(x) for x in self._prods])
class Sequence(object):
"""A Sequence of productions (Choice or single Prod)."""
def __init__(self, *prods, **options):
"""
*prods
Prod or Choice or Sequence objects
**options:
minmax = lambda: (1, 1)
callback returning number of times this sequence may run
"""
self._prods = prods
try:
minmax = options['minmax']
except KeyError:
def minmax(): return (1, 1)
self._min, self._max = minmax()
if self._max is None:
# unlimited
try:
# py2.6/3
self._max = sys.maxsize
except AttributeError:
# py<2.6
self._max = sys.maxint
self._prodcount = len(self._prods)
self.reset()
def matches(self, token):
"""Called by Choice to try to find if Sequence matches."""
for prod in self._prods:
if prod.matches(token):
return True
try:
if not prod.optional:
break
except AttributeError:
pass
return False
def reset(self):
"""Reset this Sequence if it is nested."""
self._roundstarted = False
self._i = 0
self._round = 0
def _currentName(self):
"""Return current element of Sequence, used by name"""
# TODO: current impl first only if 1st if an prod!
for prod in self._prods[self._i:]:
if not prod.optional:
return str(prod)
else:
return 'Sequence'
optional = property(lambda self: self._min == 0)
def nextProd(self, token):
"""Return
- next matching Prod or Choice
- raises ParseError if nothing matches
- raises Exhausted if sequence already done
"""
# print u'TEST for %s in %s' % (token, self)
while self._round < self._max:
# for this round
i = self._i
round = self._round
p = self._prods[i]
if i == 0:
self._roundstarted = False
# for next round
self._i += 1
if self._i == self._prodcount:
self._round += 1
self._i = 0
if p.matches(token):
self._roundstarted = True
# reset nested Choice or Prod to use from start
p.reset()
# print u'FOUND for %s: %s' % (token, p);#print
return p
elif p.optional:
continue
elif round < self._min or self._roundstarted: # or (round == 0 and self._min == 0):
raise Missing('Missing token for production %s' % p)
elif not token:
if self._roundstarted:
raise Missing('Missing token for production %s' % p)
else:
raise Done()
else:
raise NoMatch('No match for %s in %s' % (token, self))
if token:
raise Exhausted('Extra token')
def __repr__(self):
return "" % (
self.__class__.__name__, self.__str__(), self.optional, id(self))
def __str__(self):
return 'Sequence(%s)' % ', '.join([str(x) for x in self._prods])
class Prod(object):
"""Single Prod in Sequence or Choice."""
def __init__(self, name, match, optional=False,
toSeq=None, toStore=None,
stop=False, stopAndKeep=False,
stopIfNoMoreMatch=False,
nextSor=False, mayEnd=False,
storeToken=None,
exception=None):
"""
name
name used for error reporting
match callback
function called with parameters tokentype and tokenvalue
returning True, False or raising ParseError
toSeq callback (optional) or False
calling toSeq(token, tokens) returns (type_, val) == (token[0], token[1])
to be appended to seq else simply unaltered (type_, val)
if False nothing is added
toStore (optional)
key to save util.Item to store or callback(store, util.Item)
optional = False
whether Prod is optional or not
stop = False
if True stop parsing of tokens here
stopAndKeep
if True stop parsing of tokens here but return stopping
token in unused tokens
stopIfNoMoreMatch = False
stop even if more tokens available, similar to stop and keep but with
condition no more matches
nextSor=False
next is S or other like , or / (CSSValue)
mayEnd = False
no token must follow even defined by Sequence.
Used for operator ',/ ' currently only
storeToken = None
if True toStore saves simple token tuple and not and Item object
to store. Old style processing, TODO: resolve
exception = None
exception to be raised in case of error, normaly SyntaxErr
"""
self._name = name
self.match = match
self.optional = optional
self.stop = stop
self.stopAndKeep = stopAndKeep
self.stopIfNoMoreMatch = stopIfNoMoreMatch
self.nextSor = nextSor
self.mayEnd = mayEnd
self.storeToken = storeToken
self.exception = exception
def makeToStore(key):
"Return a function used by toStore."
def toStore(store, item):
"Set or append store item."
if key in store:
_v = store[key]
if not isinstance(_v, list):
store[key] = [_v]
store[key].append(item)
else:
store[key] = item
return toStore
if toSeq or toSeq is False:
# called: seq.append(toSeq(value))
self.toSeq = toSeq
else:
self.toSeq = lambda t, tokens: (t[0], t[1])
if hasattr(toStore, '__call__'):
self.toStore = toStore
elif toStore:
self.toStore = makeToStore(toStore)
else:
# always set!
self.toStore = None
def matches(self, token):
"""Return if token matches."""
if not token:
return False
type_, val, line, col = token
return self.match(type_, val)
def reset(self):
pass
def __str__(self):
return self._name
def __repr__(self):
return "" % (
self.__class__.__name__, self._name, id(self))
# global tokenizer as there is only one!
tokenizer = css_parser.tokenize2.Tokenizer()
# global: saved from subProds
savedTokens = []
class ProdParser(object):
"""Productions parser."""
def __init__(self, clear=True):
self.types = css_parser.cssproductions.CSSProductions
self._log = css_parser.log
if clear:
tokenizer.clear()
def _texttotokens(self, text):
"""Build a generator which is the only thing that is parsed!
old classes may use lists etc
"""
# under python 2.x this was basestring, but ...
if isinstance(text, string_type):
# DEFAULT, to tokenize strip space
return tokenizer.tokenize(text.strip())
elif type(text) is types.GeneratorType: # noqa
# DEFAULT, already tokenized, should be generator
return text
elif isinstance(text, tuple):
# single token
return iter([text])
elif isinstance(text, list):
# OLD: generator from list
return iter(text)
else:
# ?
return text
def _SorTokens(self, tokens, until=',/'):
"""New tokens generator which has S tokens removed,
if followed by anything in ``until``, normally a ``,``."""
for token in tokens:
if token[0] == self.types.S:
try:
next_ = next(tokens)
except StopIteration:
yield token
else:
if next_[1] in until:
# omit S as e.g. ``,`` has been found
yield next_
elif next_[0] == self.types.COMMENT:
# pass COMMENT
yield next_
else:
yield token
yield next_
elif token[0] == self.types.COMMENT:
# pass COMMENT
yield token
else:
yield token
break
# normal mode again
for token in tokens:
yield token
def parse(self, text, name, productions, keepS=False, checkS=False, store=None,
emptyOk=False, debug=False):
"""
text (or token generator)
to parse, will be tokenized if not a generator yet
may be:
- a string to be tokenized
- a single token, a tuple
- a tuple of (token, tokensGenerator)
- already tokenized so a tokens generator
name
used for logging
productions
used to parse tokens
keepS
if WS should be added to Seq or just be ignored
store UPDATED
If a Prod defines ``toStore`` the key defined there
is a key in store to be set or if store[key] is a list
the next Item is appended here.
TODO: NEEDED? :
Key ``raw`` is always added and holds all unprocessed
values found
emptyOk
if True text may be empty, hard to test before as may be generator
returns
:wellformed: True or False
:seq: a filled css_parser.util.Seq object which is NOT readonly yet
:store: filled keys defined by Prod.toStore
:unusedtokens: token generator containing tokens not used yet
"""
tokens = self._texttotokens(text)
if not tokens:
self._log.error('No content to parse.')
return False, [], None, None
seq = css_parser.util.Seq(readonly=False)
if not store: # store for specific values
store = {}
prods = [productions] # stack of productions
wellformed = True
# while no real token is found any S are ignored
started = False
stopall = False
prod = None
# flag if default S handling should be done
defaultS = True
stopIfNoMoreMatch = False
while True:
# get from savedTokens or normal tokens
try:
# print debug, "SAVED", savedTokens
token = savedTokens.pop()
except IndexError:
try:
token = next(tokens)
except StopIteration:
break
# print debug, token, stopIfNoMoreMatch
type_, val, line, col = token
# default productions
if type_ == self.types.COMMENT:
# always append COMMENT
seq.append(css_parser.css.CSSComment(val),
css_parser.css.CSSComment, line, col)
elif defaultS and type_ == self.types.S and not checkS:
# append S (but ignore starting ones)
if not keepS or not started:
continue
else:
seq.append(val, type_, line, col)
# elif type_ == self.types.ATKEYWORD:
# # @rule
# r = css_parser.css.CSSUnknownRule(cssText=val)
# seq.append(r, type(r), line, col)
elif type_ == self.types.INVALID:
# invalidate parse
wellformed = False
self._log.error('Invalid token: %r' % (token,))
break
elif type_ == 'EOF':
# do nothing? (self.types.EOF == True!)
stopall = True
else:
started = True # check S now
try:
while True:
# find next matching production
try:
prod = prods[-1].nextProd(token)
except (Exhausted, NoMatch):
# try next
prod = None
if isinstance(prod, Prod):
# found actual Prod, not a Choice or Sequence
break
elif prod:
# nested Sequence, Choice
prods.append(prod)
else:
# nested exhausted, try in parent
if len(prods) > 1:
prods.pop()
else:
raise NoMatch('No match')
except NoMatch as e:
if stopIfNoMoreMatch: # and token:
# print "\t1stopIfNoMoreMatch", e, token, prod, 'PUSHING'
# tokenizer.push(token)
savedTokens.append(token)
stopall = True
else:
wellformed = False
self._log.error('%s: %s: %r' % (name, e, token))
break
except ParseError as e:
# needed???
if stopIfNoMoreMatch: # and token:
# print "\t2stopIfNoMoreMatch", e, token, prod
tokenizer.push(token)
stopall = True
else:
wellformed = False
self._log.error('%s: %s: %r' % (name, e, token))
break
else:
# print '\t1', debug, 'PROD', prod
# may stop next time, once set stays
stopIfNoMoreMatch = prod.stopIfNoMoreMatch or stopIfNoMoreMatch
# process prod
if prod.toSeq and not prod.stopAndKeep:
type_, val = prod.toSeq(token, tokens)
if val is not None:
seq.append(val, type_, line, col)
if prod.toStore:
if not prod.storeToken:
prod.toStore(store, seq[-1])
else:
# workaround for now for old style token
# parsing!
# TODO: remove when all new style
prod.toStore(store, token)
if prod.stop:
# stop here and ignore following tokens
# EOF? or end of e.g. func ")"
break
if prod.stopAndKeep: # e.g. ;
# stop here and ignore following tokens
# but keep this token for next run
# TODO: CHECK!!!!
tokenizer.push(token)
tokens = itertools.chain(token, tokens)
stopall = True
break
if prod.nextSor:
# following is S or other token (e.g. ",")?
# remove S if
tokens = self._SorTokens(tokens, ',/')
defaultS = False
else:
defaultS = True
lastprod = prod
# print debug, 'parse done', token, stopall, '\n'
if not stopall:
# stop immediately
while True:
# all productions exhausted?
try:
prod = prods[-1].nextProd(token=None)
except Done:
# ok
prod = None
except Missing as e:
prod = None
# last was a S operator which may End a Sequence, then ok
if hasattr(lastprod, 'mayEnd') and not lastprod.mayEnd:
wellformed = False
self._log.error('%s: %s' % (name, e))
except ParseError as e:
prod = None
wellformed = False
self._log.error('%s: %s' % (name, e))
else:
if prods[-1].optional:
prod = None
elif prod and prod.optional:
# ignore optional
continue
if prod and not prod.optional:
wellformed = False
self._log.error('%s: Missing token for production %r'
% (name, text_type(prod)))
break
elif len(prods) > 1:
# nested exhausted, next in parent
prods.pop()
else:
break
if not emptyOk and not len(seq):
self._log.error('No content to parse.')
return False, [], None, None
# trim S from end
seq.rstrip()
return wellformed, seq, store, tokens
class PreDef(object):
"""Predefined Prod definition for use in productions definition
for ProdParser instances.
"""
types = css_parser.cssproductions.CSSProductions
reHexcolor = re.compile(r'^\#(?:[0-9abcdefABCDEF]{3}|[0-9abcdefABCDEF]{6})$')
@staticmethod
def calc(toSeq=None, nextSor=False):
return Prod(name='calcfunction',
match=lambda t, v: 'calc(' == css_parser.helper.normalize(v),
toSeq=toSeq,
nextSor=nextSor)
@staticmethod
def char(name='char', char=',', toSeq=None,
stop=False, stopAndKeep=False, mayEnd=False,
stopIfNoMoreMatch=False,
optional=False, # WAS: optional=True,
nextSor=False):
"any CHAR"
return Prod(name=name, match=lambda t, v: v == char, toSeq=toSeq,
stop=stop, stopAndKeep=stopAndKeep, mayEnd=mayEnd,
stopIfNoMoreMatch=stopIfNoMoreMatch,
optional=optional,
nextSor=nextSor)
@staticmethod
def comma(optional=False, toSeq=None):
return PreDef.char('comma', ',', optional=optional, toSeq=toSeq)
@staticmethod
def comment(parent=None):
return Prod(name='comment',
match=lambda t, v: t == 'COMMENT',
toSeq=lambda t, tokens: (t[0], css_parser.css.CSSComment([1], parentRule=parent)),
optional=True
)
@staticmethod
def dimension(nextSor=False, stop=False):
return Prod(name='dimension',
match=lambda t, v: t == PreDef.types.DIMENSION,
toSeq=lambda t, tokens: (t[0], css_parser.helper.normalize(t[1])),
stop=stop,
nextSor=nextSor)
@staticmethod
def function(toSeq=None, nextSor=False, toStore=None):
return Prod(name='function',
match=lambda t, v: t == PreDef.types.FUNCTION,
toStore=toStore,
toSeq=toSeq,
nextSor=nextSor)
@staticmethod
def funcEnd(stop=False, mayEnd=False):
")"
return PreDef.char('end FUNC ")"', ')', stop=stop, mayEnd=mayEnd)
@staticmethod
def hexcolor(stop=False, nextSor=False):
"#123 or #123456"
return Prod(name='HEX color',
match=lambda t, v: (
t == PreDef.types.HASH and
PreDef.reHexcolor.match(v)
),
stop=stop,
nextSor=nextSor)
@staticmethod
def ident(stop=False, toStore=None, nextSor=False):
return Prod(name='ident',
match=lambda t, v: t == PreDef.types.IDENT,
stop=stop,
toStore=toStore,
nextSor=nextSor)
@staticmethod
def number(stop=False, toSeq=None, nextSor=False):
return Prod(name='number',
match=lambda t, v: t == PreDef.types.NUMBER,
stop=stop,
toSeq=toSeq,
nextSor=nextSor)
@staticmethod
def percentage(stop=False, toSeq=None, nextSor=False):
return Prod(name='percentage',
match=lambda t, v: t == PreDef.types.PERCENTAGE,
stop=stop,
toSeq=toSeq,
nextSor=nextSor)
@staticmethod
def string(stop=False, nextSor=False):
"string delimiters are removed by default"
return Prod(name='string',
match=lambda t, v: t == PreDef.types.STRING,
toSeq=lambda t, tokens: (t[0], css_parser.helper.stringvalue(t[1])),
stop=stop,
nextSor=nextSor)
@staticmethod
def S(name='whitespace', toSeq=None, optional=False):
return Prod(name=name,
match=lambda t, v: t == PreDef.types.S,
toSeq=toSeq,
optional=optional,
mayEnd=True)
@staticmethod
def unary(stop=False, toSeq=None, nextSor=False):
"+ or -"
return Prod(name='unary +-', match=lambda t, v: v in ('+', '-'),
optional=True,
stop=stop,
toSeq=toSeq,
nextSor=nextSor)
@staticmethod
def uri(stop=False, nextSor=False):
"'url(' and ')' are removed and URI is stripped"
return Prod(name='URI',
match=lambda t, v: t == PreDef.types.URI,
toSeq=lambda t, tokens: (t[0], css_parser.helper.urivalue(t[1])),
stop=stop,
nextSor=nextSor)
@staticmethod
def unicode_range(stop=False, nextSor=False):
"u+123456-abc normalized to lower `u`"
return Prod(name='unicode-range',
match=lambda t, v: t == PreDef.types.UNICODE_RANGE,
toSeq=lambda t, tokens: (t[0], t[1].lower()),
stop=stop,
nextSor=nextSor
)
@staticmethod
def ratio(stop=False, nextSor=False):
"positive integer / positive integer"
return Prod(name='ratio',
match=lambda t, v: t == PreDef.types.RATIO,
toSeq=lambda t, tokens: (t[0], t[1].lower()),
stop=stop,
nextSor=nextSor
)
@staticmethod
def variable(toSeq=None, stop=False, nextSor=False, toStore=None):
return Prod(name='variable',
match=lambda t, v: 'var(' == css_parser.helper.normalize(v),
toSeq=toSeq,
toStore=toStore,
stop=stop,
nextSor=nextSor)
# used for MarginRule for now:
@staticmethod
def unknownrule(name='@', toStore=None):
"""@rule dummy (matches ATKEYWORD to remove unknown rule tokens from
stream::
@x;
@x {...}
no nested yet!
"""
def rule(tokens):
saved = []
for t in tokens:
saved.append(t)
if (t[1] == '}' or t[1] == ';'):
return css_parser.css.CSSUnknownRule(saved)
return Prod(name=name,
match=lambda t, v: t == 'ATKEYWORD',
toSeq=lambda t, tokens: ('CSSUnknownRule',
rule(pushtoken(t, tokens))
),
toStore=toStore
)
css-parser-1.0.4/src/css_parser/profiles.py 0000644 0001750 0001750 00000104613 13407702010 021177 0 ustar kovid kovid 0000000 0000000 from __future__ import unicode_literals, division, absolute_import, print_function
from css_parser import util
import sys
import re
"""CSS profiles.
Profiles is based on code by Kevin D. Smith, orginally used as cssvalues,
thanks!
"""
__all__ = ['Profiles']
__docformat__ = 'restructuredtext'
__version__ = '$Id: cssproperties.py 1116 2008-03-05 13:52:23Z cthedot $'
if sys.version_info[0] >= 3:
string_type = str
else:
string_type = basestring
def as_list(p):
if isinstance(p, list):
return p
return list(p)
class NoSuchProfileException(Exception):
"""Raised if no profile with given name is found"""
pass
class Profiles(object):
"""
All profiles used for validation. ``css_parser.profile`` is a
preset object of this class and used by all properties for validation.
Predefined profiles are (use
:meth:`~css_parser.profiles.Profiles.propertiesByProfile` to
get a list of defined properties):
:attr:`~css_parser.profiles.Profiles.CSS_LEVEL_2`
Properties defined by CSS2.1
:attr:`~css_parser.profiles.Profiles.CSS3_BASIC_USER_INTERFACE`
Currently resize and outline properties only
:attr:`~css_parser.profiles.Profiles.CSS3_BOX`
Currently overflow related properties only
:attr:`~css_parser.profiles.Profiles.CSS3_COLOR`
CSS 3 color properties
:attr:`~css_parser.profiles.Profiles.CSS3_PAGED_MEDIA`
As defined at http://www.w3.org/TR/css3-page/ (at 090307)
Predefined macros are:
:attr:`~css_parser.profiles.Profiles._TOKEN_MACROS`
Macros containing the token values as defined to CSS2
:attr:`~css_parser.profiles.Profiles._MACROS`
Additional general macros.
If you want to redefine any of these macros do this in your custom
macros.
"""
CSS_LEVEL_2 = 'CSS Level 2.1'
CSS3_BACKGROUNDS_AND_BORDERS = 'CSS Backgrounds and Borders Module Level 3'
CSS3_BASIC_USER_INTERFACE = 'CSS3 Basic User Interface Module'
CSS3_BOX = CSS_BOX_LEVEL_3 = 'CSS Box Module Level 3'
CSS3_COLOR = CSS_COLOR_LEVEL_3 = 'CSS Color Module Level 3'
CSS3_FONTS = 'CSS Fonts Module Level 3'
CSS3_FONT_FACE = 'CSS Fonts Module Level 3 @font-face properties'
CSS3_PAGED_MEDIA = 'CSS3 Paged Media Module'
CSS3_TEXT = 'CSS Text Level 3'
_TOKEN_MACROS = {
'ident': r'[-]?{nmstart}{nmchar}*',
'name': r'{nmchar}+',
'nmstart': r'[_a-z]|{nonascii}|{escape}',
'nonascii': r'[^\0-\177]',
'unicode': r'\\[0-9a-f]{1,6}(\r\n|[ \n\r\t\f])?',
'escape': r'{unicode}|\\[ -~\u0080-\u01ff]',
# 'escape': r'{unicode}|\\[ -~\200-\4177777]',
'int': r'[-]?\d+',
'nmchar': r'[\w-]|{nonascii}|{escape}',
'num': r'[-]?\d+|[-]?\d*\.\d+',
'positivenum': r'\d+|\d*\.\d+',
'number': r'{num}',
'string': r'{string1}|{string2}',
'string1': r'"(\\\"|[^\"])*"',
'uri': r'url\({w}({string}|(\\\)|[^\)])+){w}\)',
'string2': r"'(\\\'|[^\'])*'",
'nl': r'\n|\r\n|\r|\f',
'w': r'\s*',
}
_MACROS = {
'hexcolor': r'#[0-9a-f]{3}|#[0-9a-f]{6}',
'rgbcolor': r'rgb\({w}{int}{w}\,{w}{int}{w}\,{w}{int}{w}\)|rgb\({w}{num}%{w}\,{w}{num}%{w}\,{w}{num}%{w}\)',
'namedcolor': r'(transparent|orange|maroon|red|orange|yellow|olive|purple|fuchsia|white|lime|green|navy|blue|aqua|teal|black|silver|gray)', # noqa
'uicolor': r'(ActiveBorder|ActiveCaption|AppWorkspace|Background|ButtonFace|ButtonHighlight|ButtonShadow|ButtonText|CaptionText|GrayText|Highlight|HighlightText|InactiveBorder|InactiveCaption|InactiveCaptionText|InfoBackground|InfoText|Menu|MenuText|Scrollbar|ThreeDDarkShadow|ThreeDFace|ThreeDHighlight|ThreeDLightShadow|ThreeDShadow|Window|WindowFrame|WindowText)', # noqa
'color': r'{namedcolor}|{hexcolor}|{rgbcolor}|{uicolor}',
'integer': r'{int}',
'length': r'0|{num}(em|ex|px|in|cm|mm|pt|pc)',
'positivelength': r'0|{positivenum}(em|ex|px|in|cm|mm|pt|pc)',
'angle': r'0|{num}(deg|grad|rad)',
'time': r'0|{num}m?s',
'frequency': r'0|{num}k?Hz',
'percentage': r'{num}%',
'shadow': '(inset)?{w}{length}{w}{length}{w}{length}?{w}{length}?{w}{color}?'
}
def __init__(self, log=None):
"""A few profiles are predefined."""
self._log = log
# macro cache
self._usedMacros = Profiles._TOKEN_MACROS.copy()
self._usedMacros.update(Profiles._MACROS.copy())
# to keep order, REFACTOR!
self._profileNames = []
# for reset if macro changes
self._rawProfiles = {}
# already compiled profiles: {profile: {property: checkfunc, ...}, ...}
self._profilesProperties = {}
self._defaultProfiles = None
self.addProfiles([(self.CSS_LEVEL_2,
properties[self.CSS_LEVEL_2],
macros[self.CSS_LEVEL_2]
),
(self.CSS3_BACKGROUNDS_AND_BORDERS,
properties[self.CSS3_BACKGROUNDS_AND_BORDERS],
macros[self.CSS3_BACKGROUNDS_AND_BORDERS]
),
(self.CSS3_BASIC_USER_INTERFACE,
properties[self.CSS3_BASIC_USER_INTERFACE],
macros[self.CSS3_BASIC_USER_INTERFACE]
),
(self.CSS3_BOX,
properties[self.CSS3_BOX],
macros[self.CSS3_BOX]
),
(self.CSS3_COLOR,
properties[self.CSS3_COLOR],
macros[self.CSS3_COLOR]
),
(self.CSS3_FONTS,
properties[self.CSS3_FONTS],
macros[self.CSS3_FONTS]
),
# new object for font-face only?
(self.CSS3_FONT_FACE,
properties[self.CSS3_FONT_FACE],
macros[self.CSS3_FONTS]
),
(self.CSS3_PAGED_MEDIA,
properties[self.CSS3_PAGED_MEDIA],
macros[self.CSS3_PAGED_MEDIA]
),
(self.CSS3_TEXT,
properties[self.CSS3_TEXT],
macros[self.CSS3_TEXT]
)
])
self.__update_knownNames()
def _expand_macros(self, dictionary, macros):
"""Expand macros in token dictionary"""
def macro_value(m):
return '(?:%s)' % macros[m.groupdict()['macro']]
for key, value in as_list(dictionary.items()):
if not hasattr(value, '__call__'):
while re.search(r'{[a-z][a-z0-9-]*}', value):
value = re.sub(r'{(?P[a-z][a-z0-9-]*)}',
macro_value, value)
dictionary[key] = value
return dictionary
def _compile_regexes(self, dictionary):
"""Compile all regular expressions into callable objects"""
for key, value in as_list(dictionary.items()):
if not hasattr(value, '__call__'):
# Compiling them now will slow down the css_parser import time,
# even if css_parser is not needed. We lazily compile them the
# first time they're needed.
# https://bitbucket.org/cthedot/css_parser/issues/72
value = util.LazyRegex('^(?:%s)$' % value, re.I)
dictionary[key] = value
return dictionary
def __update_knownNames(self):
self._knownNames = []
for properties in as_list(self._profilesProperties.values()):
self._knownNames.extend(as_list(properties.keys()))
def _getDefaultProfiles(self):
"If not explicitly set same as Profiles.profiles but in reverse order."
if not self._defaultProfiles:
return self.profiles
else:
return self._defaultProfiles
def _setDefaultProfiles(self, profiles):
"profiles may be a single or a list of profile names"
# under python 2.X this was originally basestring but given unicode literals ...
if isinstance(profiles, string_type):
self._defaultProfiles = (profiles,)
else:
self._defaultProfiles = profiles
defaultProfiles = property(_getDefaultProfiles,
_setDefaultProfiles,
doc="Names of profiles to use for validation."
"To use e.g. the CSS2 profile set "
"``css_parser.profile.defaultProfiles = "
"css_parser.profile.CSS_LEVEL_2``")
profiles = property(lambda self: self._profileNames,
doc='Names of all profiles in order as defined.')
knownNames = property(lambda self: self._knownNames,
doc="All known property names of all profiles.")
def _resetProperties(self, newMacros=None):
"reset all props from raw values as changes in macros happened"
# base
macros = Profiles._TOKEN_MACROS.copy()
macros.update(Profiles._MACROS.copy())
# former
for profile in self._profileNames:
macros.update(self._rawProfiles[profile]['macros'])
# new
if newMacros:
macros.update(newMacros)
# reset properties
self._profilesProperties.clear()
for profile in self._profileNames:
properties = self._expand_macros(
# keep raw
self._rawProfiles[profile]['properties'].copy(),
macros)
self._profilesProperties[profile] = self._compile_regexes(properties)
# save
self._usedMacros = macros
def addProfiles(self, profiles):
"""Add a list of profiles at once. Useful as if profiles define custom
macros these are used in one go. Using `addProfile` instead my be
**very** slow instead.
"""
# add macros
for profile, properties, macros in profiles:
if macros:
self._usedMacros.update(macros)
self._rawProfiles[profile] = {'macros': macros.copy()}
# only add new properties
for profile, properties, macros in profiles:
self.addProfile(profile, properties.copy(), None)
def addProfile(self, profile, properties, macros=None):
"""Add a new profile with name `profile` (e.g. 'CSS level 2')
and the given `properties`.
:param profile:
the new `profile`'s name
:param properties:
a dictionary of ``{ property-name: propery-value }`` items where
property-value is a regex which may use macros defined in given
``macros`` or the standard macros Profiles.tokens and
Profiles.generalvalues.
``propery-value`` may also be a function which takes a single
argument which is the value to validate and which should return
True or False.
Any exceptions which may be raised during this custom validation
are reported or raised as all other css_parser exceptions depending
on css_parser.log.raiseExceptions which e.g during parsing normally
is False so the exceptions would be logged only.
:param macros:
may be used in the given properties definitions. There are some
predefined basic macros which may always be used in
:attr:`Profiles._TOKEN_MACROS` and :attr:`Profiles._MACROS`.
"""
if macros:
# check if known macros would change and if yes reset properties
if len(set(macros.keys()).intersection(as_list(self._usedMacros.keys()))):
self._resetProperties(newMacros=macros)
else:
# no replacement, simply continue
self._usedMacros.update(macros)
else:
# might have been set by addProfiles before
try:
macros = self._rawProfiles[profile]['macros']
except KeyError:
macros = {}
# save name and raw props/macros if macros change to completely reset
self._profileNames.append(profile)
self._rawProfiles[profile] = {'properties': properties.copy(),
'macros': macros.copy()}
# prepare and save properties
properties = self._expand_macros(properties, self._usedMacros)
self._profilesProperties[profile] = self._compile_regexes(properties)
self.__update_knownNames()
def removeProfile(self, profile=None, all=False):
"""Remove `profile` or remove `all` profiles.
If the removed profile used custom macros all remaining profiles
are reset to reflect the macro changes. This may be quite an expensive
operation!
:param profile:
profile name to remove
:param all:
if ``True`` removes all profiles to start with a clean state
:exceptions:
- :exc:`css_parser.profiles.NoSuchProfileException`:
If given `profile` cannot be found.
"""
if all:
self._profilesProperties.clear()
self._rawProfiles.clear()
del self._profileNames[:]
else:
reset = False
try:
if (self._rawProfiles[profile]['macros']):
reset = True
del self._profilesProperties[profile]
del self._rawProfiles[profile]
del self._profileNames[self._profileNames.index(profile)]
except KeyError:
raise NoSuchProfileException('No profile %r.' % profile)
else:
if reset:
# reset properties as macros were removed
self._resetProperties()
self.__update_knownNames()
def propertiesByProfile(self, profiles=None):
"""Generator: Yield property names, if no `profiles` is given all
profile's properties are used.
:param profiles:
a single profile name or a list of names.
"""
if not profiles:
profiles = self.profiles
elif isinstance(profiles, string_type):
profiles = (profiles, )
try:
for profile in sorted(profiles):
for name in sorted(self._profilesProperties[profile].keys()):
yield name
except KeyError as e:
raise NoSuchProfileException(e)
def validate(self, name, value):
"""Check if `value` is valid for given property `name` using **any**
profile.
:param name:
a property name
:param value:
a CSS value (string)
:returns:
if the `value` is valid for the given property `name` in any
profile
"""
for profile in self.profiles:
if name in self._profilesProperties[profile]:
try:
# custom validation errors are caught
r = bool(self._profilesProperties[profile][name](value))
except Exception as e:
# TODO: more specific exception?
# Validate should not be fatal though!
self._log.error(e, error=Exception)
r = False
if r:
return r
return False
def validateWithProfile(self, name, value, profiles=None):
"""Check if `value` is valid for given property `name` returning
``(valid, profile)``.
:param name:
a property name
:param value:
a CSS value (string)
:param profiles:
internal parameter used by Property.validate only
:returns:
``valid, matching, profiles`` where ``valid`` is if the `value`
is valid for the given property `name` in any profile,
``matching==True`` if it is valid in the given `profiles`
and ``profiles`` the profile names for which the value is valid
(or ``[]`` if not valid at all)
Example::
>>> css_parser.profile.defaultProfiles = css_parser.profile.CSS_LEVEL_2
>>> print css_parser.profile.validateWithProfile('color', 'rgba(1,1,1,1)')
(True, False, Profiles.CSS3_COLOR)
"""
if name not in self.knownNames:
return False, False, []
else:
if not profiles:
profiles = self.defaultProfiles
elif isinstance(profiles, string_type):
profiles = (profiles, )
for profilename in reversed(profiles):
# check given profiles
if name in self._profilesProperties[profilename]:
validate = self._profilesProperties[profilename][name]
try:
if validate(value):
return True, True, [profilename]
except Exception as e:
self._log.error(e, error=Exception)
for profilename in (p for p in self._profileNames
if p not in profiles):
# check remaining profiles as well
if name in self._profilesProperties[profilename]:
validate = self._profilesProperties[profilename][name]
try:
if validate(value):
return True, False, [profilename]
except Exception as e:
self._log.error(e, error=Exception)
names = []
for profilename, properties in as_list(self._profilesProperties.items()):
# return profile to which name belongs
if name in as_list(properties.keys()):
names.append(profilename)
names.sort()
return False, False, names
properties = {}
macros = {}
"""
Define some regular expression fragments that will be used as
macros within the CSS property value regular expressions.
"""
macros[Profiles.CSS_LEVEL_2] = {
'background-color': r'{color}|transparent|inherit',
'background-image': r'{uri}|none|inherit',
'background-position': r'({percentage}|{length}|left|center|right)(\s*({percentage}|{length}|top|center|bottom))?|((top|center|bottom)\s*(left|center|right)?)|((left|center|right)\s*(top|center|bottom)?)|inherit', # noqa
'background-repeat': r'repeat|repeat-x|repeat-y|no-repeat|inherit',
'background-attachment': r'scroll|fixed|inherit',
'shape': r'rect\(({w}({length}|auto}){w},){3}{w}({length}|auto){w}\)',
'counter': r'counter\({w}{ident}{w}(?:,{w}{list-style-type}{w})?\)',
'identifier': r'{ident}',
'family-name': r'{string}|({ident}(\s+{ident})*)',
'generic-family': r'serif|sans-serif|cursive|fantasy|monospace',
'absolute-size': r'(x?x-)?(small|large)|medium',
'relative-size': r'smaller|larger',
'font-family': r'({family-name}({w},{w}{family-name})*)|inherit',
'font-size': r'{absolute-size}|{relative-size}|{positivelength}|{percentage}|inherit',
'font-style': r'normal|italic|oblique|inherit',
'font-variant': r'normal|small-caps|inherit',
'font-weight': r'normal|bold|bolder|lighter|[1-9]00|inherit',
'line-height': r'normal|{number}|{length}|{percentage}|inherit',
'list-style-image': r'{uri}|none|inherit',
'list-style-position': r'inside|outside|inherit',
'list-style-type': r'disc|circle|square|decimal|decimal-leading-zero|lower-roman|upper-roman|lower-greek|lower-(latin|alpha)|upper-(latin|alpha)|armenian|georgian|none|inherit', # noqa
'margin-width': r'{length}|{percentage}|auto',
'padding-width': r'{length}|{percentage}',
'specific-voice': r'{ident}',
'generic-voice': r'male|female|child',
'content': r'{string}|{uri}|{counter}|attr\({w}{ident}{w}\)|open-quote|close-quote|no-open-quote|no-close-quote',
'background-attrs': r'{background-color}|{background-image}|{background-repeat}|{background-attachment}|{background-position}', # noqa
'list-attrs': r'{list-style-type}|{list-style-position}|{list-style-image}',
'font-attrs': r'{font-style}|{font-variant}|{font-weight}',
'text-attrs': r'underline|overline|line-through|blink',
'overflow': r'visible|hidden|scroll|auto|inherit',
}
"""
Define the regular expressions for validation all CSS values
"""
properties[Profiles.CSS_LEVEL_2] = {
'azimuth': r'{angle}|(behind\s+)?(left-side|far-left|left|center-left|center|center-right|right|far-right|right-side)(\s+behind)?|behind|leftwards|rightwards|inherit', # noqa
'background-attachment': r'{background-attachment}',
'background-color': r'{background-color}',
'background-image': r'{background-image}',
'background-position': r'{background-position}',
'background-repeat': r'{background-repeat}',
# Each piece should only be allowed one time
'background': r'{background-attrs}(\s+{background-attrs})*|inherit',
'border-collapse': r'collapse|separate|inherit',
'border-spacing': r'{length}(\s+{length})?|inherit',
'bottom': r'{length}|{percentage}|auto|inherit',
'caption-side': r'top|bottom|inherit',
'clear': r'none|left|right|both|inherit',
'clip': r'{shape}|auto|inherit',
'color': r'{color}|inherit',
'content': r'none|normal|{content}(\s+{content})*|inherit',
'counter-increment': r'({ident}(\s+{integer})?)(\s+({ident}(\s+{integer})?))*|none|inherit',
'counter-reset': r'({ident}(\s+{integer})?)(\s+({ident}(\s+{integer})?))*|none|inherit',
'cue-after': r'{uri}|none|inherit',
'cue-before': r'{uri}|none|inherit',
'cue': r'({uri}|none|inherit){1,2}|inherit',
'direction': r'ltr|rtl|inherit',
'display': r'inline|block|list-item|run-in|inline-block|table|inline-table|table-row-group|table-header-group|table-barter-group|table-row|table-column-group|table-column|table-cell|table-caption|none|inherit', # noqa
'elevation': r'{angle}|below|level|above|higher|lower|inherit',
'empty-cells': r'show|hide|inherit',
'float': r'left|right|none|inherit',
'font-family': r'{font-family}',
'font-size': r'{font-size}',
'font-style': r'{font-style}',
'font-variant': r'{font-variant}',
'font-weight': r'{font-weight}',
'font': r'(({font-attrs}\s+)*{font-size}({w}/{w}{line-height})?\s+{font-family})|caption|icon|menu|message-box|small-caption|status-bar|inherit', # noqa
'height': r'{length}|{percentage}|auto|inherit',
'left': r'{length}|{percentage}|auto|inherit',
'letter-spacing': r'normal|{length}|inherit',
'line-height': r'{line-height}',
'list-style-image': r'{list-style-image}',
'list-style-position': r'{list-style-position}',
'list-style-type': r'{list-style-type}',
'list-style': r'{list-attrs}(\s+{list-attrs})*|inherit',
'margin-right': r'{margin-width}|inherit',
'margin-left': r'{margin-width}|inherit',
'margin-top': r'{margin-width}|inherit',
'margin-bottom': r'{margin-width}|inherit',
'margin': r'{margin-width}(\s+{margin-width}){0,3}|inherit',
'max-height': r'{length}|{percentage}|none|inherit',
'max-width': r'{length}|{percentage}|none|inherit',
'min-height': r'{length}|{percentage}|none|inherit',
'min-width': r'{length}|{percentage}|none|inherit',
'orphans': r'{integer}|inherit',
'overflow': r'{overflow}',
'padding-top': r'{padding-width}|inherit',
'padding-right': r'{padding-width}|inherit',
'padding-bottom': r'{padding-width}|inherit',
'padding-left': r'{padding-width}|inherit',
'padding': r'{padding-width}(\s+{padding-width}){0,3}|inherit',
'page-break-after': r'auto|always|avoid|left|right|inherit',
'page-break-before': r'auto|always|avoid|left|right|inherit',
'page-break-inside': r'avoid|auto|inherit',
'pause-after': r'{time}|{percentage}|inherit',
'pause-before': r'{time}|{percentage}|inherit',
'pause': r'({time}|{percentage}){1,2}|inherit',
'pitch-range': r'{number}|inherit',
'pitch': r'{frequency}|x-low|low|medium|high|x-high|inherit',
'play-during': r'{uri}(\s+(mix|repeat))*|auto|none|inherit',
'position': r'static|relative|absolute|fixed|inherit',
'quotes': r'({string}\s+{string})(\s+{string}\s+{string})*|none|inherit',
'richness': r'{number}|inherit',
'right': r'{length}|{percentage}|auto|inherit',
'speak-header': r'once|always|inherit',
'speak-numeral': r'digits|continuous|inherit',
'speak-punctuation': r'code|none|inherit',
'speak': r'normal|none|spell-out|inherit',
'speech-rate': r'{number}|x-slow|slow|medium|fast|x-fast|faster|slower|inherit',
'stress': r'{number}|inherit',
'table-layout': r'auto|fixed|inherit',
'text-align': r'left|right|center|justify|inherit',
'text-decoration': r'none|{text-attrs}(\s+{text-attrs})*|inherit',
'text-indent': r'{length}|{percentage}|inherit',
'text-transform': r'capitalize|uppercase|lowercase|none|inherit',
'top': r'{length}|{percentage}|auto|inherit',
'unicode-bidi': r'normal|embed|bidi-override|inherit',
'vertical-align': r'baseline|sub|super|top|text-top|middle|bottom|text-bottom|{percentage}|{length}|inherit',
'visibility': r'visible|hidden|collapse|inherit',
'voice-family': r'({specific-voice}|{generic-voice}{w},{w})*({specific-voice}|{generic-voice})|inherit',
'volume': r'{number}|{percentage}|silent|x-soft|soft|medium|loud|x-loud|inherit',
'white-space': r'normal|pre|nowrap|pre-wrap|pre-line|inherit',
'widows': r'{integer}|inherit',
'width': r'{length}|{percentage}|auto|inherit',
'word-spacing': r'normal|{length}|inherit',
'z-index': r'auto|{integer}|inherit',
}
macros[Profiles.CSS3_BACKGROUNDS_AND_BORDERS] = {
'border-style': 'none|hidden|dotted|dashed|solid|double|groove|ridge|inset|outset',
'border-width': '{length}|thin|medium|thick',
'b1': r'{border-width}?({w}{border-style})?({w}{color})?',
'b2': r'{border-width}?({w}{color})?({w}{border-style})?',
'b3': r'{border-style}?({w}{border-width})?({w}{color})?',
'b4': r'{border-style}?({w}{color})?({w}{border-width})?',
'b5': r'{color}?({w}{border-style})?({w}{border-width})?',
'b6': r'{color}?({w}{border-width})?({w}{border-style})?',
'border-attrs': r'{b1}|{b2}|{b3}|{b4}|{b5}|{b6}',
'border-radius-part': r'({length}|{percentage})(\s+({length}|{percentage}))?'
}
properties[Profiles.CSS3_BACKGROUNDS_AND_BORDERS] = {
'border-color': r'({color}|transparent)(\s+({color}|transparent)){0,3}|inherit',
'border-style': r'{border-style}(\s+{border-style}){0,3}|inherit',
'border-top': r'{border-attrs}|inherit',
'border-right': r'{border-attrs}|inherit',
'border-bottom': r'{border-attrs}|inherit',
'border-left': r'{border-attrs}|inherit',
'border-top-color': r'{color}|transparent|inherit',
'border-right-color': r'{color}|transparent|inherit',
'border-bottom-color': r'{color}|transparent|inherit',
'border-left-color': r'{color}|transparent|inherit',
'border-top-style': r'{border-style}|inherit',
'border-right-style': r'{border-style}|inherit',
'border-bottom-style': r'{border-style}|inherit',
'border-left-style': r'{border-style}|inherit',
'border-top-width': r'{border-width}|inherit',
'border-right-width': r'{border-width}|inherit',
'border-bottom-width': r'{border-width}|inherit',
'border-left-width': r'{border-width}|inherit',
'border-width': r'{border-width}(\s+{border-width}){0,3}|inherit',
'border': r'{border-attrs}|inherit',
'border-top-right-radius': '{border-radius-part}',
'border-bottom-right-radius': '{border-radius-part}',
'border-bottom-left-radius': '{border-radius-part}',
'border-top-left-radius': '{border-radius-part}',
'border-radius': '({length}{w}|{percentage}{w}){1,4}(/{w}({length}{w}|{percentage}{w}){1,4})?',
'box-shadow': 'none|{shadow}({w},{w}{shadow})*',
}
# CSS3 Basic User Interface Module
macros[Profiles.CSS3_BASIC_USER_INTERFACE] = {
'border-style': macros[Profiles.CSS3_BACKGROUNDS_AND_BORDERS]['border-style'],
'border-width': macros[Profiles.CSS3_BACKGROUNDS_AND_BORDERS]['border-width'],
'outline-1': r'{outline-color}(\s+{outline-style})?(\s+{outline-width})?',
'outline-2': r'{outline-color}(\s+{outline-width})?(\s+{outline-style})?',
'outline-3': r'{outline-style}(\s+{outline-color})?(\s+{outline-width})?',
'outline-4': r'{outline-style}(\s+{outline-width})?(\s+{outline-color})?',
'outline-5': r'{outline-width}(\s+{outline-color})?(\s+{outline-style})?',
'outline-6': r'{outline-width}(\s+{outline-style})?(\s+{outline-color})?',
'outline-color': r'{color}|invert|inherit',
'outline-style': r'auto|{border-style}|inherit',
'outline-width': r'{border-width}|inherit',
}
properties[Profiles.CSS3_BASIC_USER_INTERFACE] = {
'box-sizing': r'content-box|border-box',
'cursor': r'((({uri}{w}({number}{w}{number}{w})?,{w})*)?(auto|default|none|context-menu|help|pointer|progress|wait|cell|crosshair|text|vertical-text|alias|copy|move|no-drop|not-allowed|(e|n|ne|nw|s|se|sw|w|ew|ns|nesw|nwse|col|row)-resize|all-scroll))|inherit', # noqa
'nav-index': r'auto|{number}|inherit',
'outline-color': r'{outline-color}',
'outline-style': r'{outline-style}',
'outline-width': r'{outline-width}',
'outline-offset': r'{length}|inherit',
# 'outline': r'{outline-attrs}(\s+{outline-attrs})*|inherit',
'outline': r'{outline-1}|{outline-2}|{outline-3}|{outline-4}|{outline-5}|{outline-6}|inherit',
'resize': 'none|both|horizontal|vertical|inherit',
}
# CSS Box Module Level 3
macros[Profiles.CSS3_BOX] = {
'overflow': macros[Profiles.CSS_LEVEL_2]['overflow']
}
properties[Profiles.CSS3_BOX] = {
'overflow': '{overflow}{w}{overflow}?|inherit',
'overflow-x': '{overflow}|inherit',
'overflow-y': '{overflow}|inherit'
}
# CSS Color Module Level 3
macros[Profiles.CSS3_COLOR] = {
# orange and transparent in CSS 2.1
'namedcolor': r'(currentcolor|transparent|aqua|black|blue|fuchsia|gray|green|lime|maroon|navy|olive|orange|purple|red|silver|teal|white|yellow)', # noqa
# orange?
'rgbacolor': r'rgba\({w}{int}{w}\,{w}{int}{w}\,{w}{int}{w}\,{w}{num}{w}\)|rgba\({w}{num}%{w}\,{w}{num}%{w}\,{w}{num}%{w}\,{w}{num}{w}\)', # noqa
'hslcolor': r'hsl\({w}{int}{w}\,{w}{num}%{w}\,{w}{num}%{w}\)|hsla\({w}{int}{w}\,{w}{num}%{w}\,{w}{num}%{w}\,{w}{num}{w}\)', # noqa
'x11color': r'aliceblue|antiquewhite|aqua|aquamarine|azure|beige|bisque|black|blanchedalmond|blue|blueviolet|brown|burlywood|cadetblue|chartreuse|chocolate|coral|cornflowerblue|cornsilk|crimson|cyan|darkblue|darkcyan|darkgoldenrod|darkgray|darkgreen|darkgrey|darkkhaki|darkmagenta|darkolivegreen|darkorange|darkorchid|darkred|darksalmon|darkseagreen|darkslateblue|darkslategray|darkslategrey|darkturquoise|darkviolet|deeppink|deepskyblue|dimgray|dimgrey|dodgerblue|firebrick|floralwhite|forestgreen|fuchsia|gainsboro|ghostwhite|gold|goldenrod|gray|green|greenyellow|grey|honeydew|hotpink|indianred|indigo|ivory|khaki|lavender|lavenderblush|lawngreen|lemonchiffon|lightblue|lightcoral|lightcyan|lightgoldenrodyellow|lightgray|lightgreen|lightgrey|lightpink|lightsalmon|lightseagreen|lightskyblue|lightslategray|lightslategrey|lightsteelblue|lightyellow|lime|limegreen|linen|magenta|maroon|mediumaquamarine|mediumblue|mediumorchid|mediumpurple|mediumseagreen|mediumslateblue|mediumspringgreen|mediumturquoise|mediumvioletred|midnightblue|mintcream|mistyrose|moccasin|navajowhite|navy|oldlace|olive|olivedrab|orange|orangered|orchid|palegoldenrod|palegreen|paleturquoise|palevioletred|papayawhip|peachpuff|peru|pink|plum|powderblue|purple|red|rosybrown|royalblue|saddlebrown|salmon|sandybrown|seagreen|seashell|sienna|silver|skyblue|slateblue|slategray|slategrey|snow|springgreen|steelblue|tan|teal|thistle|tomato|turquoise|violet|wheat|white|whitesmoke|yellow|yellowgreen', # noqa
'uicolor': r'(ActiveBorder|ActiveCaption|AppWorkspace|Background|ButtonFace|ButtonHighlight|ButtonShadow|ButtonText|CaptionText|GrayText|Highlight|HighlightText|InactiveBorder|InactiveCaption|InactiveCaptionText|InfoBackground|InfoText|Menu|MenuText|Scrollbar|ThreeDDarkShadow|ThreeDFace|ThreeDHighlight|ThreeDLightShadow|ThreeDShadow|Window|WindowFrame|WindowText)', # noqa
'color': r'{namedcolor}|{hexcolor}|{rgbcolor}|{rgbacolor}|{hslcolor}|{x11color}|inherit',
}
properties[Profiles.CSS3_COLOR] = {
'opacity': r'{num}|inherit',
}
# CSS Fonts Module Level 3 http://www.w3.org/TR/css3-fonts/
macros[Profiles.CSS3_FONTS] = {
# 'family-name': r'{string}|{ident}',
'family-name': r'{string}|({ident}(\s+{ident})*)',
'font-face-name': r'local\({w}{family-name}{w}\)',
'font-stretch-names': r'(ultra-condensed|extra-condensed|condensed|semi-condensed|semi-expanded|expanded|extra-expanded|ultra-expanded)', # noqa
'unicode-range': r'[uU]\+[0-9A-Fa-f?]{1,6}(\-[0-9A-Fa-f]{1,6})?'
}
properties[Profiles.CSS3_FONTS] = {
'font-size-adjust': r'{number}|none|inherit',
'font-stretch': r'normal|wider|narrower|{font-stretch-names}|inherit'
}
properties[Profiles.CSS3_FONT_FACE] = {
'font-family': '{family-name}',
'font-stretch': r'{font-stretch-names}',
'font-style': r'normal|italic|oblique',
'font-weight': r'normal|bold|[1-9]00',
'src': r'({uri}{w}(format\({w}{string}{w}(\,{w}{string}{w})*\))?|{font-face-name})({w},{w}({uri}{w}(format\({w}{string}{w}(\,{w}{string}{w})*\))?|{font-face-name}))*', # noqa
'unicode-range': '{unicode-range}({w},{w}{unicode-range})*'
}
# CSS3 Paged Media
macros[Profiles.CSS3_PAGED_MEDIA] = {
'page-size': 'a5|a4|a3|b5|b4|letter|legal|ledger',
'page-orientation': 'portrait|landscape',
'page-1': '{page-size}(?:{w}{page-orientation})?',
'page-2': '{page-orientation}(?:{w}{page-size})?',
'page-size-orientation': '{page-1}|{page-2}',
'pagebreak': 'auto|always|avoid|left|right'
}
properties[Profiles.CSS3_PAGED_MEDIA] = {
'fit': 'fill|hidden|meet|slice',
'fit-position': r'auto|(({percentage}|{length})(\s*({percentage}|{length}))?|((top|center|bottom)\s*(left|center|right)?)|((left|center|right)\s*(top|center|bottom)?))', # noqa
'image-orientation': 'auto|{angle}',
'orphans': r'{integer}|inherit',
'page': 'auto|{ident}',
'page-break-before': '{pagebreak}|inherit',
'page-break-after': '{pagebreak}|inherit',
'page-break-inside': 'auto|avoid|inherit',
'size': '({length}{w}){1,2}|auto|{page-size-orientation}',
'widows': r'{integer}|inherit'
}
macros[Profiles.CSS3_TEXT] = {
}
properties[Profiles.CSS3_TEXT] = {
'text-shadow': 'none|{shadow}({w},{w}{shadow})*',
}
css-parser-1.0.4/src/css_parser/sac.py 0000644 0001750 0001750 00000041726 13407702010 020127 0 ustar kovid kovid 0000000 0000000 #!/usr/bin/env python
from __future__ import unicode_literals, division, absolute_import, print_function
from . import tokenize2
from . import errorhandler
import codecs
from . import helper
"""A validating CSSParser"""
__docformat__ = 'restructuredtext'
__version__ = '$Id: parse.py 1754 2009-05-30 14:50:13Z cthedot $'
import sys
PY2 = sys.version_info[0] == 2
class ErrorHandler(object):
"""Basic class for CSS error handlers.
This class class provides a default implementation ignoring warnings and
recoverable errors and throwing a SAXParseException for fatal errors.
If a CSS application needs to implement customized error handling, it must
extend this class and then register an instance with the CSS parser
using the parser's setErrorHandler method. The parser will then report all
errors and warnings through this interface.
The parser shall use this class instead of throwing an exception: it is
up to the application whether to throw an exception for different types of
errors and warnings. Note, however, that there is no requirement that the
parser continue to provide useful information after a call to fatalError
(in other words, a CSS driver class could catch an exception and report a
fatalError).
"""
def __init__(self):
self._log = errorhandler.ErrorHandler()
def error(self, exception, token=None):
self._log.error(exception, token, neverraise=True)
def fatal(self, exception, token=None):
self._log.fatal(exception, token)
def warn(self, exception, token=None):
self._log.warn(exception, token, neverraise=True)
class DocumentHandler(object):
"""
void endFontFace()
Receive notification of the end of a font face statement.
void endMedia(SACMediaList media)
Receive notification of the end of a media statement.
void endPage(java.lang.String name, java.lang.String pseudo_page)
Receive notification of the end of a media statement.
void importStyle(java.lang.String uri, SACMediaList media, java.lang.String defaultNamespaceURI)
Receive notification of a import statement in the style sheet.
void startFontFace()
Receive notification of the beginning of a font face statement.
void startMedia(SACMediaList media)
Receive notification of the beginning of a media statement.
void startPage(java.lang.String name, java.lang.String pseudo_page)
Receive notification of the beginning of a page statement.
"""
def __init__(self):
def log(msg):
sys.stderr.write('INFO\t%s\n' % msg)
self._log = log
def comment(self, text, line=None, col=None):
"Receive notification of a comment."
self._log("comment %r at [%s, %s]" % (text, line, col))
def startDocument(self, encoding):
"Receive notification of the beginning of a style sheet."
# source
self._log("startDocument encoding=%s" % encoding)
def endDocument(self, source=None, line=None, col=None):
"Receive notification of the end of a document."
self._log("endDocument EOF")
def importStyle(self, uri, media, name, line=None, col=None):
"Receive notification of a import statement in the style sheet."
# defaultNamespaceURI???
self._log("importStyle at [%s, %s]" % (line, col))
def namespaceDeclaration(self, prefix, uri, line=None, col=None):
"Receive notification of an unknown rule t-rule not supported by this parser."
# prefix might be None!
self._log("namespaceDeclaration at [%s, %s]" % (line, col))
def startSelector(self, selectors=None, line=None, col=None):
"Receive notification of the beginning of a rule statement."
# TODO selectorList!
self._log("startSelector at [%s, %s]" % (line, col))
def endSelector(self, selectors=None, line=None, col=None):
"Receive notification of the end of a rule statement."
self._log("endSelector at [%s, %s]" % (line, col))
def property(self, name, value='TODO', important=False, line=None, col=None):
"Receive notification of a declaration."
# TODO: value is LexicalValue?
self._log("property %r at [%s, %s]" % (name, line, col))
def ignorableAtRule(self, atRule, line=None, col=None):
"Receive notification of an unknown rule t-rule not supported by this parser."
self._log("ignorableAtRule %r at [%s, %s]" % (atRule, line, col))
class EchoHandler(DocumentHandler):
"Echos all input to property `out`"
def __init__(self):
super(EchoHandler, self).__init__()
self._out = []
out = property(lambda self: ''.join(self._out))
def startDocument(self, encoding):
super(EchoHandler, self).startDocument(encoding)
if 'utf-8' != encoding:
self._out.append('@charset "%s";\n' % encoding)
# def comment(self, text, line=None, col=None):
# self._out.append(u'/*%s*/' % text)
def importStyle(self, uri, media, name, line=None, col=None):
"Receive notification of a import statement in the style sheet."
# defaultNamespaceURI???
super(EchoHandler, self).importStyle(uri, media, name, line, col)
self._out.append('@import %s%s%s;\n' % (helper.string(uri),
'%s ' % media if media else '',
'%s ' % name if name else '')
)
def namespaceDeclaration(self, prefix, uri, line=None, col=None):
super(EchoHandler, self).namespaceDeclaration(prefix, uri, line, col)
self._out.append('@namespace %s%s;\n' % ('%s ' % prefix if prefix else '',
helper.string(uri)))
def startSelector(self, selectors=None, line=None, col=None):
super(EchoHandler, self).startSelector(selectors, line, col)
if selectors:
self._out.append(', '.join(selectors))
self._out.append(' {\n')
def endSelector(self, selectors=None, line=None, col=None):
self._out.append(' }')
def property(self, name, value, important=False, line=None, col=None):
super(EchoHandler, self).property(name, value, line, col)
self._out.append(' %s: %s%s;\n' % (name, value,
' !important' if important else ''))
class Parser(object):
"""
java.lang.String getParserVersion()
Returns a string about which CSS language is supported by this parser.
boolean parsePriority(InputSource source)
Parse a CSS priority value (e.g.
LexicalUnit parsePropertyValue(InputSource source)
Parse a CSS property value.
void parseRule(InputSource source)
Parse a CSS rule.
SelectorList parseSelectors(InputSource source)
Parse a comma separated list of selectors.
void parseStyleDeclaration(InputSource source)
Parse a CSS style declaration (without '{' and '}').
void parseStyleSheet(InputSource source)
Parse a CSS document.
void parseStyleSheet(java.lang.String uri)
Parse a CSS document from a URI.
void setConditionFactory(ConditionFactory conditionFactory)
void setDocumentHandler(DocumentHandler handler)
Allow an application to register a document event handler.
void setErrorHandler(ErrorHandler handler)
Allow an application to register an error event handler.
void setLocale(java.util.Locale locale)
Allow an application to request a locale for errors and warnings.
void setSelectorFactory(SelectorFactory selectorFactory)
"""
def __init__(self, documentHandler=None, errorHandler=None):
self._tokenizer = tokenize2.Tokenizer()
if documentHandler:
self.setDocumentHandler(documentHandler)
else:
self.setDocumentHandler(DocumentHandler())
if errorHandler:
self.setErrorHandler(errorHandler)
else:
self.setErrorHandler(ErrorHandler())
def parseString(self, cssText, encoding=None):
if isinstance(cssText, str):
cssText = codecs.getdecoder('css')(cssText, encoding=encoding)[0]
tokens = self._tokenizer.tokenize(cssText, fullsheet=True)
def COMMENT(val, line, col):
self._handler.comment(val[2:-2], line, col)
def EOF(val, line, col):
self._handler.endDocument(val, line, col)
def simple(t):
map = {'COMMENT': COMMENT,
'S': lambda val, line, col: None,
'EOF': EOF}
type_, val, line, col = t
if type_ in map:
map[type_](val, line, col)
return True
else:
return False
# START PARSING
t = next(tokens)
type_, val, line, col = t
encoding = 'utf-8'
if 'CHARSET_SYM' == type_:
# @charset "encoding";
# S
encodingtoken = next(tokens)
semicolontoken = next(tokens)
if 'STRING' == type_:
encoding = helper.stringvalue(val)
# ;
if 'STRING' == encodingtoken[0] and semicolontoken:
encoding = helper.stringvalue(encodingtoken[1])
else:
self._errorHandler.fatal('Invalid @charset')
t = next(tokens)
type_, val, line, col = t
self._handler.startDocument(encoding)
while True:
start = (line, col)
try:
if simple(t):
pass
elif 'ATKEYWORD' == type_ or type_ in ('PAGE_SYM', 'MEDIA_SYM', 'FONT_FACE_SYM'):
atRule = [val]
braces = 0
while True:
# read till end ;
# TODO: or {}
t = next(tokens)
type_, val, line, col = t
atRule.append(val)
if ';' == val and not braces:
break
elif '{' == val:
braces += 1
elif '}' == val:
braces -= 1
if braces == 0:
break
self._handler.ignorableAtRule(''.join(atRule), *start)
elif 'IMPORT_SYM' == type_:
# import URI or STRING media? name?
uri, media, name = None, None, None
while True:
t = next(tokens)
type_, val, line, col = t
if 'STRING' == type_:
uri = helper.stringvalue(val)
elif 'URI' == type_:
uri = helper.urivalue(val)
elif ';' == val:
break
if uri:
self._handler.importStyle(uri, media, name)
else:
self._errorHandler.error('Invalid @import'
' declaration at %r'
% (start,))
elif 'NAMESPACE_SYM' == type_:
prefix, uri = None, None
while True:
t = next(tokens)
type_, val, line, col = t
if 'IDENT' == type_:
prefix = val
elif 'STRING' == type_:
uri = helper.stringvalue(val)
elif 'URI' == type_:
uri = helper.urivalue(val)
elif ';' == val:
break
if uri:
self._handler.namespaceDeclaration(prefix, uri, *start)
else:
self._errorHandler.error('Invalid @namespace'
' declaration at %r'
% (start,))
else:
# CSSSTYLERULE
selector = []
selectors = []
while True:
# selectors[, selector]* {
if 'S' == type_:
selector.append(' ')
elif simple(t):
pass
elif ',' == val:
selectors.append(''.join(selector).strip())
selector = []
elif '{' == val:
selectors.append(''.join(selector).strip())
self._handler.startSelector(selectors, *start)
break
else:
selector.append(val)
t = next(tokens)
type_, val, line, col = t
end = None
while True:
# name: value [!important][;name: value [!important]]*;?
name, value, important = None, [], False
while True:
# name:
t = next(tokens)
type_, val, line, col = t
if 'S' == type_:
pass
elif simple(t):
pass
elif 'IDENT' == type_:
if name:
self._errorHandler.error('more than one property name', t)
else:
name = val
elif ':' == val:
if not name:
self._errorHandler.error('no property name', t)
break
elif ';' == val:
self._errorHandler.error('premature end of property', t)
end = val
break
elif '}' == val:
if name:
self._errorHandler.error('premature end of property', t)
end = val
break
else:
self._errorHandler.error('unexpected property name token %r' % val, t)
while not ';' == end and not '}' == end:
# value !;}
t = next(tokens)
type_, val, line, col = t
if 'S' == type_:
value.append(' ')
elif simple(t):
pass
elif '!' == val or ';' == val or '}' == val:
value = ''.join(value).strip()
if not value:
self._errorHandler.error('premature end of property (no value)', t)
end = val
break
else:
value.append(val)
while '!' == end:
# !important
t = next(tokens)
type_, val, line, col = t
if simple(t):
pass
elif 'IDENT' == type_ and not important:
important = True
elif ';' == val or '}' == val:
end = val
break
else:
self._errorHandler.error('unexpected priority token %r' % val)
if name and value:
self._handler.property(name, value, important)
if '}' == end:
self._handler.endSelector(selectors, line=line, col=col)
break
else:
# reset
end = None
else:
self._handler.endSelector(selectors, line=line, col=col)
t = next(tokens)
type_, val, line, col = t
except StopIteration:
break
def setDocumentHandler(self, handler):
"Allow an application to register a document event `handler`."
self._handler = handler
def setErrorHandler(self, handler):
"TODO"
self._errorHandler = handler
css-parser-1.0.4/src/css_parser/script.py 0000644 0001750 0001750 00000031532 13407702533 020671 0 ustar kovid kovid 0000000 0000000 from __future__ import (absolute_import, division, print_function,
unicode_literals)
import codecs
import errno
import logging
import os
import sys
import css_parser
import css_parser.encutils as encutils
"""classes and functions used by css_parser scripts
"""
__all__ = ['CSSCapture', 'csscombine']
__docformat__ = 'restructuredtext'
__version__ = '$Id: parse.py 1323 2008-07-06 18:13:57Z cthedot $'
if sys.version_info[0] >= 3:
from html.parser import HTMLParser as htmlparser_HTMLParser
from urllib.parse import urlsplit as urllib_urlsplit
from urllib.parse import urljoin as urllib_urljoin
from urllib.request import urlopen as urllib_urlopen
from urllib.request import Request as urllib_Request
from urllib.error import HTTPError as urllib_HTTPError
else:
from HTMLParser import HTMLParser as htmlparser_HTMLParser
from urlparse import urlsplit as urllib_urlsplit
from urlparse import urljoin as urllib_urljoin
from urllib2 import urlopen as urllib_urlopen
from urllib2 import HTTPError as urllib_HTTPError
from urllib2 import Request as urllib_Request
# types of sheets in HTML
LINK = 0 #
STYLE = 1 #
class CSSCaptureHTMLParser(htmlparser_HTMLParser):
"""CSSCapture helper: Parse given data for link and style elements"""
curtag = ''
sheets = [] # (type, [atts, cssText])
def _loweratts(self, atts):
return dict([(a.lower(), v.lower()) for a, v in atts])
def handle_starttag(self, tag, atts):
if tag == 'link':
atts = self._loweratts(atts)
if 'text/css' == atts.get('type', ''):
self.sheets.append((LINK, atts))
elif tag == 'style':
# also get content of style
atts = self._loweratts(atts)
if 'text/css' == atts.get('type', ''):
self.sheets.append((STYLE, [atts, '']))
self.curtag = tag
else:
# close as only intersting