def parse_stylesheets(tree, url, session=None):
"""Find and parse the stylesheets in ``tree``.
Return two :class:`cssselect2.Matcher` objects,
for normal and !important declarations.
"""
normal_matcher = cssselect2.Matcher()
important_matcher = cssselect2.Matcher()
for stylesheet in find_stylesheets(tree):
for rule in find_stylesheets_rules(tree,
stylesheet,
url,
session=session):
normal_declarations, important_declarations = parse_declarations(
rule.content)
for selector in cssselect2.compile_selector_list(rule.prelude):
if (selector.pseudo_element is None
and not selector.never_matches):
if normal_declarations:
normal_matcher.add_selector(selector,
normal_declarations)
if important_declarations:
important_matcher.add_selector(selector,
important_declarations)
return normal_matcher, important_matcher
def parse_stylesheets(tree, url):
"""Find stylesheets and return rule matchers in given tree."""
normal_matcher = cssselect2.Matcher()
important_matcher = cssselect2.Matcher()
# Find stylesheets
# TODO: support contentStyleType on <svg>
stylesheets = []
for element in tree.etree_element.iter():
# http://www.w3.org/TR/SVG/styling.html#StyleElement
if (element.tag == '{http://www.w3.org/2000/svg}style' and
element.get('type', 'text/css') == 'text/css' and
element.text):
# TODO: pass href for relative URLs
# TODO: support media types
# TODO: what if <style> has children elements?
stylesheets.append(tinycss2.parse_stylesheet(
element.text, skip_comments=True, skip_whitespace=True))
# Parse rules and fill matchers
for stylesheet in stylesheets:
for rule in find_stylesheets_rules(tree, stylesheet, url):
normal_declarations, important_declarations = parse_declarations(
rule.content)
for selector in cssselect2.compile_selector_list(rule.prelude):
if (selector.pseudo_element is None and
not selector.never_matches):
if normal_declarations:
normal_matcher.add_selector(
selector, normal_declarations)
if important_declarations:
important_matcher.add_selector(
selector, important_declarations)
return normal_matcher, important_matcher
def __init__(self, guess=None, filename=None, url=None, file_obj=None,
string=None, encoding=None, base_url=None,
url_fetcher=default_url_fetcher, _check_mime_type=False,
media_type='print', font_config=None, counter_style=None,
matcher=None, page_rules=None):
PROGRESS_LOGGER.info(
'Step 2 - Fetching and parsing CSS - %s',
filename or url or getattr(file_obj, 'name', 'CSS string'))
result = _select_source(
guess, filename, url, file_obj, string,
base_url=base_url, url_fetcher=url_fetcher,
check_css_mime_type=_check_mime_type)
with result as (source_type, source, base_url, protocol_encoding):
if source_type == 'string' and not isinstance(source, bytes):
# unicode, no encoding
stylesheet = tinycss2.parse_stylesheet(source)
else:
if source_type == 'file_obj':
source = source.read()
stylesheet, encoding = tinycss2.parse_stylesheet_bytes(
source, environment_encoding=encoding,
protocol_encoding=protocol_encoding)
self.base_url = base_url
self.matcher = matcher or cssselect2.Matcher()
self.page_rules = [] if page_rules is None else page_rules
self.fonts = []
preprocess_stylesheet(
media_type, base_url, stylesheet, url_fetcher, self.matcher,
self.page_rules, self.fonts, font_config, counter_style)
def select(rule: str, element: et.Element) -> 'PQuery':
rules = tinycss2.parse_stylesheet(rule + ' {}', skip_whitespace=True)
matcher = cssselect2.Matcher()
for rule in rules:
selectors = cssselect2.compile_selector_list(rule.prelude)
for selector in selectors:
matcher.add_selector(selector, None)
wrapper = cssselect2.ElementWrapper.from_html_root(element)
matching_elements: List[et.Element] = []
for element in wrapper.iter_subtree():
matches = matcher.match(element)
if matches:
matching_elements.append(element.etree_element)
return PQuery(matching_elements)
from xml.etree import ElementTree
import cssselect2
import tinycss2
# Parse CSS and add rules to the matcher
matcher = cssselect2.Matcher()
rules = tinycss2.parse_stylesheet('''
body { font-size: 2em }
body p { background: red }
p { color: blue }
''', skip_whitespace=True)
for rule in rules:
selectors = cssselect2.compile_selector_list(rule.prelude)
selector_string = tinycss2.serialize(rule.prelude)
content_string = tinycss2.serialize(rule.content)
payload = (selector_string, content_string)
for selector in selectors:
matcher.add_selector(selector, payload)
# Parse HTML and find CSS rules applying to each tag
html_tree = ElementTree.fromstring('''
<html>
<body>
<p>Test</p>
</body>
