def preprocess_stylesheet(device_media_type, base_url, rules, url_fetcher):
"""Do the work that can be done early on stylesheet, before they are
in a document.
"""
selector_to_xpath = cssselect.HTMLTranslator().selector_to_xpath
for rule in rules:
if not rule.at_keyword:
declarations = list(
preprocess_declarations(base_url, rule.declarations))
if declarations:
selector_string = rule.selector.as_css()
try:
selector_list = []
for selector in cssselect.parse(selector_string):
xpath = selector_to_xpath(selector)
try:
lxml_xpath = lxml.etree.XPath(xpath)
except ValueError as exc:
# TODO: Some characters are not supported by lxml's
# XPath implementation (including control
# characters), but these characters are valid in
# the CSS2.1 specification.
raise cssselect.SelectorError(str(exc))
selector_list.append(
Selector((0, ) + selector.specificity(),
selector.pseudo_element, lxml_xpath))
for selector in selector_list:
if selector.pseudo_element not in PSEUDO_ELEMENTS:
raise cssselect.ExpressionError(
'Unknown pseudo-element: %s' %
selector.pseudo_element)
except cssselect.SelectorError as exc:
LOGGER.warn("Invalid or unsupported selector '%s', %s",
selector_string, exc)
continue
yield rule, selector_list, declarations
elif rule.at_keyword == '@import':
if not evaluate_media_query(rule.media, device_media_type):
continue
url = url_join(base_url, rule.uri, '@import at %s:%s', rule.line,
rule.column)
if url is not None:
try:
stylesheet = CSS(url=url,
url_fetcher=url_fetcher,
media_type=device_media_type)
except URLFetchingError as exc:
LOGGER.warn('Failed to load stylesheet at %s : %s', url,
exc)
else:
for result in stylesheet.rules:
yield result
elif rule.at_keyword == '@media':
if not evaluate_media_query(rule.media, device_media_type):
continue
for result in preprocess_stylesheet(device_media_type, base_url,
rule.rules, url_fetcher):
yield result
elif rule.at_keyword == '@page':
page_name, pseudo_class = rule.selector
# TODO: support named pages (see CSS3 Paged Media)
if page_name is not None:
LOGGER.warn(
'Named pages are not supported yet, the whole '
'@page %s rule was ignored.',
page_name + (':' + pseudo_class if pseudo_class else ''))
continue
declarations = list(
preprocess_declarations(base_url, rule.declarations))
# Use a double lambda to have a closure that holds page_types
match = (lambda page_types: lambda _document: page_types)(
PAGE_PSEUDOCLASS_TARGETS[pseudo_class])
specificity = rule.specificity
if declarations:
selector_list = [Selector(specificity, None, match)]
yield rule, selector_list, declarations
for margin_rule in rule.at_rules:
declarations = list(
preprocess_declarations(base_url,
margin_rule.declarations))
if declarations:
selector_list = [
Selector(specificity, margin_rule.at_keyword, match)
]
yield margin_rule, selector_list, declarations
def preprocess_stylesheet(device_media_type, base_url, stylesheet_rules,
url_fetcher, rules, fonts, font_config):
"""Do the work that can be done early on stylesheet, before they are
in a document.
"""
selector_to_xpath = cssselect.HTMLTranslator().selector_to_xpath
for rule in stylesheet_rules:
if rule.type == 'qualified-rule':
declarations = list(
preprocess_declarations(
base_url, tinycss2.parse_declaration_list(rule.content)))
if declarations:
selector_string = tinycss2.serialize(rule.prelude)
try:
selector_list = []
for selector in cssselect.parse(selector_string):
xpath = selector_to_xpath(selector)
try:
lxml_xpath = lxml.etree.XPath(xpath)
except ValueError as exc:
# TODO: Some characters are not supported by lxml's
# XPath implementation (including control
# characters), but these characters are valid in
# the CSS2.1 specification.
raise cssselect.SelectorError(str(exc))
selector_list.append(
Selector((0, ) + selector.specificity(),
selector.pseudo_element, lxml_xpath))
for selector in selector_list:
if selector.pseudo_element not in PSEUDO_ELEMENTS:
raise cssselect.ExpressionError(
'Unknown pseudo-element: %s' %
selector.pseudo_element)
except cssselect.SelectorError as exc:
LOGGER.warning("Invalid or unsupported selector '%s', %s",
selector_string, exc)
continue
rules.append((rule, selector_list, declarations))
elif rule.type == 'at-rule' and rule.at_keyword == 'import':
tokens = remove_whitespace(rule.prelude)
if tokens and tokens[0].type in ('url', 'string'):
url = tokens[0].value
else:
continue
media = parse_media_query(tokens[1:])
if media is None:
LOGGER.warning(
'Invalid media type "%s" '
'the whole @import rule was ignored at %s:%s.',
tinycss2.serialize(rule.prelude), rule.source_line,
rule.source_column)
if not evaluate_media_query(media, device_media_type):
continue
url = url_join(base_url,
url,
allow_relative=False,
context='@import at %s:%s',
context_args=(rule.source_line, rule.source_column))
if url is not None:
try:
stylesheet = CSS(url=url,
url_fetcher=url_fetcher,
media_type=device_media_type,
font_config=font_config)
except URLFetchingError as exc:
LOGGER.warning('Failed to load stylesheet at %s : %s', url,
exc)
else:
for result in stylesheet.rules:
rules.append(result)
elif rule.type == 'at-rule' and rule.at_keyword == 'media':
media = parse_media_query(rule.prelude)
if media is None:
LOGGER.warning(
'Invalid media type "%s" '
'the whole @media rule was ignored at %s:%s.',
tinycss2.serialize(rule.prelude), rule.source_line,
rule.source_column)
continue
if not evaluate_media_query(media, device_media_type):
continue
content_rules = tinycss2.parse_rule_list(rule.content)
preprocess_stylesheet(device_media_type, base_url, content_rules,
url_fetcher, rules, fonts, font_config)
elif rule.type == 'at-rule' and rule.at_keyword == 'page':
tokens = remove_whitespace(rule.prelude)
# TODO: support named pages (see CSS3 Paged Media)
if not tokens:
pseudo_class = None
specificity = (0, 0)
elif (len(tokens) == 2 and tokens[0].type == 'literal'
and tokens[0].value == ':' and tokens[1].type == 'ident'):
pseudo_class = tokens[1].lower_value
specificity = {
'first': (1, 0),
'blank': (1, 0),
'left': (0, 1),
'right': (0, 1),
}.get(pseudo_class)
if not specificity:
LOGGER.warning(
'Unknown @page pseudo-class "%s", '
'the whole @page rule was ignored '
'at %s:%s.', pseudo_class, rule.source_line,
rule.source_column)
continue
else:
LOGGER.warning(
'Unsupported @page selector "%s", '
'the whole @page rule was ignored at %s:%s.',
tinycss2.serialize(rule.prelude), rule.source_line,
rule.source_column)
continue
content = tinycss2.parse_declaration_list(rule.content)
declarations = list(preprocess_declarations(base_url, content))
# Use a double lambda to have a closure that holds page_types
match = (lambda page_types: lambda _document: page_types)(
PAGE_PSEUDOCLASS_TARGETS[pseudo_class])
if declarations:
selector_list = [Selector(specificity, None, match)]
rules.append((rule, selector_list, declarations))
for margin_rule in content:
if margin_rule.type != 'at-rule':
continue
declarations = list(
preprocess_declarations(
base_url,
tinycss2.parse_declaration_list(margin_rule.content)))
if declarations:
selector_list = [
Selector(specificity, '@' + margin_rule.at_keyword,
match)
]
rules.append((margin_rule, selector_list, declarations))
elif rule.type == 'at-rule' and rule.at_keyword == 'font-face':
content = tinycss2.parse_declaration_list(rule.content)
rule_descriptors = dict(preprocess_descriptors(base_url, content))
for key in ('src', 'font_family'):
if key not in rule_descriptors:
LOGGER.warning(
"Missing %s descriptor in '@font-face' rule at %s:%s",
key.replace('_', '-'), rule.source_line,
rule.source_column)
break
else:
if font_config is not None:
font_filename = font_config.add_font_face(
rule_descriptors, url_fetcher)
if font_filename:
fonts.append(font_filename)
