Kozea/WeasyPrint
1
1
Fork 0
mirror of https://github.com/Kozea/WeasyPrint.git synced 2024-10-04 07:57:52 +03:00
Code Issues Projects Releases Wiki Activity

Add support for data: URL scheme stylesheets.

Browse Source
This commit is contained in:
Simon Sapin 2011-08-16 15:11:35 +02:00
parent fc1bb2dd29
commit f6940496f1
6 changed files with 35 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
weasy/css/__init__.py
View File

@ -83,6 +83,7 @@ def find_stylesheets(document):
Find and parse stylesheets in a Document object. Return an iterable of
stylesheets, in tree order.
"""
parser = document.css_parser
for element in document.dom.iter():
if element.tag not in ('style', 'link'):
continue
@ -102,15 +103,17 @@ def find_stylesheets(document):
content = ''.join(content)
# lxml should give us either unicode or ASCII-only bytestrings, so
# we don't need `encoding` here.
yield parseString(content, href=element.base_url,
media=media_attr, title=element.get('title'))
yield parser.parseString(content,
href=element.base_url,
media=media_attr, title=element.get('title'))
elif element.tag == 'link' and element.get('href') \
and ' stylesheet ' in ' %s ' % element.get('rel', ''):
# URLs should NOT have been made absolute earlier
# TODO: support the <base> HTML element, but do not use
# lxml.html.HtmlElement.make_links_absolute() that changes the tree
href = get_url_attribute(element, 'href')
yield parseUrl(href, media=media_attr, title=element.get('title'))
yield parser.parseUrl(href,
media=media_attr, title=element.get('title'))
def find_style_attributes(document):

5
weasy/document.py
View File

@ -21,7 +21,7 @@ import os.path
import math
import logging
from cssutils import parseFile
from cssutils import parseFile, CSSParser
import lxml.html
import cairo
@ -54,6 +54,9 @@ class Document(object):
# Go through the property setter which calls ensure_url()
self.base_url = self.base_url
self.css_parser = CSSParser()
self.css_parser.setFetcher(utils.urllib_fetcher)
self.user_stylesheets = user_stylesheets or []
self.user_agent_stylesheets = user_agent_stylesheets or []

5
weasy/tests/resources/doc1.html
View File

@ -2,10 +2,11 @@
<head>
<meta http-equiv=Content-Type content=text/html;charset=utf8>
<link rel=stylesheet href="sheet1.css" media=screen>
<!-- currentColor means 'inherit' on color itself. -->
<link rel=stylesheet href="data:text/css,a%7Bcolor%3AcurrentColor%7D">
<style media=print>
@import url(sheet1.css);
li { color: red; }
a { color: currentColor; /* means 'inherit' on color itself. */ }
@import url(data:text/css,li%7Bcolor%3Ared%7D);
a:after {
content: " [" attr(href) "]"
}

9
weasy/tests/test_css.py
View File

@ -56,17 +56,18 @@ def test_find_stylesheets():
document = parse_html('doc1.html')
sheets = list(css.find_stylesheets(document))
assert len(sheets) == 2
assert len(sheets) == 3
# Also test that stylesheets are in tree order
assert [s.href.rsplit('/', 1)[-1] for s in sheets] \
== ['sheet1.css', 'doc1.html']
assert [s.href.rsplit('/', 1)[-1].rsplit(',', 1)[-1] for s in sheets] \
== ['sheet1.css', 'a%7Bcolor%3AcurrentColor%7D',
'doc1.html']
rules = list(rule for sheet in sheets
for rule in css.effective_rules(sheet, 'print'))
assert len(rules) == 9
# Also test appearance order
assert [rule.selectorText for rule in rules] \
== ['li', 'p', 'ul', 'li', 'a', 'a:after', ':first', 'ul',
== ['a', 'li', 'p', 'ul', 'li', 'a:after', ':first', 'ul',
'body > h1:first-child']

2
weasy/tests/test_utils.py
View File

@ -31,7 +31,7 @@ def test_multifunction():
@utils.MultiFunction
def foo():
"Hello, docstrings!"
assert foo.__name__ == 'foo'
assert foo.__doc__ == 'Hello, docstrings!'
assert foo.__module__ == __name__

16
weasy/utils.py
View File

@ -16,6 +16,8 @@
# You should have received a copy of the GNU Affero General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
import urllib
try:
from urlparse import urljoin, urlparse
except ImportError:
@ -72,3 +74,17 @@ def ensure_url(filename_or_url):
return filename_or_url
else:
return path2url(filename_or_url)
def urllib_fetcher(url):
"""
A "fetcher" for cssutils, based on urllib instead of urllib2, since
urllib has support for the "data" URL scheme.
"""
result = urllib.urlopen(url)
info = result.info()
if info.gettype() != 'text/css':
# TODO: warn
return None
charset = info.getparam('charset')
return charset, result.read()