1
0
mirror of https://github.com/django/django.git synced 2025-03-03 05:24:24 +00:00

Fixed our HTMLParser patches for python 2.7.4

This commit is contained in:
Florian Apolloner 2012-06-16 20:49:50 +02:00
parent a54a8bab0c
commit 6bc1b22299

View File

@ -1,6 +1,7 @@
import HTMLParser as _HTMLParser import HTMLParser as _HTMLParser
import re import re
tagfind = re.compile('([a-zA-Z][-.a-zA-Z0-9:_]*)(?:\s|/(?!>))*')
class HTMLParser(_HTMLParser.HTMLParser): class HTMLParser(_HTMLParser.HTMLParser):
""" """
@ -33,10 +34,10 @@ class HTMLParser(_HTMLParser.HTMLParser):
# Now parse the data between i+1 and j into a tag and attrs # Now parse the data between i+1 and j into a tag and attrs
attrs = [] attrs = []
match = _HTMLParser.tagfind.match(rawdata, i + 1) match = tagfind.match(rawdata, i + 1)
assert match, 'unexpected call to parse_starttag()' assert match, 'unexpected call to parse_starttag()'
k = match.end() k = match.end()
self.lasttag = tag = rawdata[i + 1:k].lower() self.lasttag = tag = match.group(1).lower()
while k < endpos: while k < endpos:
m = _HTMLParser.attrfind.match(rawdata, k) m = _HTMLParser.attrfind.match(rawdata, k)
@ -48,6 +49,7 @@ class HTMLParser(_HTMLParser.HTMLParser):
elif attrvalue[:1] == '\'' == attrvalue[-1:] or \ elif attrvalue[:1] == '\'' == attrvalue[-1:] or \
attrvalue[:1] == '"' == attrvalue[-1:]: attrvalue[:1] == '"' == attrvalue[-1:]:
attrvalue = attrvalue[1:-1] attrvalue = attrvalue[1:-1]
if attrvalue:
attrvalue = self.unescape(attrvalue) attrvalue = self.unescape(attrvalue)
attrs.append((attrname.lower(), attrvalue)) attrs.append((attrname.lower(), attrvalue))
k = m.end() k = m.end()