Stowage/cpython
2
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2025-11-03 23:21:29 +00:00
Code Issues Projects Releases Packages Wiki Activity

Merged revisions 87542 via svnmerge from

Browse source 
svn+ssh://pythondev@svn.python.org/python/branches/py3k

........
  r87542 | senthil.kumaran | 2010-12-28 23:55:16 +0800 (Tue, 28 Dec 2010) | 3 lines

  Fix Issue10759 - html.parser.unescape() fails on HTML entities with incorrect syntax
........
This commit is contained in:
Senthil Kumaran 2010-12-28 16:10:56 +00:00
parent 18f6b1987f
commit 6c85838489
2 changed files with 14 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

3
Lib/html/parser.py
View file

@ -367,6 +367,7 @@ def unescape(self, s):
return s return s
def replaceEntities(s): def replaceEntities(s):
s = s.groups()[0] s = s.groups()[0]
try:
if s[0] == "#": if s[0] == "#":
s = s[1:] s = s[1:]
if s[0] in ['x','X']: if s[0] in ['x','X']:
@ -374,6 +375,8 @@ def replaceEntities(s):
else: else:
c = int(s) c = int(s)
return chr(c) return chr(c)
except ValueError:
return '&#'+ s +';'
else: else:
# Cannot use name2codepoint directly, because HTMLParser # Cannot use name2codepoint directly, because HTMLParser
# supports apos, which is not part of HTML 4 # supports apos, which is not part of HTML 4

4
Lib/test/test_htmlparser.py
View file

@ -319,6 +319,10 @@ def test_entityrefs_in_attributes(self):
self._run_check("<html foo='&euro;&amp;&#97;&#x61;&unsupported;'>", [ self._run_check("<html foo='&euro;&amp;&#97;&#x61;&unsupported;'>", [
("starttag", "html", [("foo", "\u20AC&aa&unsupported;")]) ("starttag", "html", [("foo", "\u20AC&aa&unsupported;")])
]) ])
def test_unescape_function(self):
p = html.parser.HTMLParser()
self.assertEqual(p.unescape('&#bad;'),'&#bad;')
self.assertEqual(p.unescape('&#0038;'),'&')
def test_main(): def test_main():