Mirror
/
cpython
mirror of https://github.com/python/cpython
4
1
Fork 0
Code Issues Releases Wiki Activity

Merged revisions 83209 via svnmerge from 

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

........
  r83209 | senthil.kumaran | 2010-07-28 21:57:56 +0530 (Wed, 28 Jul 2010) | 3 lines

  Fix Issue6325 - robotparse to honor urls with query strings.
........
This commit is contained in:
Senthil Kumaran 2010-07-28 16:30:46 +00:00
parent b8f96c1603
commit 42b0c2f740
2 changed files with 15 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
Lib/test/test_robotparser.py
View File

@ -205,6 +205,17 @@ bad = ['/folder1/anotherfile.html']
RobotTest(13, doc, good, bad, agent="googlebot") RobotTest(13, doc, good, bad, agent="googlebot")
# 14. For issue #6325 (query string support)
doc = """
User-agent: *
Disallow: /some/path?name=value
"""
good = ['/some/path']
bad = ['/some/path?name=value']
RobotTest(14, doc, good, bad)
class NetworkTestCase(unittest.TestCase): class NetworkTestCase(unittest.TestCase):

6
Lib/urllib/robotparser.py
View File

@ -129,8 +129,10 @@ class RobotFileParser:
return True return True
# search for given user agent matches # search for given user agent matches
# the first match counts # the first match counts
url = urllib.parse.quote( parsed_url = urllib.parse.urlparse(urllib.parse.unquote(url))
urllib.parse.urlparse(urllib.parse.unquote(url))[2]) url = urllib.parse.urlunparse(('','',parsed_url.path,
parsed_url.params,parsed_url.query, parsed_url.fragment))
url = urllib.parse.quote(url)
if not url: if not url:
url = "/" url = "/"
for entry in self.entries: for entry in self.entries: