[Python-checkins] r86676 - in python/branches/py3k: Lib/test/test_urllib2.py Lib/urllib/parse.py Lib/urllib/request.py Misc/NEWS
senthil.kumaran
python-checkins at python.org
Mon Nov 22 05:48:26 CET 2010
Author: senthil.kumaran
Date: Mon Nov 22 05:48:26 2010
New Revision: 86676
Log:
Fix Issue4493 - urllib2 adds '/' to the path component of url, when it does not
starts with one. This behavior is exhibited by browser and other clients.
Modified:
python/branches/py3k/Lib/test/test_urllib2.py
python/branches/py3k/Lib/urllib/parse.py
python/branches/py3k/Lib/urllib/request.py
python/branches/py3k/Misc/NEWS
Modified: python/branches/py3k/Lib/test/test_urllib2.py
==============================================================================
--- python/branches/py3k/Lib/test/test_urllib2.py (original)
+++ python/branches/py3k/Lib/test/test_urllib2.py Mon Nov 22 05:48:26 2010
@@ -848,6 +848,25 @@
p_ds_req = h.do_request_(ds_req)
self.assertEqual(p_ds_req.unredirected_hdrs["Host"],"example.com")
+ def test_fixpath_in_weirdurls(self):
+ # Issue4493: urllib2 to supply '/' when to urls where path does not
+ # start with'/'
+
+ h = urllib.request.AbstractHTTPHandler()
+ o = h.parent = MockOpener()
+
+ weird_url = 'http://www.python.org?getspam'
+ req = Request(weird_url)
+ newreq = h.do_request_(req)
+ self.assertEqual(newreq.host,'www.python.org')
+ self.assertEqual(newreq.selector,'/?getspam')
+
+ url_without_path = 'http://www.python.org'
+ req = Request(url_without_path)
+ newreq = h.do_request_(req)
+ self.assertEqual(newreq.host,'www.python.org')
+ self.assertEqual(newreq.selector,'')
+
def test_errors(self):
h = urllib.request.HTTPErrorProcessor()
Modified: python/branches/py3k/Lib/urllib/parse.py
==============================================================================
--- python/branches/py3k/Lib/urllib/parse.py (original)
+++ python/branches/py3k/Lib/urllib/parse.py Mon Nov 22 05:48:26 2010
@@ -699,7 +699,12 @@
_hostprog = re.compile('^//([^/?]*)(.*)$')
match = _hostprog.match(url)
- if match: return match.group(1, 2)
+ if match:
+ host_port = match.group(1)
+ path = match.group(2)
+ if path and not path.startswith('/'):
+ path = '/' + path
+ return host_port, path
return None, url
_userprog = None
Modified: python/branches/py3k/Lib/urllib/request.py
==============================================================================
--- python/branches/py3k/Lib/urllib/request.py (original)
+++ python/branches/py3k/Lib/urllib/request.py Mon Nov 22 05:48:26 2010
@@ -105,7 +105,7 @@
# check for SSL
try:
import ssl
-except:
+except ImportError:
_have_ssl = False
else:
_have_ssl = True
Modified: python/branches/py3k/Misc/NEWS
==============================================================================
--- python/branches/py3k/Misc/NEWS (original)
+++ python/branches/py3k/Misc/NEWS Mon Nov 22 05:48:26 2010
@@ -32,6 +32,9 @@
Library
-------
+- Issue #4493: urllib2 adds '/' in front of path components which does not
+ start with '/. Common behavior exhibited by browsers and other clients.
+
- Issue #6378: idle.bat now runs with the appropriate Python version rather than
the system default. Patch by Sridhar Ratnakumar.
More information about the Python-checkins
mailing list