Merged revisions 83818 via svnmerge from
svn+ssh://pythondev@svn.python.org/python/branches/py3k
........
r83818 | senthil.kumaran | 2010-08-08 16:57:53 +0530 (Sun, 08 Aug 2010) | 4 lines
Fix Issue8280 - urllib2's Request method will remove fragements in the url.
This is how it should work,wget and curl work like this way too. Old behavior was wrong.
........
diff --git a/Lib/test/test_urllib2.py b/Lib/test/test_urllib2.py
index 23631f6..8703081 100644
--- a/Lib/test/test_urllib2.py
+++ b/Lib/test/test_urllib2.py
@@ -1237,6 +1237,16 @@
self.assertEqual("www.python.org", self.get.get_origin_req_host())
self.assertEqual("www.perl.org", self.get.get_host())
+ def test_wrapped_url(self):
+ req = Request("<URL:http://www.python.org>")
+ self.assertEqual("www.python.org", req.get_host())
+
+ def test_urlwith_fragment(self):
+ req = Request("http://www.python.org/?qs=query#fragment=true")
+ self.assertEqual("/?qs=query", req.get_selector())
+ req = Request("http://www.python.org/#fun=true")
+ self.assertEqual("/", req.get_selector())
+
def test_main(verbose=None):
from test import test_urllib2
diff --git a/Lib/test/test_urllib2net.py b/Lib/test/test_urllib2net.py
index d2762dc..4da38ef 100644
--- a/Lib/test/test_urllib2net.py
+++ b/Lib/test/test_urllib2net.py
@@ -154,6 +154,13 @@
## self._test_urls(urls, self._extra_handlers()+[bauth, dauth])
+ def test_urlwithfrag(self):
+ urlwith_frag = "http://docs.python.org/glossary.html#glossary"
+ req = urllib2.Request(urlwith_frag)
+ res = urllib2.urlopen(req)
+ self.assertEqual(res.geturl(),
+ "http://docs.python.org/glossary.html")
+
def _test_urls(self, urls, handlers, retry=True):
import time
import logging
diff --git a/Lib/urllib2.py b/Lib/urllib2.py
index f6f7be9..d0e81a8 100644
--- a/Lib/urllib2.py
+++ b/Lib/urllib2.py
@@ -109,7 +109,7 @@
from StringIO import StringIO
from urllib import (unwrap, unquote, splittype, splithost, quote,
- addinfourl, splitport,
+ addinfourl, splitport, splittag,
splitattr, ftpwrapper, splituser, splitpasswd, splitvalue)
# support for FileHandler, proxies via environment variables
@@ -190,6 +190,7 @@
origin_req_host=None, unverifiable=False):
# unwrap('<URL:type://host/path>') --> 'type://host/path'
self.__original = unwrap(url)
+ self.__original, fragment = splittag(self.__original)
self.type = None
# self.__r_type is what's left after doing the splittype
self.host = None