From 4c88db77a057ebe0a5a5496152d2099cf760e837 Mon Sep 17 00:00:00 2001 From: Senthil Kumaran Date: Sun, 8 Aug 2010 11:30:58 +0000 Subject: 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. ........ --- Lib/test/test_urllib2.py | 10 ++++++++++ Lib/test/test_urllib2net.py | 7 +++++++ Lib/urllib/request.py | 3 ++- 3 files changed, 19 insertions(+), 1 deletion(-) diff --git a/Lib/test/test_urllib2.py b/Lib/test/test_urllib2.py index 9d1ed9b..83bd467 100644 --- a/Lib/test/test_urllib2.py +++ b/Lib/test/test_urllib2.py @@ -1250,6 +1250,16 @@ class RequestTests(unittest.TestCase): 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("") + 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 0f109dc..eab9306 100644 --- a/Lib/test/test_urllib2net.py +++ b/Lib/test/test_urllib2net.py @@ -152,6 +152,13 @@ class OtherNetworkTests(unittest.TestCase): ## self._test_urls(urls, self._extra_handlers()+[bauth, dauth]) + def test_urlwithfrag(self): + urlwith_frag = "http://docs.python.org/glossary.html#glossary" + req = urllib.request.Request(urlwith_frag) + res = urllib.request.urlopen(req) + self.assertEqual(res.geturl(), + "http://docs.python.org/glossary.html") + def _test_urls(self, urls, handlers, retry=True): import socket import time diff --git a/Lib/urllib/request.py b/Lib/urllib/request.py index a26b763..3ea38f2 100644 --- a/Lib/urllib/request.py +++ b/Lib/urllib/request.py @@ -99,7 +99,7 @@ from urllib.error import URLError, HTTPError, ContentTooShortError from urllib.parse import ( urlparse, urlsplit, urljoin, unwrap, quote, unquote, splittype, splithost, splitport, splituser, splitpasswd, - splitattr, splitquery, splitvalue, to_bytes, urlunparse) + splitattr, splitquery, splitvalue, splittag, to_bytes, urlunparse) from urllib.response import addinfourl, addclosehook # check for SSL @@ -163,6 +163,7 @@ class Request: origin_req_host=None, unverifiable=False): # unwrap('') --> 'type://host/path' self.full_url = unwrap(url) + self.full_url, fragment = splittag(self.full_url) self.data = data self.headers = {} self._tunnel_host = None -- cgit v0.12