mirrors/cpython
1
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2025-08-04 00:48:58 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 8

Issue #22118: Switch urllib.parse to use RFC 3986 semantics for the resolution of relative URLs, rather than RFCs 1808 and 2396.

Browse source 
Patch by Demian Brecht.
This commit is contained in:
Antoine Pitrou 2014-08-21 19:16:17 -04:00
parent a7eb746278
commit 55ac5b3f7b
4 changed files with 71 additions and 41 deletions
Download patch file Download diff file Expand all files Collapse all files

40
Lib/test/test_urlparse.py
View file

@ -211,10 +211,6 @@ class UrlParseTestCase(unittest.TestCase):
# "abnormal" cases from RFC 1808:
self.checkJoin(RFC1808_BASE, '', 'http://a/b/c/d;p?q#f')
self.checkJoin(RFC1808_BASE, '../../../g', 'http://a/../g')
self.checkJoin(RFC1808_BASE, '../../../../g', 'http://a/../../g')
self.checkJoin(RFC1808_BASE, '/./g', 'http://a/./g')
self.checkJoin(RFC1808_BASE, '/../g', 'http://a/../g')
self.checkJoin(RFC1808_BASE, 'g.', 'http://a/b/c/g.')
self.checkJoin(RFC1808_BASE, '.g', 'http://a/b/c/.g')
self.checkJoin(RFC1808_BASE, 'g..', 'http://a/b/c/g..')
@ -229,6 +225,13 @@ class UrlParseTestCase(unittest.TestCase):
#self.checkJoin(RFC1808_BASE, 'http:g', 'http:g')
#self.checkJoin(RFC1808_BASE, 'http:', 'http:')
# XXX: The following tests are no longer compatible with RFC3986
# self.checkJoin(RFC1808_BASE, '../../../g', 'http://a/../g')
# self.checkJoin(RFC1808_BASE, '../../../../g', 'http://a/../../g')
# self.checkJoin(RFC1808_BASE, '/./g', 'http://a/./g')
# self.checkJoin(RFC1808_BASE, '/../g', 'http://a/../g')
def test_RFC2368(self):
# Issue 11467: path that starts with a number is not parsed correctly
self.assertEqual(urllib.parse.urlparse('mailto:1337@example.org'),
@ -259,10 +262,6 @@ class UrlParseTestCase(unittest.TestCase):
self.checkJoin(RFC2396_BASE, '../../', 'http://a/')
self.checkJoin(RFC2396_BASE, '../../g', 'http://a/g')
self.checkJoin(RFC2396_BASE, '', RFC2396_BASE)
self.checkJoin(RFC2396_BASE, '../../../g', 'http://a/../g')
self.checkJoin(RFC2396_BASE, '../../../../g', 'http://a/../../g')
self.checkJoin(RFC2396_BASE, '/./g', 'http://a/./g')
self.checkJoin(RFC2396_BASE, '/../g', 'http://a/../g')
self.checkJoin(RFC2396_BASE, 'g.', 'http://a/b/c/g.')
self.checkJoin(RFC2396_BASE, '.g', 'http://a/b/c/.g')
self.checkJoin(RFC2396_BASE, 'g..', 'http://a/b/c/g..')
@ -278,10 +277,17 @@ class UrlParseTestCase(unittest.TestCase):
self.checkJoin(RFC2396_BASE, 'g#s/./x', 'http://a/b/c/g#s/./x')
self.checkJoin(RFC2396_BASE, 'g#s/../x', 'http://a/b/c/g#s/../x')
# XXX: The following tests are no longer compatible with RFC3986
# self.checkJoin(RFC2396_BASE, '../../../g', 'http://a/../g')
# self.checkJoin(RFC2396_BASE, '../../../../g', 'http://a/../../g')
# self.checkJoin(RFC2396_BASE, '/./g', 'http://a/./g')
# self.checkJoin(RFC2396_BASE, '/../g', 'http://a/../g')
def test_RFC3986(self):
# Test cases from RFC3986
self.checkJoin(RFC3986_BASE, '?y','http://a/b/c/d;p?y')
self.checkJoin(RFC2396_BASE, ';x', 'http://a/b/c/;x')
self.checkJoin(RFC3986_BASE, ';x', 'http://a/b/c/;x')
self.checkJoin(RFC3986_BASE, 'g:h','g:h')
self.checkJoin(RFC3986_BASE, 'g','http://a/b/c/g')
self.checkJoin(RFC3986_BASE, './g','http://a/b/c/g')
@ -305,17 +311,17 @@ class UrlParseTestCase(unittest.TestCase):
self.checkJoin(RFC3986_BASE, '../..','http://a/')
self.checkJoin(RFC3986_BASE, '../../','http://a/')
self.checkJoin(RFC3986_BASE, '../../g','http://a/g')
self.checkJoin(RFC3986_BASE, '../../../g', 'http://a/g')
#Abnormal Examples
# The 'abnormal scenarios' are incompatible with RFC2986 parsing
# Tests are here for reference.
#self.checkJoin(RFC3986_BASE, '../../../g','http://a/g')
#self.checkJoin(RFC3986_BASE, '../../../../g','http://a/g')
#self.checkJoin(RFC3986_BASE, '/./g','http://a/g')
#self.checkJoin(RFC3986_BASE, '/../g','http://a/g')
self.checkJoin(RFC3986_BASE, '../../../g','http://a/g')
self.checkJoin(RFC3986_BASE, '../../../../g','http://a/g')
self.checkJoin(RFC3986_BASE, '/./g','http://a/g')
self.checkJoin(RFC3986_BASE, '/../g','http://a/g')
self.checkJoin(RFC3986_BASE, 'g.','http://a/b/c/g.')
self.checkJoin(RFC3986_BASE, '.g','http://a/b/c/.g')
self.checkJoin(RFC3986_BASE, 'g..','http://a/b/c/g..')
@ -355,10 +361,8 @@ class UrlParseTestCase(unittest.TestCase):
self.checkJoin(SIMPLE_BASE, '../g','http://a/b/g')
self.checkJoin(SIMPLE_BASE, '../..','http://a/')
self.checkJoin(SIMPLE_BASE, '../../g','http://a/g')
self.checkJoin(SIMPLE_BASE, '../../../g','http://a/../g')
self.checkJoin(SIMPLE_BASE, './../g','http://a/b/g')
self.checkJoin(SIMPLE_BASE, './g/.','http://a/b/c/g/')
self.checkJoin(SIMPLE_BASE, '/./g','http://a/./g')
self.checkJoin(SIMPLE_BASE, 'g/./h','http://a/b/c/g/h')
self.checkJoin(SIMPLE_BASE, 'g/../h','http://a/b/c/h')
self.checkJoin(SIMPLE_BASE, 'http:g','http://a/b/c/g')
@ -372,6 +376,10 @@ class UrlParseTestCase(unittest.TestCase):
self.checkJoin('svn://pathtorepo/dir1', 'dir2', 'svn://pathtorepo/dir2')
self.checkJoin('svn+ssh://pathtorepo/dir1', 'dir2', 'svn+ssh://pathtorepo/dir2')
# XXX: The following tests are no longer compatible with RFC3986
# self.checkJoin(SIMPLE_BASE, '../../../g','http://a/../g')
# self.checkJoin(SIMPLE_BASE, '/./g','http://a/./g')
def test_RFC2732(self):
str_cases = [
('http://Test.python.org:5432/foo/', 'test.python.org', 5432),