# HG changeset patch
# Parent 56f71f02206ebe3413caa53b614403617ac544ed
Issue #18828: Allow urljoin() to work with any scheme not in non_hierarchical

diff -r 56f71f02206e Doc/library/urllib.parse.rst
--- a/Doc/library/urllib.parse.rst	Tue Dec 16 18:17:18 2014 -0800
+++ b/Doc/library/urllib.parse.rst	Thu Mar 26 10:25:26 2015 +0000
@@ -270,7 +270,9 @@
 
    .. versionchanged:: 3.5
 
-      Behaviour updated to match the semantics defined in :rfc:`3986`.
+      Behaviour updated to match the semantics defined in :rfc:`3986`.  Now
+      also supports arbitrary schemes other than ``hdl``, ``mailto``,
+      ``news``, ``sip``, ``sips``, ``snews``, ``tel``, and ``telnet``.
 
 
 .. function:: urldefrag(url)
diff -r 56f71f02206e Lib/test/test_urlparse.py
--- a/Lib/test/test_urlparse.py	Tue Dec 16 18:17:18 2014 -0800
+++ b/Lib/test/test_urlparse.py	Thu Mar 26 10:25:26 2015 +0000
@@ -375,6 +375,10 @@
         self.checkJoin('', 'http://a/./g', 'http://a/./g')
         self.checkJoin('svn://pathtorepo/dir1', 'dir2', 'svn://pathtorepo/dir2')
         self.checkJoin('svn+ssh://pathtorepo/dir1', 'dir2', 'svn+ssh://pathtorepo/dir2')
+        self.checkJoin('x-arbitrary-scheme://netloc/a', '/b',
+            'x-arbitrary-scheme://netloc/b')
+        self.checkJoin('x-arbitrary-scheme://netloc/path', '?query',
+            'x-arbitrary-scheme://netloc/path?query')
 
         # XXX: The following tests are no longer compatible with RFC3986
         # self.checkJoin(SIMPLE_BASE, '../../../g','http://a/../g')
@@ -392,6 +396,35 @@
         self.checkJoin('http://a/b/c/d/e/', '../../f/g', 'http://a/b/c/f/g')
         self.checkJoin('http://a/b/', '../../f/g/', 'http://a/f/g/')
 
+        # Schemes previously listed as non-hierarchical, but joined anyway:
+        self.checkJoin('gopher://example/0path/old-file', 'new-file',
+            'gopher://example/0path/new-file')
+        self.checkJoin('wais://host/database/wtype/old-path', 'new-path',
+            'wais://host/database/wtype/new-path')
+        self.checkJoin('imap://joe@example.com/INBOX/;uid=20', ';UID=21',
+            'imap://joe@example.com/INBOX/;UID=21')
+
+    def test_urljoin_non_hier(self):
+        # Relative URL returned verbatim (without a scheme) for base schemes
+        # in non_hierarchical
+        tests = (
+            ('hdl:cnri.dlib/december95', 'cnri.dlib/november95'),
+            ('mailto:infobot@example.com', '?subject=current-issue'),
+            ('news:comp.lang.python', 'message@id'),
+            ('telnet://host/', '//other-host/'),
+            ('snews://server/comp.lang.python', 'message@id'),
+            ('sip:user@host', '?Subject=foo'),
+            ('sips:user@host', '?Subject=foo'),
+            ('tel:+1-212-555-0101', '+1-800-555-0191'),
+        )
+        schemes = set()
+        for base, relurl in tests:
+            schemes.add(urllib.parse.urlsplit(base).scheme)
+            with self.subTest((base, relurl)):
+                self.checkJoin(base, relurl, relurl)
+        # Ensure we tested each relevant scheme
+        self.assertCountEqual(urllib.parse.non_hierarchical, schemes)
+
     def test_RFC2732(self):
         str_cases = [
             ('http://Test.python.org:5432/foo/', 'test.python.org', 5432),
diff -r 56f71f02206e Lib/urllib/parse.py
--- a/Lib/urllib/parse.py	Tue Dec 16 18:17:18 2014 -0800
+++ b/Lib/urllib/parse.py	Thu Mar 26 10:25:26 2015 +0000
@@ -37,10 +37,6 @@
            "unquote", "unquote_plus", "unquote_to_bytes"]
 
 # A classification of schemes ('' means apply by default)
-uses_relative = ['ftp', 'http', 'gopher', 'nntp', 'imap',
-                 'wais', 'file', 'https', 'shttp', 'mms',
-                 'prospero', 'rtsp', 'rtspu', '', 'sftp',
-                 'svn', 'svn+ssh']
 uses_netloc = ['ftp', 'http', 'gopher', 'nntp', 'telnet',
                'imap', 'wais', 'file', 'mms', 'https', 'shttp',
                'snews', 'prospero', 'rtsp', 'rtspu', 'rsync', '',
@@ -48,11 +44,15 @@
 uses_params = ['ftp', 'hdl', 'prospero', 'http', 'imap',
                'https', 'shttp', 'rtsp', 'rtspu', 'sip', 'sips',
                'mms', '', 'sftp', 'tel']
+non_hierarchical = ['hdl', 'mailto', 'news',
+                    'tel', 'telnet', 'snews', 'sip', 'sips']
 
 # These are not actually used anymore, but should stay for backwards
 # compatibility.  (They are undocumented, but have a public-looking name.)
-non_hierarchical = ['gopher', 'hdl', 'mailto', 'news',
-                    'telnet', 'wais', 'imap', 'snews', 'sip', 'sips']
+uses_relative = ['ftp', 'http', 'gopher', 'nntp', 'imap',
+                 'wais', 'file', 'https', 'shttp', 'mms',
+                 'prospero', 'rtsp', 'rtspu', '', 'sftp',
+                 'svn', 'svn+ssh']
 uses_query = ['http', 'wais', 'imap', 'https', 'shttp', 'mms',
               'gopher', 'rtsp', 'rtspu', 'sip', 'sips', '']
 uses_fragment = ['ftp', 'hdl', 'http', 'gopher', 'news',
@@ -416,13 +416,12 @@
     scheme, netloc, path, params, query, fragment = \
             urlparse(url, bscheme, allow_fragments)
 
-    if scheme != bscheme or scheme not in uses_relative:
+    if scheme != bscheme or scheme in non_hierarchical:
         return _coerce_result(url)
-    if scheme in uses_netloc:
-        if netloc:
-            return _coerce_result(urlunparse((scheme, netloc, path,
-                                              params, query, fragment)))
-        netloc = bnetloc
+    if netloc:
+        return _coerce_result(urlunparse((scheme, netloc, path,
+                                          params, query, fragment)))
+    netloc = bnetloc
 
     if not path and not params:
         path = bpath