Issue #20270: urllib and urlparse now support empty ports.

diff --git a/Lib/test/test_urllib.py b/Lib/test/test_urllib.py
index 3018ab198b4aeec36de2824247e2f021f5f9eb06..fa88b61a11647c4f754a063652aaca7b5e83761b 100644 (file)
--- a/Lib/test/test_urllib.py
+++ b/Lib/test/test_urllib.py
@@ -789,6 +789,26 @@ class Utility_Tests(unittest.TestCase):
         self.assertEqual(('user 2', 'ab'),urllib.splitpasswd('user 2:ab'))
         self.assertEqual(('user+1', 'a+b'),urllib.splitpasswd('user+1:a+b'))
 
+    def test_splitport(self):
+        splitport = urllib.splitport
+        self.assertEqual(splitport('parrot:88'), ('parrot', '88'))
+        self.assertEqual(splitport('parrot'), ('parrot', None))
+        self.assertEqual(splitport('parrot:'), ('parrot', None))
+        self.assertEqual(splitport('127.0.0.1'), ('127.0.0.1', None))
+        self.assertEqual(splitport('parrot:cheese'), ('parrot:cheese', None))
+
+    def test_splitnport(self):
+        splitnport = urllib.splitnport
+        self.assertEqual(splitnport('parrot:88'), ('parrot', 88))
+        self.assertEqual(splitnport('parrot'), ('parrot', -1))
+        self.assertEqual(splitnport('parrot', 55), ('parrot', 55))
+        self.assertEqual(splitnport('parrot:'), ('parrot', -1))
+        self.assertEqual(splitnport('parrot:', 55), ('parrot', 55))
+        self.assertEqual(splitnport('127.0.0.1'), ('127.0.0.1', -1))
+        self.assertEqual(splitnport('127.0.0.1', 55), ('127.0.0.1', 55))
+        self.assertEqual(splitnport('parrot:cheese'), ('parrot', None))
+        self.assertEqual(splitnport('parrot:cheese', 55), ('parrot', None))
+
 
 class URLopener_Tests(unittest.TestCase):
     """Testcase to test the open method of URLopener class."""

diff --git a/Lib/test/test_urlparse.py b/Lib/test/test_urlparse.py
index 13dd297c49bee64f30c7ddcc27d3f11c002d4524..b3ad7cd513bbd8bcf3af7105dba1f8d7eb942f3a 100644 (file)
--- a/Lib/test/test_urlparse.py
+++ b/Lib/test/test_urlparse.py
@@ -362,6 +362,16 @@ class UrlParseTestCase(unittest.TestCase):
             ('http://[::12.34.56.78]/foo/', '::12.34.56.78', None),
             ('http://[::ffff:12.34.56.78]/foo/',
              '::ffff:12.34.56.78', None),
+            ('http://Test.python.org:/foo/', 'test.python.org', None),
+            ('http://12.34.56.78:/foo/', '12.34.56.78', None),
+            ('http://[::1]:/foo/', '::1', None),
+            ('http://[dead:beef::1]:/foo/', 'dead:beef::1', None),
+            ('http://[dead:beef::]:/foo/', 'dead:beef::', None),
+            ('http://[dead:beef:cafe:5417:affe:8FA3:deaf:feed]:/foo/',
+             'dead:beef:cafe:5417:affe:8fa3:deaf:feed', None),
+            ('http://[::12.34.56.78]:/foo/', '::12.34.56.78', None),
+            ('http://[::ffff:12.34.56.78]:/foo/',
+             '::ffff:12.34.56.78', None),
             ]:
             urlparsed = urlparse.urlparse(url)
             self.assertEqual((urlparsed.hostname, urlparsed.port) , (hostname, port))

diff --git a/Lib/urllib.py b/Lib/urllib.py
index 244cb05374d1b49f4251b03afd589725542112f1..ac5d797e33a0a8bce806a58217819a282352359b 100644 (file)
--- a/Lib/urllib.py
+++ b/Lib/urllib.py
@@ -1125,10 +1125,13 @@ def splitport(host):
     global _portprog
     if _portprog is None:
         import re
-        _portprog = re.compile('^(.*):([0-9]+)$')
+        _portprog = re.compile('^(.*):([0-9]*)$')
 
     match = _portprog.match(host)
-    if match: return match.group(1, 2)
+    if match:
+        host, port = match.groups()
+        if port:
+            return host, port
     return host, None
 
 _nportprog = None
@@ -1145,12 +1148,12 @@ def splitnport(host, defport=-1):
     match = _nportprog.match(host)
     if match:
         host, port = match.group(1, 2)
-        try:
-            if not port: raise ValueError, "no digits"
-            nport = int(port)
-        except ValueError:
-            nport = None
-        return host, nport
+        if port:
+            try:
+                nport = int(port)
+            except ValueError:
+                nport = None
+            return host, nport
     return host, defport
 
 _queryprog = None

diff --git a/Lib/urlparse.py b/Lib/urlparse.py
index 4ce982e8fdd4e52e403a41520ec566f17f0d1f5a..4cd3d6743a960f9e88f1de18ded659f6a84cc04d 100644 (file)
--- a/Lib/urlparse.py
+++ b/Lib/urlparse.py
@@ -109,10 +109,11 @@ class ResultMixin(object):
         netloc = self.netloc.split('@')[-1].split(']')[-1]
         if ':' in netloc:
             port = netloc.split(':')[1]
-            port = int(port, 10)
-            # verify legal port
-            if (0 <= port <= 65535):
-                return port
+            if port:
+                port = int(port, 10)
+                # verify legal port
+                if (0 <= port <= 65535):
+                    return port
         return None
 
 from collections import namedtuple

diff --git a/Misc/NEWS b/Misc/NEWS
index abafed787b5ea53d8a9677487b9ce87d534ca09a..40a9c5d190e524fe89075694665fa60842c774fc 100644 (file)
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -35,6 +35,8 @@ Core and Builtins
 Library
 -------
 
+- Issue #20270: urllib and urlparse now support empty ports.
+
 - Issue #20243: TarFile no longer raise ReadError when opened in write mode.
 
 - Issue #20245: The open functions in the tarfile module now correctly handle