GH-126601: pathname2url(): handle NTFS alternate data streams

Adjust `pathname2url()` to encode embedded colon characters in Windows
paths, rather than bailing out with an `OSError`.

Author: barneygale

Lib/nturl2path.py

@@ -41,6 +41,7 @@ def pathname2url(p):
     #   C:\foo\bar\spam.foo
     # becomes
     #   ///C:/foo/bar/spam.foo
+    import ntpath
     import urllib.parse
     # First, clean up some special forms. We are going to sacrifice
     # the additional information anyway
@@ -49,16 +50,10 @@ def pathname2url(p):
         p = p[4:]
         if p[:4].upper() == 'UNC/':
             p = '//' + p[4:]
-        elif p[1:2] != ':':
-            raise OSError('Bad path: ' + p)
-    if not ':' in p:
+    drive, tail = ntpath.splitdrive(p)
+    if drive[1:2] != ':':
         # No DOS drive specified, just quote the pathname
         return urllib.parse.quote(p)
-    comp = p.split(':', maxsplit=2)
-    if len(comp) != 2 or len(comp[0]) > 1:
-        error = 'Bad path: ' + p
-        raise OSError(error)
-
-    drive = urllib.parse.quote(comp[0].upper())
-    tail = urllib.parse.quote(comp[1])
+    drive = urllib.parse.quote(drive[0].upper())
+    tail = urllib.parse.quote(tail)
     return '///' + drive + ':' + tail

Lib/test/test_urllib.py

@@ -1533,8 +1533,9 @@ def test_pathname2url_win(self):
         self.assertEqual(fn('C:\\a\\b%#c'), '///C:/a/b%25%23c')
         self.assertEqual(fn('C:\\a\\b\xe9'), '///C:/a/b%C3%A9')
         self.assertEqual(fn('C:\\foo\\bar\\spam.foo'), "///C:/foo/bar/spam.foo")
-        # Long drive letter
-        self.assertRaises(IOError, fn, "XX:\\")
+        # NTFS alternate data streams
+        self.assertEqual(fn('C:\\foo:bar'), '///C:/foo%3Abar')
+        self.assertEqual(fn('foo:bar'), 'foo%3Abar')
         # No drive letter
         self.assertEqual(fn("\\folder\\test\\"), '/folder/test/')
         self.assertEqual(fn("\\\\folder\\test\\"), '//folder/test/')

Misc/NEWS.d/next/Library/2024-11-12-20-05-09.gh-issue-126601.Nj7bA9.rst

@@ -0,0 +1,3 @@
+Fix issue where :func:`urllib.request.pathname2url` raised :exc:`OSError`
+when given a Windows path containing a colon character not following a
+drive letter, such as before an NTFS alternate data stream.