From 99f62bcb23cee6121475c702f36ed34d54602bbd Mon Sep 17 00:00:00 2001 From: barneygale Date: Thu, 2 Feb 2023 22:37:21 +0000 Subject: [PATCH 1/5] GH-78079: Fix UNC device path root normalization in pathlib We no longer add a root to device paths such as `//./PhysicalDrive0`, `//?/BootPartition` and `//./c:` while normalizing. We also avoid adding a root to incomplete UNC share paths, like `//`, `//a` and `//a/`. --- Lib/pathlib.py | 11 ++++++++--- Lib/test/test_ntpath.py | 8 ++++++++ Lib/test/test_pathlib.py | 11 +++++++++++ .../2023-02-17-21-14-40.gh-issue-78079.z3Szr6.rst | 3 +++ 4 files changed, 30 insertions(+), 3 deletions(-) create mode 100644 Misc/NEWS.d/next/Library/2023-02-17-21-14-40.gh-issue-78079.z3Szr6.rst diff --git a/Lib/pathlib.py b/Lib/pathlib.py index dde573592fddce..b05b461c6719d8 100644 --- a/Lib/pathlib.py +++ b/Lib/pathlib.py @@ -281,9 +281,14 @@ def _parse_parts(cls, parts): if altsep: path = path.replace(altsep, sep) drv, root, rel = cls._flavour.splitroot(path) - if drv.startswith(sep): - # pathlib assumes that UNC paths always have a root. - root = sep + if drv and not root and not drv.endswith(sep): + drv_parts = drv.split(sep) + if len(drv_parts) == 4 and drv_parts[2] not in '?.': + # e.g. //server/share + root = sep + elif len(drv_parts) == 6: + # e.g. //?/unc/server/share + root = sep unfiltered_parsed = [drv + root] + rel.split(sep) parsed = [sys.intern(x) for x in unfiltered_parsed if x and x != '.'] return drv, root, parsed diff --git a/Lib/test/test_ntpath.py b/Lib/test/test_ntpath.py index 08c8a7a1f94b95..c304123812824f 100644 --- a/Lib/test/test_ntpath.py +++ b/Lib/test/test_ntpath.py @@ -168,6 +168,7 @@ def test_splitroot(self): # gh-81790: support device namespace, including UNC drives. tester('ntpath.splitroot("//?/c:")', ("//?/c:", "", "")) + tester('ntpath.splitroot("//./c:")', ("//./c:", "", "")) tester('ntpath.splitroot("//?/c:/")', ("//?/c:", "/", "")) tester('ntpath.splitroot("//?/c:/dir")', ("//?/c:", "/", "dir")) tester('ntpath.splitroot("//?/UNC")', ("//?/UNC", "", "")) @@ -178,8 +179,12 @@ def test_splitroot(self): tester('ntpath.splitroot("//?/VOLUME{00000000-0000-0000-0000-000000000000}/spam")', ('//?/VOLUME{00000000-0000-0000-0000-000000000000}', '/', 'spam')) tester('ntpath.splitroot("//?/BootPartition/")', ("//?/BootPartition", "/", "")) + tester('ntpath.splitroot("//./BootPartition/")', ("//./BootPartition", "/", "")) + tester('ntpath.splitroot("//./PhysicalDrive0")', ("//./PhysicalDrive0", "", "")) + tester('ntpath.splitroot("//./nul")', ("//./nul", "", "")) tester('ntpath.splitroot("\\\\?\\c:")', ("\\\\?\\c:", "", "")) + tester('ntpath.splitroot("\\\\.\\c:")', ("\\\\.\\c:", "", "")) tester('ntpath.splitroot("\\\\?\\c:\\")', ("\\\\?\\c:", "\\", "")) tester('ntpath.splitroot("\\\\?\\c:\\dir")', ("\\\\?\\c:", "\\", "dir")) tester('ntpath.splitroot("\\\\?\\UNC")', ("\\\\?\\UNC", "", "")) @@ -192,6 +197,9 @@ def test_splitroot(self): tester('ntpath.splitroot("\\\\?\\VOLUME{00000000-0000-0000-0000-000000000000}\\spam")', ('\\\\?\\VOLUME{00000000-0000-0000-0000-000000000000}', '\\', 'spam')) tester('ntpath.splitroot("\\\\?\\BootPartition\\")', ("\\\\?\\BootPartition", "\\", "")) + tester('ntpath.splitroot("\\\\.\\BootPartition\\")', ("\\\\.\\BootPartition", "\\", "")) + tester('ntpath.splitroot("\\\\.\\PhysicalDrive0")', ("\\\\.\\PhysicalDrive0", "", "")) + tester('ntpath.splitroot("\\\\.\\nul")', ("\\\\.\\nul", "", "")) # gh-96290: support partial/invalid UNC drives tester('ntpath.splitroot("//")', ("//", "", "")) # empty server & missing share diff --git a/Lib/test/test_pathlib.py b/Lib/test/test_pathlib.py index 4de91d52c6d10c..2455173e94d4ea 100644 --- a/Lib/test/test_pathlib.py +++ b/Lib/test/test_pathlib.py @@ -95,6 +95,9 @@ def test_parse_parts(self): check(['c:/a'], ('c:', '\\', ['c:\\', 'a'])) check(['/a'], ('', '\\', ['\\', 'a'])) # UNC paths. + check(['//'], ('\\\\', '', ['\\\\'])) + check(['//a'], ('\\\\a', '', ['\\\\a'])) + check(['//a/'], ('\\\\a\\', '', ['\\\\a\\'])) check(['//a/b'], ('\\\\a\\b', '\\', ['\\\\a\\b\\'])) check(['//a/b/'], ('\\\\a\\b', '\\', ['\\\\a\\b\\'])) check(['//a/b/c'], ('\\\\a\\b', '\\', ['\\\\a\\b\\', 'c'])) @@ -108,12 +111,20 @@ def test_parse_parts(self): # UNC paths. check(['a', '//b/c//', 'd'], ('\\\\b\\c', '\\', ['\\\\b\\c\\', 'd'])) # Extended paths. + check(['//./c:'], ('\\\\.\\c:', '', ['\\\\.\\c:'])) check(['//?/c:/'], ('\\\\?\\c:', '\\', ['\\\\?\\c:\\'])) check(['//?/c:/a'], ('\\\\?\\c:', '\\', ['\\\\?\\c:\\', 'a'])) check(['//?/c:/a', '/b'], ('\\\\?\\c:', '\\', ['\\\\?\\c:\\', 'b'])) # Extended UNC paths (format is "\\?\UNC\server\share"). + check(['//?/UNC/'], ('\\\\?\\UNC\\', '', ['\\\\?\\UNC\\'])) check(['//?/UNC/b/c'], ('\\\\?\\UNC\\b\\c', '\\', ['\\\\?\\UNC\\b\\c\\'])) check(['//?/UNC/b/c/d'], ('\\\\?\\UNC\\b\\c', '\\', ['\\\\?\\UNC\\b\\c\\', 'd'])) + # UNC device paths + check(['//./BootPartition/'], ('\\\\.\\BootPartition', '\\', ['\\\\.\\BootPartition\\'])) + check(['//?/BootPartition/'], ('\\\\?\\BootPartition', '\\', ['\\\\?\\BootPartition\\'])) + check(['//./PhysicalDrive0'], ('\\\\.\\PhysicalDrive0', '', ['\\\\.\\PhysicalDrive0'])) + check(['//?/Volume{}/'], ('\\\\?\\Volume{}', '\\', ['\\\\?\\Volume{}\\'])) + check(['//./nul'], ('\\\\.\\nul', '', ['\\\\.\\nul'])) # Second part has a root but not drive. check(['a', '/b', 'c'], ('', '\\', ['\\', 'b', 'c'])) check(['Z:/a', '/b', 'c'], ('Z:', '\\', ['Z:\\', 'b', 'c'])) diff --git a/Misc/NEWS.d/next/Library/2023-02-17-21-14-40.gh-issue-78079.z3Szr6.rst b/Misc/NEWS.d/next/Library/2023-02-17-21-14-40.gh-issue-78079.z3Szr6.rst new file mode 100644 index 00000000000000..bbb9ac3e3f8faa --- /dev/null +++ b/Misc/NEWS.d/next/Library/2023-02-17-21-14-40.gh-issue-78079.z3Szr6.rst @@ -0,0 +1,3 @@ +Fix incorrect normalization of UNC device path roots, and partial UNC share +path roots, in :class:`pathlib.PurePath`. Pathlib no longer appends a +trailing slash to such paths. From 6651b67bf7db05da0463677a8e8c84132d98b67a Mon Sep 17 00:00:00 2001 From: barneygale Date: Mon, 20 Feb 2023 01:20:14 +0000 Subject: [PATCH 2/5] A few more test cases --- Lib/test/test_pathlib.py | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/Lib/test/test_pathlib.py b/Lib/test/test_pathlib.py index 2455173e94d4ea..bedd9d1a167fe4 100644 --- a/Lib/test/test_pathlib.py +++ b/Lib/test/test_pathlib.py @@ -116,8 +116,14 @@ def test_parse_parts(self): check(['//?/c:/a'], ('\\\\?\\c:', '\\', ['\\\\?\\c:\\', 'a'])) check(['//?/c:/a', '/b'], ('\\\\?\\c:', '\\', ['\\\\?\\c:\\', 'b'])) # Extended UNC paths (format is "\\?\UNC\server\share"). + check(['//?'], ('\\\\?', '', ['\\\\?'])) + check(['//?/'], ('\\\\?\\', '', ['\\\\?\\'])) + check(['//?/UNC'], ('\\\\?\\UNC', '', ['\\\\?\\UNC'])) check(['//?/UNC/'], ('\\\\?\\UNC\\', '', ['\\\\?\\UNC\\'])) + check(['//?/UNC/b'], ('\\\\?\\UNC\\b', '', ['\\\\?\\UNC\\b'])) + check(['//?/UNC/b/'], ('\\\\?\\UNC\\b\\', '', ['\\\\?\\UNC\\b\\'])) check(['//?/UNC/b/c'], ('\\\\?\\UNC\\b\\c', '\\', ['\\\\?\\UNC\\b\\c\\'])) + check(['//?/UNC/b/c/'], ('\\\\?\\UNC\\b\\c', '\\', ['\\\\?\\UNC\\b\\c\\'])) check(['//?/UNC/b/c/d'], ('\\\\?\\UNC\\b\\c', '\\', ['\\\\?\\UNC\\b\\c\\', 'd'])) # UNC device paths check(['//./BootPartition/'], ('\\\\.\\BootPartition', '\\', ['\\\\.\\BootPartition\\'])) From c444a3cbd4d13f97e05cca2913ab53e0f55c0782 Mon Sep 17 00:00:00 2001 From: Barney Gale Date: Fri, 10 Mar 2023 19:33:48 +0000 Subject: [PATCH 3/5] Update Lib/pathlib.py Co-authored-by: Eryk Sun --- Lib/pathlib.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Lib/pathlib.py b/Lib/pathlib.py index b05b461c6719d8..dfa9da12408096 100644 --- a/Lib/pathlib.py +++ b/Lib/pathlib.py @@ -281,7 +281,7 @@ def _parse_parts(cls, parts): if altsep: path = path.replace(altsep, sep) drv, root, rel = cls._flavour.splitroot(path) - if drv and not root and not drv.endswith(sep): + if not root and drv.startswith(sep) and not drv.endswith(sep): drv_parts = drv.split(sep) if len(drv_parts) == 4 and drv_parts[2] not in '?.': # e.g. //server/share From 36623c0b3439089d7d706eeae565f928c0fb7b88 Mon Sep 17 00:00:00 2001 From: barneygale Date: Fri, 10 Mar 2023 22:10:55 +0000 Subject: [PATCH 4/5] Fix handling of UNC paths with incomplete drives in glob(), walk(), etc. --- Lib/pathlib.py | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/Lib/pathlib.py b/Lib/pathlib.py index dfa9da12408096..9e14e364d6aabc 100644 --- a/Lib/pathlib.py +++ b/Lib/pathlib.py @@ -718,8 +718,15 @@ def __new__(cls, *args, **kwargs): def _make_child_relpath(self, part): # This is an optimization used for dir walking. `part` must be # a single part relative to this path. - parts = self._parts + [part] - return self._from_parsed_parts(self._drv, self._root, parts) + drv = self._drv + root = self._root + parts = self._parts + if drv and not root and not parts: + sep = self._flavour.sep + if drv.startswith(sep) and not drv.endswith(sep): + # Incomplete UNC drive like '//foo'. + return self._from_parts([self, part]) + return self._from_parsed_parts(drv, root, parts + [part]) def __enter__(self): # In previous versions of pathlib, __exit__() marked this path as From aea236e5f429f86e041bb7b2808783e47909320e Mon Sep 17 00:00:00 2001 From: barneygale Date: Fri, 10 Mar 2023 22:17:32 +0000 Subject: [PATCH 5/5] Test handling of UNC paths with incomplete drives in joinpath() --- Lib/test/test_pathlib.py | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/Lib/test/test_pathlib.py b/Lib/test/test_pathlib.py index bedd9d1a167fe4..dc94fc9ba6fb62 100644 --- a/Lib/test/test_pathlib.py +++ b/Lib/test/test_pathlib.py @@ -1331,6 +1331,13 @@ def test_join(self): self.assertEqual(pp, P('C:/a/b/x/y')) pp = p.joinpath('c:/x/y') self.assertEqual(pp, P('C:/x/y')) + # Joining onto a UNC path with no root + pp = P('//').joinpath('server') + self.assertEqual(pp, P('//server')) + pp = P('//server').joinpath('share') + self.assertEqual(pp, P('//server/share')) + pp = P('//./BootPartition').joinpath('Windows') + self.assertEqual(pp, P('//./BootPartition/Windows')) def test_div(self): # Basically the same as joinpath().