gh-101000: Add os.path.splitroot()

python · AlexWaygood · Jan 27, 2023 · Jan 12, 2023 · Jan 12, 2023 · Jan 12, 2023
commit 21c0ba9069f496912f6cc6fc31f21055b772d1ea
diff --git a/Doc/library/os.path.rst b/Doc/library/os.path.rst
@@ -488,6 +488,26 @@ the :mod:`glob` module.)
       Accepts a :term:`path-like object`.
 
 
+.. function:: splitroot(path)
+
+   Split the pathname *path* into a triad ``(drive, root, tail)`` where:
+
+   1. *drive* is an optional mount point, exactly like :func:`splitdrive`;
+   2. *root* is an optional sequence of separators following the drive; and
+   3. *tail* is anything after the root.
+
+   On Posix, *drive* is always empty. The *root* may be empty (relative path),
+   a single forward slash (absolute path), or two forward slashes
+   (implementation-defined per the POSIX standard).
+
+   On Windows, *drive* may be a UNC sharepoint or a traditional DOS drive. The
+   *root* may be empty, a forward slash, or a backward slash.
+
+   In all cases, ``drive + root + tail`` will be the same as *path*.
+
+   .. versionadded:: 3.12
+
+
 .. function:: splitext(path)
 
    Split the pathname *path* into a pair ``(root, ext)``  such that ``root + ext ==

diff --git a/Lib/ntpath.py b/Lib/ntpath.py
@@ -24,7 +24,7 @@
 from genericpath import *
 
 
-__all__ = ["normcase","isabs","join","splitdrive","split","splitext",
+__all__ = ["normcase","isabs","join","splitdrive","splitroot","split","splitext",
            "basename","dirname","commonprefix","getsize","getmtime",
            "getatime","getctime", "islink","exists","lexists","isdir","isfile",
            "ismount", "expanduser","expandvars","normpath","abspath",
@@ -169,35 +169,58 @@ def splitdrive(p):
 
     Paths cannot contain both a drive letter and a UNC path.
 
+    """
+    drive, root, tail = splitroot(p)
+    return drive, root + tail
+
+
+def splitroot(p):
+    """Split a pathname into drive, root and tail. The drive is defined
+    exactly as in splitdrive(). On Windows, the root may be a single path
+    separator or an empty string. The tail contains anything after the root.
+    For example:
+
+        splitroot('//server/share/') == ('//server/share', '/', '')
+        splitroot('C:/Users/Barney') == ('C:', '/', 'Users/Barney')
+        splitroot('Windows') == ('', '', 'Windows')
     """
     p = os.fspath(p)
-    if len(p) >= 2:
-        if isinstance(p, bytes):
-            sep = b'\\'
-            altsep = b'/'
-            colon = b':'
-            unc_prefix = b'\\\\?\\UNC\\'
-        else:
-            sep = '\\'
-            altsep = '/'
-            colon = ':'
-            unc_prefix = '\\\\?\\UNC\\'
-        normp = p.replace(altsep, sep)
-        if normp[0:2] == sep * 2:
+    if isinstance(p, bytes):
+        sep = b'\\'
+        altsep = b'/'
+        colon = b':'
+        unc_prefix = b'\\\
8000
\?\\UNC\\'
+    else:
+        sep = '\\'
+        altsep = '/'
+        colon = ':'
+        unc_prefix = '\\\\?\\UNC\\'
+    normp = p.replace(altsep, sep)
+    if normp[:1] == sep:
+        if normp[1:2] == sep:
             # UNC drives, e.g. \\server\share or \\?\UNC\server\share
             # Device drives, e.g. \\.\device or \\?\device
             start = 8 if normp[:8].upper() == unc_prefix else 2
             index = normp.find(sep, start)
             if index == -1:
-                return p, p[:0]
+                return p, p[:0], p[:0]
             index2 = normp.find(sep, index + 1)
             if index2 == -1:
-                return p, p[:0]
-            return p[:index2], p[index2:]
-        if normp[1:2] == colon:
-            # Drive-letter drives, e.g. X:
-            return p[:2], p[2:]
-    return p[:0], p
+                return p, p[:0], p[:0]
+            return p[:index2], p[index2:index2 + 1], p[index2 + 1:]
+        else:
+            # Relative path with root, e.g. \Windows
+            return p[:0], p[:1], p[1:]
+    elif normp[1:2] == colon:
+        if normp[2:3] == sep:
+            # Absolute drive-letter path, e.g. X:\Windows
+            return p[:2], p[2:3], p[3:]
+        else:
+            # Relative path with drive, e.g. X:Windows
+            return p[:2], p[:0], p[2:]
+    else:
+        # Relative path, e.g. Windows
+        return p[:0], p[:0], p
 
 
 # Split a path in head (everything up to the last '/') and tail (the

diff --git a/Lib/posixpath.py b/Lib/posixpath.py
@@ -28,7 +28,7 @@
 import genericpath
 from genericpath import *
 
-__all__ = ["normcase","isabs","join","splitdrive","split","splitext",
+__all__ = ["normcase","isabs","join","splitdrive","splitroot","split","splitext",
            "basename","dirname","commonprefix","getsize","getmtime",
            "getatime","getctime","islink","exists","lexists","isdir","isfile",
            "ismount", "expanduser","expandvars","normpath","abspath",
@@ -135,6 +135,30 @@ def splitdrive(p):
     return p[:0], p
 
 
+def splitroot(p):
+    """Split a pathname into drive, root and tail. On Posix, drive is always
+    empty; the root may be empty, a single slash, or two slashes. The tail
+    contains anything after the root. For example:
+
+        splitdrive('foo/bar&
8000
#39;) == ('', '', 'foo/bar')
+        splitdrive('/foo/bar') == ('', '/', 'foo/bar')
+    """
+    p = os.fspath(p)
+    sep = b'/' if isinstance(p, bytes) else '/'
+    if p[:1] != sep:
+        # Relative path, e.g.: 'foo'
+        return p[:0], p[:0], p
+    elif p[1:2] != sep:
+        # Absolute path, e.g.: '/foo'
+        return p[:0], p[:1], p[1:]
+    elif p[2:3] != sep:
+        # Implementation defined per POSIX standard, e.g.: '//foo'
+        return p[:0], p[:2], p[2:]
+    else:
+        # Absolute path with extraneous slashes, e.g.: '///foo', '////foo', etc.
+        return p[:0], p[:1], p[1:]
+
+
 # Return the tail (basename) part of a path, same as split(path)[1].
 
 def basename(p):

diff --git a/Lib/test/test_ntpath.py b/Lib/test/test_ntpath.py
@@ -150,6 +150,22 @@ def test_splitdrive(self):
         tester('ntpath.splitdrive("//x")', ("//x", ""))  # non-empty server & missing share
         tester('ntpath.splitdrive("//x/")', ("//x/", ""))  # non-empty server & empty share
 
+    def test_splitroot(self):
+        tester("ntpath.splitroot('')", ('', '', ''))
+        tester("ntpath.splitroot('a')", ('', '', 'a'))
+        tester("ntpath.splitroot('a\\b')", ('', '', 'a\\b'))
+        tester("ntpath.splitroot('\\a')", ('', '\\', 'a'))
+        tester("ntpath.splitroot('\\a\\b')", ('', '\\', 'a\\b'))
+        tester("ntpath.splitroot('c:a\\b')", ('c:', '', 'a\\b'))
+        tester("ntpath.splitroot('c:\\a\\b')", ('c:', '\\', 'a\\b'))
+        # Redundant slashes are not included in the root.
+        tester("ntpath.splitroot('c:\\\\a')", ('c:', '\\', '\\a'))
+        tester("ntpath.splitroot('c:\\\\\\a/b')", ('c:', '\\', '\\\\a/b'))
+        # Valid UNC paths.
+        tester("ntpath.splitroot('\\\\a\\b')", ('\\\\a\\b', '', ''))
+        tester("ntpath.splitroot('\\\\a\\b\\')", ('\\\\a\\b', '\\', ''))
+        tester("ntpath.splitroot('\\\\a\\b\\c\\d')", ('\\\\a\\b', '\\', 'c\\d'))
+
     def test_split(self):
         tester('ntpath.split("c:\\foo\\bar")', ('c:\\foo', 'bar'))
         tester('ntpath.split("\\\\conky\\mountpoint\\foo\\bar")',

diff --git a/Lib/test/test_posixpath.py b/Lib/test/test_posixpath.py
@@ -115,6 +115,26 @@ def test_splitext(self):
         self.splitextTest("........", "........", "")
         self.splitextTest("", "", "")
 
+    def test_splitroot(self):
+        f = posixpath.splitroot
+        self.assertEqual(f(''), ('', '', ''))
+        self.assertEqual(f('a'), ('', '', 'a'))
+        self.assertEqual(f('a/b'), ('', '', 'a/b'))
+        self.assertEqual(f('a/b/'), ('', '', 'a/b/'))
+        self.assertEqual(f('/a'), ('', '/', 'a'))
+        self.assertEqual(f('/a/b'), ('', '/', 'a/b'))
+        self.assertEqual(f('/a/b/'), ('', '/', 'a/b/'))
+        # The root is collapsed when there are redundant slashes
+        # except when there are exactly two leading slashes, which
+        # is a special case in POSIX.
+        self.assertEqual(f('//a'), ('', '//', 'a'))
+        self.assertEqual(f('///a'), ('', '/', '//a'))
+        self.assertEqual(f('///a/b'), ('', '/', '//a/b'))
+        # Paths which look like NT paths aren't treated specially.
+        self.assertEqual(f('c:/a/b'), ('', '', 'c:/a/b'))
+        self.assertEqual(f('\\/a/b'), ('', '', '\\/a/b'))
+        self.assertEqual(f('\\a\\b'), ('', '', '\\a\\b'))
+
     def test_isabs(self):
         self.assertIs(posixpath.isabs(""), False)
         self.assertIs(posixpath.isabs("/"), True)

diff --git a/Misc/NEWS.d/next/Library/2023-01-12-21-22-20.gh-issue-101000.wz4Xgc.rst b/Misc/NEWS.d/next/Library/2023-01-12-21-22-20.gh-issue-101000.wz4Xgc.rst
@@ -0,0 +1,2 @@
+Add :func:`os.path.splitroot()`, which splits a path into a triad of
+``(drive, root, tail)``.
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1,2 @@
		Add :func:`os.path.splitroot()`, which splits a path into a triad of
		``(drive, root, tail)``.
AlexWaygood marked this conversation as resolved. Show resolved Hide resolved