gh-102511: Speed up os.path.splitroot() with native helpers (GH-118089)

python · nineteendo · Apr 25, 2024 · May 28, 2024 · May 28, 2024 · May 31, 2024
commit 6872edd4264b45f608dda459c6b37997d3c7429c
diff --git a/Include/internal/pycore_fileutils.h b/Include/internal/pycore_fileutils.h
@@ -264,6 +264,8 @@ extern wchar_t *_Py_normpath_and_size(wchar_t *path, Py_ssize_t size, Py_ssize_t
 extern HRESULT PathCchSkipRoot(const wchar_t *pszPath, const wchar_t **ppszRootEnd);
 #endif /* defined(MS_WINDOWS_GAMES) && !defined(MS_WINDOWS_DESKTOP) */
 
+extern void _Py_skiproot(const wchar_t *path, Py_ssize_t size, Py_ssize_t *drvsize, Py_ssize_t *rootsize);
+
 // Macros to protect CRT calls against instant termination when passed an
 // invalid parameter (bpo-23524). IPH stands for Invalid Parameter Handler.
 // Usage:

diff --git a/Lib/ntpath.py b/Lib/ntpath.py
@@ -176,56 +176,76 @@ def splitdrive(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('C:///spam///ham') == ('C:', '/', '//spam///ham')
-        splitroot('Windows/notepad') == ('', '', 'Windows/notepad')
-    """
-    p = os.fspath(p)
-    if isinstance(p, bytes):
-        sep = b'\\'
-        altsep = b'/'
-        colon = b':'
-        unc_prefix = b'\\\\?\\UNC\\'
-        empty = b''
-    else:
-        sep = '\\'
-        altsep = '/'
-        colon = ':'
-        unc_prefix = '\\\\?\\UNC\\'
-        empty = ''
-    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, empty, empty
-            index2 = normp.find(sep, index + 1)
-            if index2 == -1:
-                return p, empty, empty
-            return p[:index2], p[index2:index2 + 1], p[index2 + 1:]
+try:
+    from nt import _path_splitroot_ex
+except ImportError:
+    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('C:///spam///ham') == ('C:', '/', '//spam///ham')
+            splitroot('Windows/notepad') == ('', '', 'Windows/notepad')
+        """
+        p = os.fspath(p)
+        if isinstance(p, bytes):
+            sep = b'\\'
+            altsep = b'/'
+            colon = b':'
+            unc_prefix = b'\\\\?\\UNC\\'
+            empty = b''
         else:
-            # Relative path with root, e.g. \Windows
-            return empty, 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:]
+            sep = '\\'
+            altsep = '/'
+            colon = ':'
+            unc_prefix = '\\\\?\\UNC\\'
+            empty = ''
+        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, empty, empty
+                index2 = normp.find(sep, index + 1)
+                if index2 == -1:
+                    return p, empty, empty
+                return p[:index2], p[index2:index2 + 1], p[index2 + 1:]
+            else:
+                # Relative path with root, e.g. \Windows
+                return empty, 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], empty, p[2:]
         else:
-            # Relative path with drive, e.g. X:Windows
-            return p[:2], empty, p[2:]
-    else:
-        # Relative path, e.g. Windows
-        return empty, empty, p
+            # Relative path, e.g. Windows
+            return empty, empty, p
+else:
+    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('C:///spam///ham') == ('C:', '/', '//spam///ham')
+            splitroot('Windows/notepad') == ('', '', 'Windows/notepad')
+        """
+        p = os.fspath(p)
+        if isinstance(p, bytes):
+            drive, root, tail = _path_splitroot_ex(os.fsdecode(p))
+            return os.fsencode(drive), os.fsencode(root), os.fsencode(tail)
+        return _path_splitroot_ex(p)
 
 
 # Split a path in head (everything up to the last '/') and tail (the

diff --git a/Lib/posixpath.py b/Lib/posixpath.py
@@ -135,33 +135,53 @@ 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:
-
-        splitroot('foo/bar') == ('', '', 'foo/bar')
-        splitroot('/foo/bar') == ('', '/', 'foo/bar')
-        splitroot('//foo/bar') == ('', '//', 'foo/bar')
-        splitroot('///foo/bar') == ('', '/', '//foo/bar')
-    """
-    p = os.fspath(p)
-    if isinstance(p, bytes):
-        sep = b'/'
-        empty = b''
-    else:
-        sep = '/'
-        empty = ''
-    if p[:1] != sep:
-        # Relative path, e.g.: 'foo'
-        return empty, empty, p
-    elif p[1:2] != sep or p[2:3] == sep:
-        # Absolute path, e.g.: '/foo', '///foo', '////foo', etc.
-        return empty, sep, p[1:]
-    else:
-        # Precisely two leading slashes, e.g.: '//foo'. Implementation defined per POSIX, see
-        # https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap04.html#tag_04_13
-        return empty, p[:2], p[2:]
+try:
+    from posix import _path_splitroot_ex
+except ImportError:
+    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:
+
+            splitroot('foo/bar') == ('', '', 'foo/bar')
+            splitroot('/foo/bar') == ('', '/', 'foo/bar')
+            splitroot('//foo/bar') == ('', '//', 'foo/bar')
+            splitroot('///foo/bar') == ('', '/', '//foo/bar')
+        """
+        p = os.fspath(p)
+        if isinstance(p, bytes):
+            sep = b'/'
+            empty = b''
+        else:
+            sep = '/'
+            empty = ''
+        if p[:1] != sep:
+            # Relative path, e.g.: 'foo'
+            return empty, empty, p
+        elif p[1:2] != sep or p[2:3] == sep:
+            # Absolute path, e.g.: '/foo', '///foo', '////foo', etc.
+            return empty, sep, p[1:]
+        else:
+            # Precisely two leading slashes, e.g.: '//foo'. Implementation defined per POSIX, see
+            # https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap04.html#tag_04_13
+            return empty, p[:2], p[2:]
+else:
+    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:
+
+            splitroot('foo/bar') == ('', '', 'foo/bar')
+            splitroot('/foo/bar') == ('', '/', 'foo/bar')
+            splitroot('//foo/bar') == ('', '//', 'foo/bar')
+            splitroot('///foo/bar') == ('', '/', '//foo/bar')
+        """
+        p = os.fspath(p)
+        if isinstance(p, bytes):
+            # Optimisation: the drive is always empty
+            _, root, tail = _path_splitroot_ex(os.fsdecode(p))
+            return b'', os.fsencode(root), os.fsencode(tail)
+        return _path_splitroot_ex(p)
 
 
 # Return the tail (basename) part of a path, same as split(path)[1].

diff --git a/Lib/test/test_ntpath.py b/Lib/test/test_ntpath.py
@@ -354,6 +354,7 @@ def test_normpath(self):
         tester("ntpath.normpath('\\\\foo\\')", '\\\\foo\\')
         tester("ntpath.normpath('\\\\foo')", '\\\\foo')
         tester("ntpath.normpath('\\\\')", '\\\\')
+        tester("ntpath.normpath('//?/UNC/server/share/..')", '\\\\?\\UNC\\server\\share\\')
 
     def test_realpath_curdir(self):
         expected = ntpath.normpath(os.getcwd())

diff --git a/Misc/NEWS.d/next/Core and Builtins/2024-04-19-08-50-48.gh-issue-102511.qDEB66.rst b/Misc/NEWS.d/next/Core and Builtins/2024-04-19-08-50-48.gh-issue-102511.qDEB66.rst
@@ -0,0 +1 @@
+Speed up :func:`os.path.splitroot` with a native implementation.
diff --git a/Modules/clinic/posixmodule.c.h b/Modules/clinic/posixmodule.c.h
diff --git a/Modules/posixmodule.c b/Modules/posixmodule.c
@@ -5286,6 +5286,49 @@ os__path_islink_impl(PyObject *module, PyObject *path)
 #endif /* MS_WINDOWS */
 
 
+/*[clinic input]
+os._path_splitroot_ex
+
+    path: unicode
+
+[clinic start generated code]*/
+
+static PyObject *
+os__path_splitroot_ex_impl(PyObject *module, PyObject *path)
+/*[clinic end generated code: output=de97403d3dfebc40 input=f1470e12d899f9ac]*/
+{
+    Py_ssize_t len, drvsize, rootsize;
+    PyObject *drv = NULL, *root = NULL, *tail = NULL, *result = NULL;
+
+    wchar_t *buffer = PyUnicode_AsWideCharString(path, &len);
+    if (!buffer) {
+        goto exit;
+    }
+
+    _Py_skiproot(buffer, len, &drvsize, &rootsize);
+    drv = PyUnicode_FromWideChar(buffer, drvsize);
+    if (drv == NULL) {
+        goto exit;
+    }
+    root = PyUnicode_FromWideChar(&buffer[drvsize], rootsize);
+    if (root == NULL) {
+        goto exit;
+    }
+    tail = PyUnicode_FromWideChar(&buffer[drvsize + rootsize],
+                                  len - drvsize - rootsize);
+    if (tail == NULL) {
+        goto exit;
+    }
+    result = Py_BuildValue("(OOO)", drv, root, tail);
+exit:
+    PyMem_Free(buffer);
+    Py_XDECREF(drv);
+    Py_XDECREF(root);
+    Py_XDECREF(tail);
+    return result;
+}
+
+
 /*[clinic input]
 os._path_normpath
 
@@ -16029,6 +16072,7 @@ static PyMethodDef posix_methods[] = {
     OS__GETFINALPATHNAME_METHODDEF
     OS__GETVOLUMEPATHNAME_METHODDEF
     OS__PATH_SPLITROOT_METHODDEF
+    OS__PATH_SPLITROOT_EX_METHODDEF
     OS__PATH_NORMPATH_METHODDEF
     OS_GETLOADAVG_METHODDEF
     OS_URANDOM_METHODDEF
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1 @@
		Speed up :func:`os.path.splitroot` with a native implementation.