GH-114847: Speed up posixpath.realpath()

Lib/posixpath.py

@@ -412,49 +412,55 @@ def realpath(filename, *, strict=False):
     """Return the canonical path of the specified filename, eliminating any
 symbolic links encountered in the path."""
     filename = os.fspath(filename)
-    path, ok = _joinrealpath(filename[:0], filename, strict, {})
-    return abspath(path)
-
-# Join two paths, normalizing and eliminating any symbolic links
-# encountered in the second path.
-def _joinrealpath(path, rest, strict, seen):
-    if isinstance(path, bytes):
+    if isinstance(filename, bytes):
         sep = b'/'
         curdir = b'.'
         pardir = b'..'
+        getcwd = os.getcwdb
     else:
         sep = '/'
         curdir = '.'
         pardir = '..'
+        getcwd = os.getcwd
+
+    # The stack of unresolved path parts. When popped, a special value of None
+    # indicates that a symlink target has been resolved, and that the original
+    # symlink path can be retrieved by popping again. The [::-1] slice is a
+    # very fast way of spelling list(reversed(...)).
+    rest = filename.split(sep)[::-1]
 
-    if isabs(rest):
-        rest = rest[1:]
-        path = sep
+    # The resolved path, which is absolute throughout this function.
+    # Note: getcwd() returns a normalized and symlink-free path.
+    path = sep if filename.startswith(sep) else getcwd()
+
+    # Mapping from symlink paths to *fully resolved* symlink targets. If a
+    # symlink is encountered but not yet resolved, the value is None. This is
+    # used both to detect symlink loops and to speed up repeated traversals of
+    # the same links.
+    seen = {}
+
+    # Whether we're calling lstat() and readlink() to resolve symlinks. If we
+    # encounter an OSError in non-strict mode, this is switched off.
+    querying = True
 
     while rest:
-        name, _, rest = rest.partition(sep)
+        name = rest.pop()
+        if name is None:
+            # resolved symlink target
+            seen[rest.pop()] = path
+            continue
         if not name or name == curdir:
             # current dir
             continue
         if name == pardir:
             # parent dir
-            if path:
-                path, name = split(path)
-                if name == pardir:
-                    path = join(path, pardir, pardir)
-            else:
-                path = pardir
+            path = path[:path.rfind(sep)] or sep
             continue
-        newpath = join(path, name)
-        try:
-            st = os.lstat(newpath)
-        except OSError:
-            if strict:
-                raise
-            is_link = False
+        if path == sep:
+            newpath = path + name
         else:
-            is_link = stat.S_ISLNK(st.st_mode)
-        if not is_link:
+            newpath = path + sep + name
+        if not querying:
             path = newpath
             continue
         # Resolve the symbolic link
@@ -470,14 +476,36 @@ def _joinrealpath(path, rest, strict, seen):
                 os.stat(newpath)
             else:
                 # Return already resolved part + rest of the path unchanged.
-                return join(newpath, rest), False
+                path = newpath
+                querying = False
+                continue
+        try:
+            st = os.lstat(newpath)
+            if not stat.S_ISLNK(st.st_mode):
+                path = newpath
+                continue
+            target = os.readlink(newpath)
+        except OSError:
+            if strict:
+                raise
+            else:
+                # Return already resolved part + rest of the path unchanged.
+                path = newpath
+                querying = False
+                continue
+        if target.startswith(sep):
+            # Symlink target is absolute; reset resolved path.
+            path = sep
         seen[newpath] = None # not resolved symlink
-        path, ok = _joinrealpath(path, os.readlink(newpath), strict, seen)
-        if not ok:
-            return join(path, rest), False
-        seen[newpath] = path # resolved symlink
+        # Push the symlink path onto the stack, and signal its specialness by
+        # also pushing None. When these entries are popped, we'll record the
+        # fully-resolved symlink target in the 'seen' mapping.
+        rest.append(newpath)
+        rest.append(None)
+        # Push the unresolved symlink target parts onto the stack.
+        rest.extend(target.split(sep)[::-1])
 
-    return path, True
+    return path
 
 
 supports_unicode_filenames = (sys.platform == 'darwin')

Misc/NEWS.d/next/Library/2024-02-01-08-09-20.gh-issue-114847.-JrWrR.rst

@@ -0,0 +1 @@
+Speed up :func:`os.path.realpath` on non-Windows platforms.