Sync from upstream.

CHANGES.rst

@@ -6,6 +6,7 @@ Unreleased
 
 - Add ``PurePathBase.full_match()``.
 - Revert ``match()`` back to 3.12 behaviour (no recursive wildcards).
+- Disallow passing ``bytes`` to initialisers.
 - Improve walking and globbing performance.
 - Expand test coverage.
 - Clarify that we're using the PSF license.

pathlib_abc/__init__.py

@@ -152,39 +152,39 @@ class PathModuleBase:
     """
 
     @classmethod
-    def _unsupported(cls, attr):
-        raise UnsupportedOperation(f"{cls.__name__}.{attr} is unsupported")
+    def _unsupported_msg(cls, attribute):
+        return f"{cls.__name__}.{attribute} is unsupported"
 
     @property
     def sep(self):
         """The character used to separate path components."""
-        self._unsupported('sep')
+        raise UnsupportedOperation(self._unsupported_msg('sep'))
 
     def join(self, path, *paths):
         """Join path segments."""
-        self._unsupported('join()')
+        raise UnsupportedOperation(self._unsupported_msg('join()'))
 
     def split(self, path):
         """Split the path into a pair (head, tail), where *head* is everything
         before the final path separator, and *tail* is everything after.
         Either part may be empty.
         """
-        self._unsupported('split()')
+        raise UnsupportedOperation(self._unsupported_msg('split()'))
 
     def splitdrive(self, path):
         """Split the path into a 2-item tuple (drive, tail), where *drive* is
         a device name or mount point, and *tail* is everything after the
         drive. Either part may be empty."""
-        self._unsupported('splitdrive()')
+        raise UnsupportedOperation(self._unsupported_msg('splitdrive()'))
 
     def normcase(self, path):
         """Normalize the case of the path."""
-        self._unsupported('normcase()')
+        raise UnsupportedOperation(self._unsupported_msg('normcase()'))
 
     def isabs(self, path):
         """Returns whether the path is absolute, i.e. unaffected by the
         current directory or drive."""
-        self._unsupported('isabs()')
+        raise UnsupportedOperation(self._unsupported_msg('isabs()'))
 
 
 class PurePathBase(abc.ABC):
@@ -210,6 +210,9 @@ class PurePathBase(abc.ABC):
 
     def __init__(self, path, *paths):
         self._raw_path = self.pathmod.join(path, *paths) if paths else path
+        if not isinstance(self._raw_path, str):
+            raise TypeError(
+                f"path should be a str, not {type(self._raw_path).__name__!r}")
         self._resolving = False
 
     def with_segments(self, *pathsegments):
@@ -302,10 +305,13 @@ def with_suffix(self, suffix):
         has no suffix, add given suffix.  If the given suffix is an empty
         string, remove the suffix from the path.
         """
+        stem = self.stem
         if not suffix:
-            return self.with_name(self.stem)
+            return self.with_name(stem)
+        elif not stem:
+            raise ValueError(f"{self!r} has an empty name")
         elif suffix.startswith('.') and len(suffix) > 1:
-            return self.with_name(self.stem + suffix)
+            return self.with_name(stem + suffix)
         else:
             raise ValueError(f"Invalid suffix {suffix!r}")
 
@@ -321,8 +327,6 @@ def relative_to(self, other, *, walk_up=False):
             other = self.with_segments(other)
         anchor0, parts0 = self._stack
         anchor1, parts1 = other._stack
-        if isinstance(anchor0, str) != isinstance(anchor1, str):
-            raise TypeError(f"{self._raw_path!r} and {other._raw_path!r} have different types")
         if anchor0 != anchor1:
             raise ValueError(f"{self._raw_path!r} and {other._raw_path!r} have different anchors")
         while parts0 and parts1 and parts0[-1] == parts1[-1]:
@@ -346,8 +350,6 @@ def is_relative_to(self, other):
             other = self.with_segments(other)
         anchor0, parts0 = self._stack
         anchor1, parts1 = other._stack
-        if isinstance(anchor0, str) != isinstance(anchor1, str):
-            raise TypeError(f"{self._raw_path!r} and {other._raw_path!r} have different types")
         if anchor0 != anchor1:
             return False
         while parts0 and parts1 and parts0[-1] == parts1[-1]:
@@ -505,16 +507,15 @@ class PathBase(PurePathBase):
     _max_symlinks = 40
 
     @classmethod
-    def _unsupported(cls, method_name):
-        msg = f"{cls.__name__}.{method_name}() is unsupported"
-        raise UnsupportedOperation(msg)
+    def _unsupported_msg(cls, attribute):
+        return f"{cls.__name__}.{attribute} is unsupported"
 
     def stat(self, *, follow_symlinks=True):
         """
         Return the result of the stat() system call on this path, like
         os.stat() does.
         """
-        self._unsupported("stat")
+        raise UnsupportedOperation(self._unsupported_msg('stat()'))
 
     def lstat(self):
         """
@@ -703,7 +704,7 @@ def open(self, mode='r', buffering=-1, encoding=None,
         Open the file pointed by this path and return a file object, as
         the built-in open() function does.
         """
-        self._unsupported("open")
+        raise UnsupportedOperation(self._unsupported_msg('open()'))
 
     def read_bytes(self):
         """
@@ -744,7 +745,7 @@ def iterdir(self):
         The children are yielded in arbitrary order, and the
         special entries '.' and '..' are not included.
         """
-        self._unsupported("iterdir")
+        raise UnsupportedOperation(self._unsupported_msg('iterdir()'))
 
     def _scandir(self):
         # Emulate os.scandir(), which returns an object that can be used as a
@@ -871,7 +872,7 @@ def absolute(self):
 
         Use resolve() to resolve symlinks and remove '..' segments.
         """
-        self._unsupported("absolute")
+        raise UnsupportedOperation(self._unsupported_msg('absolute()'))
 
     @classmethod
     def cwd(cls):
@@ -886,7 +887,7 @@ def expanduser(self):
         """ Return a new path with expanded ~ and ~user constructs
         (as returned by os.path.expanduser)
         """
-        self._unsupported("expanduser")
+        raise UnsupportedOperation(self._unsupported_msg('expanduser()'))
 
     @classmethod
     def home(cls):
@@ -898,7 +899,7 @@ def readlink(self):
         """
         Return the path to which the symbolic link points.
         """
-        self._unsupported("readlink")
+        raise UnsupportedOperation(self._unsupported_msg('readlink()'))
     readlink._supported = False
 
     def resolve(self, strict=False):
@@ -973,27 +974,27 @@ def symlink_to(self, target, target_is_directory=False):
         Make this path a symlink pointing to the target path.
         Note the order of arguments (link, target) is the reverse of os.symlink.
         """
-        self._unsupported("symlink_to")
+        raise UnsupportedOperation(self._unsupported_msg('symlink_to()'))
 
     def hardlink_to(self, target):
         """
         Make this path a hard link pointing to the same file as *target*.
 
         Note the order of arguments (self, target) is the reverse of os.link's.
         """
-        self._unsupported("hardlink_to")
+        raise UnsupportedOperation(self._unsupported_msg('hardlink_to()'))
 
     def touch(self, mode=0o666, exist_ok=True):
         """
         Create this file with the given access mode, if it doesn't exist.
         """
-        self._unsupported("touch")
+        raise UnsupportedOperation(self._unsupported_msg('touch()'))
 
     def mkdir(self, mode=0o777, parents=False, exist_ok=False):
         """
         Create a new directory at this given path.
         """
-        self._unsupported("mkdir")
+        raise UnsupportedOperation(self._unsupported_msg('mkdir()'))
 
     def rename(self, target):
         """
@@ -1005,7 +1006,7 @@ def rename(self, target):
 
         Returns the new Path instance pointing to the target path.
         """
-        self._unsupported("rename")
+        raise UnsupportedOperation(self._unsupported_msg('rename()'))
 
     def replace(self, target):
         """
@@ -1017,13 +1018,13 @@ def replace(self, target):
 
         Returns the new Path instance pointing to the target path.
         """
-        self._unsupported("replace")
+        raise UnsupportedOperation(self._unsupported_msg('replace()'))
 
     def chmod(self, mode, *, follow_symlinks=True):
         """
         Change the permissions of the path, like os.chmod().
         """
-        self._unsupported("chmod")
+        raise UnsupportedOperation(self._unsupported_msg('chmod()'))
 
     def lchmod(self, mode):
         """
@@ -1037,34 +1038,34 @@ def unlink(self, missing_ok=False):
         Remove this file or link.
         If the path is a directory, use rmdir() instead.
         """
-        self._unsupported("unlink")
+        raise UnsupportedOperation(self._unsupported_msg('unlink()'))
 
     def rmdir(self):
         """
         Remove this directory.  The directory must be empty.
         """
-        self._unsupported("rmdir")
+        raise UnsupportedOperation(self._unsupported_msg('rmdir()'))
 
     def owner(self, *, follow_symlinks=True):
         """
         Return the login name of the file owner.
         """
-        self._unsupported("owner")
+        raise UnsupportedOperation(self._unsupported_msg('owner()'))
 
     def group(self, *, follow_symlinks=True):
         """
         Return the group name of the file gid.
         """
-        self._unsupported("group")
+        raise UnsupportedOperation(self._unsupported_msg('group()'))
 
     @classmethod
     def from_uri(cls, uri):
         """Return a new path from the given 'file' URI."""
-        cls._unsupported("from_uri")
+        raise UnsupportedOperation(cls._unsupported_msg('from_uri()'))
 
     def as_uri(self):
         """Return the path as a URI."""
-        self._unsupported("as_uri")
+        raise UnsupportedOperation(self._unsupported_msg('as_uri()'))
 
 
 PurePathBase.register(pathlib.PurePath)

tests/test_pathlib_abc.py

@@ -155,6 +155,31 @@ def test_constructor_common(self):
         P('a/b/c')
         P('/a/b/c')
 
+    def test_bytes(self):
+        P = self.cls
+        with self.assertRaises(TypeError):
+            P(b'a')
+        with self.assertRaises(TypeError):
+            P(b'a', 'b')
+        with self.assertRaises(TypeError):
+            P('a', b'b')
+        with self.assertRaises(TypeError):
+            P('a').joinpath(b'b')
+        with self.assertRaises(TypeError):
+            P('a') / b'b'
+        with self.assertRaises(TypeError):
+            b'a' / P('b')
+        with self.assertRaises(TypeError):
+            P('a').match(b'b')
+        with self.assertRaises(TypeError):
+            P('a').relative_to(b'b')
+        with self.assertRaises(TypeError):
+            P('a').with_name(b'b')
+        with self.assertRaises(TypeError):
+            P('a').with_stem(b'b')
+        with self.assertRaises(TypeError):
+            P('a').with_suffix(b'b')
+
     def _check_str_subclass(self, *args):
         # Issue #21127: it should be possible to construct a PurePath object
         # from a str subclass instance, and it then gets converted to
@@ -977,9 +1002,8 @@ def test_with_suffix_windows(self):
     def test_with_suffix_empty(self):
         P = self.cls
         # Path doesn't have a "filename" component.
-        self.assertEqual(P('').with_suffix('.gz'), P('.gz'))
-        self.assertEqual(P('.').with_suffix('.gz'), P('..gz'))
-        self.assertEqual(P('/').with_suffix('.gz'), P('/.gz'))
+        self.assertRaises(ValueError, P('').with_suffix, '.gz')
+        self.assertRaises(ValueError, P('/').with_suffix, '.gz')
 
     def test_with_suffix_seps(self):
         P = self.cls
@@ -1766,16 +1790,26 @@ def _check(path, glob, expected):
         _check(p, "*/fileB", ["dirB/fileB", "linkB/fileB"])
         _check(p, "*/", ["dirA/", "dirB/", "dirC/", "dirE/", "linkB/"])
         _check(p, "dir*/*/..", ["dirC/dirD/..", "dirA/linkC/..", "dirB/linkD/.."])
+        _check(p, "dir*/**", [
+            "dirA", "dirA/linkC", "dirA/linkC/fileB", "dirA/linkC/linkD", "dirA/linkC/linkD/fileB",
+            "dirB", "dirB/fileB", "dirB/linkD", "dirB/linkD/fileB",
+            "dirC", "dirC/fileC", "dirC/dirD",  "dirC/dirD/fileD", "dirC/novel.txt",
+            "dirE"])
         _check(p, "dir*/**/", ["dirA/", "dirA/linkC/", "dirA/linkC/linkD/", "dirB/", "dirB/linkD/",
                                "dirC/", "dirC/dirD/", "dirE/"])
         _check(p, "dir*/**/..", ["dirA/..", "dirA/linkC/..", "dirB/..",
                                  "dirB/linkD/..", "dirA/linkC/linkD/..",
                                  "dirC/..", "dirC/dirD/..", "dirE/.."])
+        _check(p, "dir*/*/**", [
+            "dirA/linkC", "dirA/linkC/linkD", "dirA/linkC/fileB", "dirA/linkC/linkD/fileB",
+            "dirB/linkD", "dirB/linkD/fileB",
+            "dirC/dirD", "dirC/dirD/fileD"])
         _check(p, "dir*/*/**/", ["dirA/linkC/", "dirA/linkC/linkD/", "dirB/linkD/", "dirC/dirD/"])
         _check(p, "dir*/*/**/..", ["dirA/linkC/..", "dirA/linkC/linkD/..",
                                    "dirB/linkD/..", "dirC/dirD/.."])
         _check(p, "dir*/**/fileC", ["dirC/fileC"])
         _check(p, "dir*/*/../dirD/**/", ["dirC/dirD/../dirD/"])
+        _check(p, "*/dirD/**", ["dirC/dirD", "dirC/dirD/fileD"])
         _check(p, "*/dirD/**/", ["dirC/dirD/"])
 
     @needs_symlinks
@@ -1792,12 +1826,20 @@ def _check(path, glob, expected):
         _check(p, "*/fileB", ["dirB/fileB"])
         _check(p, "*/", ["dirA/", "dirB/", "dirC/", "dirE/"])
         _check(p, "dir*/*/..", ["dirC/dirD/.."])
+        _check(p, "dir*/**", [
+            "dirA", "dirA/linkC",
+            "dirB", "dirB/fileB", "dirB/linkD",
+            "dirC", "dirC/fileC", "dirC/dirD", "dirC/dirD/fileD", "dirC/novel.txt",
+            "dirE"])
         _check(p, "dir*/**/", ["dirA/", "dirB/", "dirC/", "dirC/dirD/", "dirE/"])
         _check(p, "dir*/**/..", ["dirA/..", "dirB/..", "dirC/..", "dirC/dirD/..", "dirE/.."])
+        _check(p, "dir*/*/**", ["dirC/dirD", "dirC/dirD/fileD"])
         _check(p, "dir*/*/**/", ["dirC/dirD/"])
         _check(p, "dir*/*/**/..", ["dirC/dirD/.."])
         _check(p, "dir*/**/fileC", ["dirC/fileC"])
+        _check(p, "dir*/*/../dirD/**", ["dirC/dirD/../dirD", "dirC/dirD/../dirD/fileD"])
         _check(p, "dir*/*/../dirD/**/", ["dirC/dirD/../dirD/"])
+        _check(p, "*/dirD/**", ["dirC/dirD", "dirC/dirD/fileD"])
         _check(p, "*/dirD/**/", ["dirC/dirD/"])
 
     def test_rglob_common(self):
@@ -1834,10 +1876,13 @@ def _check(glob, expected):
                               "dirC/dirD", "dirC/dirD/fileD"])
         _check(p.rglob("file*"), ["dirC/fileC", "dirC/dirD/fileD"])
         _check(p.rglob("**/file*"), ["dirC/fileC", "dirC/dirD/fileD"])
+        _check(p.rglob("dir*/**"), ["dirC/dirD", "dirC/dirD/fileD"])
         _check(p.rglob("dir*/**/"), ["dirC/dirD/"])
         _check(p.rglob("*/*"), ["dirC/dirD/fileD"])
         _check(p.rglob("*/"), ["dirC/dirD/"])
         _check(p.rglob(""), ["dirC/", "dirC/dirD/"])
+        _check(p.rglob("**"), [
+            "dirC", "dirC/fileC", "dirC/dirD", "dirC/dirD/fileD", "dirC/novel.txt"])
         _check(p.rglob("**/"), ["dirC/", "dirC/dirD/"])
         # gh-91616, a re module regression
         _check(p.rglob("*.txt"), ["dirC/novel.txt"])