pythonGH-89812: Add pathlib._LexicalPath

barneygale · barneygale · commit 81e301b876a0 · 2023-05-23T20:31:11.000+01:00
This internal class excludes the `__fspath__()`, `__bytes__()` and
`as_uri()` methods, which must not be inherited by a future
`tarfile.TarPath` class.
diff --git a/Lib/pathlib.py b/Lib/pathlib.py
@@ -233,14 +233,10 @@ def __repr__(self):
         return "<{}.parents>".format(type(self._path).__name__)
 
 
-class PurePath(object):
-    """Base class for manipulating paths without I/O.
+class _LexicalPath(object):
+    """Base class for manipulating paths using only lexical operations.
 
-    PurePath represents a filesystem path and offers operations which
-    don't imply any actual filesystem I/O.  Depending on your system,
-    instantiating a PurePath will return either a PurePosixPath or a
-    PureWindowsPath object.  You can also instantiate either of these classes
-    directly, regardless of your system.
+    This class does not provide __fspath__(), __bytes__() or as_uri().
     """
 
     __slots__ = (
@@ -280,16 +276,6 @@ class PurePath(object):
     )
     _flavour = os.path
 
-    def __new__(cls, *args, **kwargs):
-        """Construct a PurePath from one or several strings and or existing
-        PurePath objects.  The strings and path objects are combined so as
-        to yield a canonicalized path, which is incorporated into the
-        new PurePath object.
-        """
-        if cls is PurePath:
-            cls = PureWindowsPath if os.name == 'nt' else PurePosixPath
-        return object.__new__(cls)
-
     def __reduce__(self):
         # Using the parts tuple helps share interned path parts
         # when pickling related paths.
@@ -298,7 +284,7 @@ def __reduce__(self):
     def __init__(self, *args):
         paths = []
         for arg in args:
-            if isinstance(arg, PurePath):
+            if isinstance(arg, _LexicalPath):
                 path = arg._raw_path
             else:
                 try:
@@ -378,43 +364,15 @@ def __str__(self):
                                                   self._tail) or '.'
             return self._str
 
-    def __fspath__(self):
-        return str(self)
-
     def as_posix(self):
         """Return the string representation of the path with forward (/)
         slashes."""
         f = self._flavour
         return str(self).replace(f.sep, '/')
 
-    def __bytes__(self):
-        """Return the bytes representation of the path.  This is only
-        recommended to use under Unix."""
-        return os.fsencode(self)
-
     def __repr__(self):
         return "{}({!r})".format(self.__class__.__name__, self.as_posix())
 
-    def as_uri(self):
-        """Return the path as a 'file' URI."""
-        if not self.is_absolute():
-            raise ValueError("relative path can't be expressed as a file URI")
-
-        drive = self.drive
-        if len(drive) == 2 and drive[1] == ':':
-            # It's a path on a local drive => 'file:///c:/a/b'
-            prefix = 'file:///' + drive
-            path = self.as_posix()[2:]
-        elif drive:
-            # It's a path on a network drive => 'file://host/share/a/b'
-            prefix = 'file:'
-            path = self.as_posix()
-        else:
-            # It's a posix path => 'file:///etc/hosts'
-            prefix = 'file://'
-            path = str(self)
-        return prefix + urlquote_from_bytes(os.fsencode(path))
-
     @property
     def _str_normcase(self):
         # String with normalized case, for hashing and equality checks
@@ -434,7 +392,7 @@ def _parts_normcase(self):
             return self._parts_normcase_cached
 
     def __eq__(self, other):
-        if not isinstance(other, PurePath):
+        if not isinstance(other, _LexicalPath):
             return NotImplemented
         return self._str_normcase == other._str_normcase and self._flavour is other._flavour
 
@@ -446,22 +404,22 @@ def __hash__(self):
             return self._hash
 
     def __lt__(self, other):
-        if not isinstance(other, PurePath) or self._flavour is not other._flavour:
+        if not isinstance(other, _LexicalPath) or self._flavour is not other._flavour:
             return NotImplemented
         return self._parts_normcase < other._parts_normcase
 
     def __le__(self, other):
-        if not isinstance(other, PurePath) or self._flavour is not other._flavour:
+        if not isinstance(other, _LexicalPath) or self._flavour is not other._flavour:
             return NotImplemented
         return self._parts_normcase <= other._parts_normcase
 
     def __gt__(self, other):
-        if not isinstance(other, PurePath) or self._flavour is not other._flavour:
+        if not isinstance(other, _LexicalPath) or self._flavour is not other._flavour:
             return NotImplemented
         return self._parts_normcase > other._parts_normcase
 
     def __ge__(self, other):
-        if not isinstance(other, PurePath) or self._flavour is not other._flavour:
+        if not isinstance(other, _LexicalPath) or self._flavour is not other._flavour:
             return NotImplemented
         return self._parts_normcase >= other._parts_normcase
 
@@ -707,6 +665,57 @@ def match(self, path_pattern, *, case_sensitive=None):
                 return False
         return True
 
+
+class PurePath(_LexicalPath):
+    """Base class for manipulating paths without I/O.
+
+    PurePath represents a filesystem path and offers operations which
+    don't imply any actual filesystem I/O.  Depending on your system,
+    instantiating a PurePath will return either a PurePosixPath or a
+    PureWindowsPath object.  You can also instantiate either of these classes
+    directly, regardless of your system.
+    """
+    __slots__ = ()
+
+    def __new__(cls, *args, **kwargs):
+        """Construct a PurePath from one or several strings and or existing
+        PurePath objects.  The strings and path objects are combined so as
+        to yield a canonicalized path, which is incorporated into the
+        new PurePath object.
+        """
+        if cls is PurePath:
+            cls = PureWindowsPath if os.name == 'nt' else PurePosixPath
+        return object.__new__(cls)
+
+    def __fspath__(self):
+        return str(self)
+
+    def __bytes__(self):
+        """Return the bytes representation of the path.  This is only
+        recommended to use under Unix."""
+        return os.fsencode(self)
+
+    def as_uri(self):
+        """Return the path as a 'file' URI."""
+        if not self.is_absolute():
+            raise ValueError("relative path can't be expressed as a file URI")
+
+        drive = self.drive
+        if len(drive) == 2 and drive[1] == ':':
+            # It's a path on a local drive => 'file:///c:/a/b'
+            prefix = 'file:///' + drive
+            path = self.as_posix()[2:]
+        elif drive:
+            # It's a path on a network drive => 'file://host/share/a/b'
+            prefix = 'file:'
+            path = self.as_posix()
+        else:
+            # It's a posix path => 'file:///etc/hosts'
+            prefix = 'file://'
+            path = str(self)
+        return prefix + urlquote_from_bytes(os.fsencode(path))
+
+
 # Can't subclass os.PathLike from PurePath and keep the constructor
 # optimizations in PurePath.__slots__.
 os.PathLike.register(PurePath)
diff --git a/Lib/test/test_pathlib.py b/Lib/test/test_pathlib.py
@@ -37,7 +37,7 @@ def with_segments(self, *pathsegments):
         return type(self)(*pathsegments, session_id=self.session_id)
 
 
-class _BasePurePathTest(object):
+class _BaseLexicalPathTest(object):
 
     # Keys are canonical paths, values are list of tuples of arguments
     # supposed to produce equal paths.
@@ -227,18 +227,6 @@ def test_as_posix_common(self):
             self.assertEqual(P(pathstr).as_posix(), pathstr)
         # Other tests for as_posix() are in test_equivalences().
 
-    def test_as_bytes_common(self):
-        sep = os.fsencode(self.sep)
-        P = self.cls
-        self.assertEqual(bytes(P('a/b')), b'a' + sep + b'b')
-
-    def test_as_uri_common(self):
-        P = self.cls
-        with self.assertRaises(ValueError):
-            P('a').as_uri()
-        with self.assertRaises(ValueError):
-            P().as_uri()
-
     def test_repr_common(self):
         for pathstr in ('a', 'a/b', 'a/b/c', '/', '/a/b', '/a/b/c'):
             with self.subTest(pathstr=pathstr):
@@ -358,12 +346,6 @@ def test_parts_common(self):
         parts = p.parts
         self.assertEqual(parts, (sep, 'a', 'b'))
 
-    def test_fspath_common(self):
-        P = self.cls
-        p = P('a/b')
-        self._check_str(p.__fspath__(), ('a/b',))
-        self._check_str(os.fspath(p), ('a/b',))
-
     def test_equivalences(self):
         for k, tuples in self.equivalences.items():
             canon = k.replace('/', self.sep)
@@ -702,6 +684,30 @@ def test_pickling_common(self):
             self.assertEqual(str(pp), str(p))
 
 
+class LexicalPathTest(_BaseLexicalPathTest, unittest.TestCase):
+    cls = pathlib._LexicalPath
+
+
+class _BasePurePathTest(_BaseLexicalPathTest):
+    def test_fspath_common(self):
+        P = self.cls
+        p = P('a/b')
+        self._check_str(p.__fspath__(), ('a/b',))
+        self._check_str(os.fspath(p), ('a/b',))
+
+    def test_bytes_common(self):
+        sep = os.fsencode(self.sep)
+        P = self.cls
+        self.assertEqual(bytes(P('a/b')), b'a' + sep + b'b')
+
+    def test_as_uri_common(self):
+        P = self.cls
+        with self.assertRaises(ValueError):
+            P('a').as_uri()
+        with self.assertRaises(ValueError):
+            P().as_uri()
+
+
 class PurePosixPathTest(_BasePurePathTest, unittest.TestCase):
     cls = pathlib.PurePosixPath
 
