Update zipimport from CPython 3.10.5

Author: CPython developers

Lib/test/test_zipimport.py

@@ -467,15 +467,15 @@ def testZipImporterMethods(self):
         # PEP 302
         with warnings.catch_warnings():
             warnings.simplefilter("ignore", DeprecationWarning)
-        find_mod = zi.find_module('spam')
-        self.assertIsNotNone(find_mod)
-        self.assertIsInstance(find_mod, zipimport.zipimporter)
-        self.assertFalse(find_mod.is_package('spam'))
-        load_mod = find_mod.load_module('spam')
-        self.assertEqual(find_mod.get_filename('spam'), load_mod.__file__)
-
-        mod = zi.load_module(TESTPACK)
-        self.assertEqual(zi.get_filename(TESTPACK), mod.__file__)
+            find_mod = zi.find_module('spam')
+            self.assertIsNotNone(find_mod)
+            self.assertIsInstance(find_mod, zipimport.zipimporter)
+            self.assertFalse(find_mod.is_package('spam'))
+            load_mod = find_mod.load_module('spam')
+            self.assertEqual(find_mod.get_filename('spam'), load_mod.__file__)
+
+            mod = zi.load_module(TESTPACK)
+            self.assertEqual(zi.get_filename(TESTPACK), mod.__file__)
 
         # PEP 451
         spec = zi.find_spec('spam')
@@ -584,8 +584,8 @@ def testZipImporterMethodsInSubDirectory(self):
         # PEP 302
         with warnings.catch_warnings():
             warnings.simplefilter("ignore", DeprecationWarning)
-        mod = zi.load_module(TESTPACK2)
-        self.assertEqual(zi.get_filename(TESTPACK2), mod.__file__)
+            mod = zi.load_module(TESTPACK2)
+            self.assertEqual(zi.get_filename(TESTPACK2), mod.__file__)
         # PEP 451
         spec = zi.find_spec(TESTPACK2)
         mod = importlib.util.module_from_spec(spec)
@@ -600,13 +600,13 @@ def testZipImporterMethodsInSubDirectory(self):
         # PEP 302
         with warnings.catch_warnings():
             warnings.simplefilter("ignore", DeprecationWarning)
-        find_mod_dotted = zi2.find_module(TESTMOD)
-        self.assertIsNotNone(find_mod_dotted)
-        self.assertIsInstance(find_mod_dotted, zipimport.zipimporter)
-        self.assertFalse(zi2.is_package(TESTMOD))
-        load_mod = find_mod_dotted.load_module(TESTMOD)
-        self.assertEqual(
-            find_mod_dotted.get_filename(TESTMOD), load_mod.__file__)
+            find_mod_dotted = zi2.find_module(TESTMOD)
+            self.assertIsNotNone(find_mod_dotted)
+            self.assertIsInstance(find_mod_dotted, zipimport.zipimporter)
+            self.assertFalse(zi2.is_package(TESTMOD))
+            load_mod = find_mod_dotted.load_module(TESTMOD)
+            self.assertEqual(
+                find_mod_dotted.get_filename(TESTMOD), load_mod.__file__)
 
         # PEP 451
         spec = zi2.find_spec(TESTMOD)
@@ -846,7 +846,7 @@ def _testBogusZipFile(self):
         try:
             with warnings.catch_warnings():
                 warnings.simplefilter("ignore", DeprecationWarning)
-            self.assertRaises(TypeError, z.load_module, None)
+                self.assertRaises(TypeError, z.load_module, None)
             self.assertRaises(TypeError, z.find_module, None)
             self.assertRaises(TypeError, z.find_spec, None)
             self.assertRaises(TypeError, z.exec_module, None)
@@ -861,7 +861,7 @@ def _testBogusZipFile(self):
 
             with warnings.catch_warnings():
                 warnings.simplefilter("ignore", DeprecationWarning)
-            self.assertRaises(error, z.load_module, 'abc')
+                self.assertRaises(error, z.load_module, 'abc')
             self.assertRaises(error, z.get_code, 'abc')
             self.assertRaises(OSError, z.get_data, 'abc')
             self.assertRaises(error, z.get_source, 'abc')
@@ -871,7 +871,7 @@ def _testBogusZipFile(self):
 
 
 def tearDownModule():
-        os_helper.unlink(TESTMOD)
+    os_helper.unlink(TESTMOD)
 
 
 if __name__ == "__main__":

Lib/zipimport.py

@@ -22,6 +22,7 @@
 import marshal  # for loads
 import sys  # for modules
 import time  # for mktime
+import _warnings  # For warn()
 
 __all__ = ['ZipImportError', 'zipimporter']
 
@@ -42,7 +43,7 @@ class ZipImportError(ImportError):
 STRING_END_ARCHIVE = b'PK\x05\x06'
 MAX_COMMENT_LEN = (1 << 16) - 1
 
-class zipimporter:
+class zipimporter(_bootstrap_external._LoaderBasics):
     """zipimporter(archivepath) -> zipimporter object
 
     Create a new zipimporter instance. 'archivepath' must be a path to
@@ -115,7 +116,12 @@ def find_loader(self, fullname, path=None):
         full path name if it's possibly a portion of a namespace package,
         or None otherwise. The optional 'path' argument is ignored -- it's
         there for compatibility with the importer protocol.
+
+        Deprecated since Python 3.10. Use find_spec() instead.
         """
+        _warnings.warn("zipimporter.find_loader() is deprecated and slated for "
+                       "removal in Python 3.12; use find_spec() instead",
+                       DeprecationWarning)
         mi = _get_module_info(self, fullname)
         if mi is not None:
             # This is a module or package.
@@ -146,15 +152,46 @@ def find_module(self, fullname, path=None):
         instance itself if the module was found, or None if it wasn't.
         The optional 'path' argument is ignored -- it's there for compatibility
         with the importer protocol.
+
+        Deprecated since Python 3.10. Use find_spec() instead.
         """
+        _warnings.warn("zipimporter.find_module() is deprecated and slated for "
+                       "removal in Python 3.12; use find_spec() instead",
+                       DeprecationWarning)
         return self.find_loader(fullname, path)[0]
 
+    def find_spec(self, fullname, target=None):
+        """Create a ModuleSpec for the specified module.
+
+        Returns None if the module cannot be found.
+        """
+        module_info = _get_module_info(self, fullname)
+        if module_info is not None:
+            return _bootstrap.spec_from_loader(fullname, self, is_package=module_info)
+        else:
+            # Not a module or regular package. See if this is a directory, and
+            # therefore possibly a portion of a namespace package.
+
+            # We're only interested in the last path component of fullname
+            # earlier components are recorded in self.prefix.
+            modpath = _get_module_path(self, fullname)
+            if _is_dir(self, modpath):
+                # This is possibly a portion of a namespace
+                # package. Return the string representing its path,
+                # without a trailing separator.
+                path = f'{self.archive}{path_sep}{modpath}'
+                spec = _bootstrap.ModuleSpec(name=fullname, loader=None,
+                                             is_package=True)
+                spec.submodule_search_locations.append(path)
+                return spec
+            else:
+                return None
 
     def get_code(self, fullname):
         """get_code(fullname) -> code object.
 
         Return the code object for the specified module. Raise ZipImportError
-        if the module couldn't be found.
+        if the module couldn't be imported.
         """
         code, ispackage, modpath = _get_module_code(self, fullname)
         return code
@@ -184,7 +221,8 @@ def get_data(self, pathname):
     def get_filename(self, fullname):
         """get_filename(fullname) -> filename string.
 
-        Return the filename for the specified module.
+        Return the filename for the specified module or raise ZipImportError
+        if it couldn't be imported.
         """
         # Deciding the filename requires working out where the code
         # would come from if the module was actually loaded
@@ -236,8 +274,13 @@ def load_module(self, fullname):
 
         Load the module specified by 'fullname'. 'fullname' must be the
         fully qualified (dotted) module name. It returns the imported
-        module, or raises ZipImportError if it wasn't found.
+        module, or raises ZipImportError if it could not be imported.
+
+        Deprecated since Python 3.10. Use exec_module() instead.
         """
+        msg = ("zipimport.zipimporter.load_module() is deprecated and slated for "
+               "removal in Python 3.12; use exec_module() instead")
+        _warnings.warn(msg, DeprecationWarning)
         code, ispackage, modpath = _get_module_code(self, fullname)
         mod = sys.modules.get(fullname)
         if mod is None or not isinstance(mod, _module_type):
@@ -280,11 +323,18 @@ def get_resource_reader(self, fullname):
                 return None
         except ZipImportError:
             return None
-        if not _ZipImportResourceReader._registered:
-            from importlib.abc import ResourceReader
-            ResourceReader.register(_ZipImportResourceReader)
-            _ZipImportResourceReader._registered = True
-        return _ZipImportResourceReader(self, fullname)
+        from importlib.readers import ZipReader
+        return ZipReader(self, fullname)
+
+
+    def invalidate_caches(self):
+        """Reload the file data of the archive path."""
+        try:
+            self._files = _read_directory(self.archive)
+            _zip_directory_cache[self.archive] = self._files
+        except ZipImportError:
+            _zip_directory_cache.pop(self.archive, None)
+            self._files = {}
 
 
     def __repr__(self):
@@ -580,20 +630,15 @@ def _eq_mtime(t1, t2):
 
 
 # Given the contents of a .py[co] file, unmarshal the data
-# and return the code object. Return None if it the magic word doesn't
-# match, or if the recorded .py[co] metadata does not match the source,
-# (we do this instead of raising an exception as we fall back
-# to .py if available and we don't want to mask other errors).
+# and return the code object. Raises ImportError it the magic word doesn't
+# match, or if the recorded .py[co] metadata does not match the source.
 def _unmarshal_code(self, pathname, fullpath, fullname, data):
     exc_details = {
         'name': fullname,
         'path': fullpath,
     }
 
-    try:
-        flags = _bootstrap_external._classify_pyc(data, fullname, exc_details)
-    except ImportError:
-        return None
+    flags = _bootstrap_external._classify_pyc(data, fullname, exc_details)
 
     hash_based = flags & 0b1 != 0
     if hash_based:
@@ -607,11 +652,8 @@ def _unmarshal_code(self, pathname, fullpath, fullname, data):
                     source_bytes,
                 )
 
-                try:
-                    _bootstrap_external._validate_hash_pyc(
-                        data, source_hash, fullname, exc_details)
-                except ImportError:
-                    return None
+                _bootstrap_external._validate_hash_pyc(
+                    data, source_hash, fullname, exc_details)
     else:
         source_mtime, source_size = \
             _get_mtime_and_size_of_source(self, fullpath)
@@ -697,6 +739,7 @@ def _get_pyc_source(self, path):
 # 'fullname'.
 def _get_module_code(self, fullname):
     path = _get_module_path(self, fullname)
+    import_error = None
     for suffix, isbytecode, ispackage in _zip_searchorder:
         fullpath = path + suffix
         _bootstrap._verbose_message('trying {}{}{}', self.archive, path_sep, fullpath, verbosity=2)
@@ -707,8 +750,12 @@ def _get_module_code(self, fullname):
         else:
             modpath = toc_entry[0]
             data = _get_data(self.archive, toc_entry)
+            code = None
             if isbytecode:
-                code = _unmarshal_code(self, modpath, fullpath, fullname, data)
+                try:
+                    code = _unmarshal_code(self, modpath, fullpath, fullname, data)
+                except ImportError as exc:
+                    import_error = exc
             else:
                 code = _compile_source(modpath, data)
             if code is None:
@@ -718,75 +765,8 @@ def _get_module_code(self, fullname):
             modpath = toc_entry[0]
             return code, ispackage, modpath
     else:
-        raise ZipImportError(f"can't find module {fullname!r}", name=fullname)
-
-
-class _ZipImportResourceReader:
-    """Private class used to support ZipImport.get_resource_reader().
-
-    This class is allowed to reference all the innards and private parts of
-    the zipimporter.
-    """
-    _registered = False
-
-    def __init__(self, zipimporter, fullname):
-        self.zipimporter = zipimporter
-        self.fullname = fullname
-
-    def open_resource(self, resource):
-        fullname_as_path = self.fullname.replace('.', '/')
-        path = f'{fullname_as_path}/{resource}'
-        from io import BytesIO
-        try:
-            return BytesIO(self.zipimporter.get_data(path))
-        except OSError:
-            raise FileNotFoundError(path)
-
-    def resource_path(self, resource):
-        # All resources are in the zip file, so there is no path to the file.
-        # Raising FileNotFoundError tells the higher level API to extract the
-        # binary data and create a temporary file.
-        raise FileNotFoundError
-
-    def is_resource(self, name):
-        # Maybe we could do better, but if we can get the data, it's a
-        # resource.  Otherwise it isn't.
-        fullname_as_path = self.fullname.replace('.', '/')
-        path = f'{fullname_as_path}/{name}'
-        try:
-            self.zipimporter.get_data(path)
-        except OSError:
-            return False
-        return True
-
-    def contents(self):
-        # This is a bit convoluted, because fullname will be a module path,
-        # but _files is a list of file names relative to the top of the
-        # archive's namespace.  We want to compare file paths to find all the
-        # names of things inside the module represented by fullname.  So we
-        # turn the module path of fullname into a file path relative to the
-        # top of the archive, and then we iterate through _files looking for
-        # names inside that "directory".
-        from pathlib import Path
-        fullname_path = Path(self.zipimporter.get_filename(self.fullname))
-        relative_path = fullname_path.relative_to(self.zipimporter.archive)
-        # Don't forget that fullname names a package, so its path will include
-        # __init__.py, which we want to ignore.
-        assert relative_path.name == '__init__.py'
-        package_path = relative_path.parent
-        subdirs_seen = set()
-        for filename in self.zipimporter._files:
-            try:
-                relative = Path(filename).relative_to(package_path)
-            except ValueError:
-                continue
-            # If the path of the file (which is relative to the top of the zip
-            # namespace), relative to the package given when the resource
-            # reader was created, has a parent, then it's a name in a
-            # subdirectory and thus we skip it.
-            parent_name = relative.parent.name
-            if len(parent_name) == 0:
-                yield relative.name
-            elif parent_name not in subdirs_seen:
-                subdirs_seen.add(parent_name)
-                yield parent_name
+        if import_error:
+            msg = f"module load failed: {import_error}"
+            raise ZipImportError(msg, name=fullname) from import_error
+        else:
+            raise ZipImportError(f"can't find module {fullname!r}", name=fullname)