Merge branch 'master' into feature/pep517-metadata

Author: jaraco

.gitlab-ci.yml

@@ -22,6 +22,10 @@ coverage:
     paths:
       - coverage.xml
 
+benchmark:
+  script:
+    - tox -e perf
+
 diffcov:
   script:
     - tox -e py27-diffcov,py35-diffcov,py36-diffcov,py37-diffcov,py38-diffcov

importlib_metadata/__init__.py

@@ -10,6 +10,7 @@
 import operator
 import functools
 import itertools
+import posixpath
 import collections
 
 from ._compat import (
@@ -23,11 +24,12 @@
     NotADirectoryError,
     PermissionError,
     pathlib,
-    PYPY_OPEN_BUG,
     ModuleNotFoundError,
     MetaPathFinder,
     email_message_from_string,
     PyPy_repr,
+    unique_ordered,
+    str,
     )
 from importlib import import_module
 from itertools import starmap
@@ -95,6 +97,16 @@ def load(self):
         attrs = filter(None, (match.group('attr') or '').split('.'))
         return functools.reduce(getattr, attrs, module)
 
+    @property
+    def module(self):
+        match = self.pattern.match(self.value)
+        return match.group('module')
+
+    @property
+    def attr(self):
+        match = self.pattern.match(self.value)
+        return match.group('attr')
+
     @property
     def extras(self):
         match = self.pattern.match(self.value)
@@ -400,10 +412,6 @@ def path(self):
             """
             return vars(self).get('path', sys.path)
 
-        @property
-        def pattern(self):
-            return '.*' if self.name is None else re.escape(self.name)
-
     @abc.abstractmethod
     def find_distributions(self, context=Context()):
         """
@@ -415,6 +423,75 @@ def find_distributions(self, context=Context()):
         """
 
 
+class FastPath:
+    """
+    Micro-optimized class for searching a path for
+    children.
+    """
+
+    def __init__(self, root):
+        self.root = str(root)
+        self.base = os.path.basename(self.root).lower()
+
+    def joinpath(self, child):
+        return pathlib.Path(self.root, child)
+
+    def children(self):
+        with suppress(Exception):
+            return os.listdir(self.root or '')
+        with suppress(Exception):
+            return self.zip_children()
+        return []
+
+    def zip_children(self):
+        zip_path = zipp.Path(self.root)
+        names = zip_path.root.namelist()
+        self.joinpath = zip_path.joinpath
+
+        return unique_ordered(
+            child.split(posixpath.sep, 1)[0]
+            for child in names
+            )
+
+    def is_egg(self, search):
+        base = self.base
+        return (
+            base == search.versionless_egg_name
+            or base.startswith(search.prefix)
+            and base.endswith('.egg'))
+
+    def search(self, name):
+        for child in self.children():
+            n_low = child.lower()
+            if (n_low in name.exact_matches
+                    or n_low.startswith(name.prefix)
+                    and n_low.endswith(name.suffixes)
+                    # legacy case:
+                    or self.is_egg(name) and n_low == 'egg-info'):
+                yield self.joinpath(child)
+
+
+class Prepared:
+    """
+    A prepared search for metadata on a possibly-named package.
+    """
+    normalized = ''
+    prefix = ''
+    suffixes = '.dist-info', '.egg-info'
+    exact_matches = [''][:0]
+    versionless_egg_name = ''
+
+    def __init__(self, name):
+        self.name = name
+        if name is None:
+            return
+        self.normalized = name.lower().replace('-', '_')
+        self.prefix = self.normalized + '-'
+        self.exact_matches = [
+            self.normalized + suffix for suffix in self.suffixes]
+        self.versionless_egg_name = self.normalized + '.egg'
+
+
 @install
 class MetadataPathFinder(NullFinder, DistributionFinder):
     """A degenerate finder for distribution packages on the file system.
@@ -432,45 +509,17 @@ def find_distributions(self, context=DistributionFinder.Context()):
         (or all names if ``None`` indicated) along the paths in the list
         of directories ``context.path``.
         """
-        found = self._search_paths(context.pattern, context.path)
+        found = self._search_paths(context.name, context.path)
         return map(PathDistribution, found)
 
     @classmethod
-    def _search_paths(cls, pattern, paths):
+    def _search_paths(cls, name, paths):
         """Find metadata directories in paths heuristically."""
         return itertools.chain.from_iterable(
-            cls._search_path(path, pattern)
-            for path in map(cls._switch_path, paths)
+            path.search(Prepared(name))
+            for path in map(FastPath, paths)
             )
 
-    @staticmethod
-    def _switch_path(path):
-        if not PYPY_OPEN_BUG or os.path.isfile(path):  # pragma: no branch
-            with suppress(Exception):
-                return zipp.Path(path)
-        return pathlib.Path(path)
-
-    @classmethod
-    def _matches_info(cls, normalized, item):
-        template = r'{pattern}(-.*)?\.(dist|egg)-info'
-        manifest = template.format(pattern=normalized)
-        return re.match(manifest, item.name, flags=re.IGNORECASE)
-
-    @classmethod
-    def _matches_legacy(cls, normalized, item):
-        template = r'{pattern}-.*\.egg[\\/]EGG-INFO'
-        manifest = template.format(pattern=normalized)
-        return re.search(manifest, str(item), flags=re.IGNORECASE)
-
-    @classmethod
-    def _search_path(cls, root, pattern):
-        if not root.is_dir():
-            return ()
-        normalized = pattern.replace('-', '_')
-        return (item for item in root.iterdir()
-                if cls._matches_info(normalized, item)
-                or cls._matches_legacy(normalized, item))
-
 
 class PathDistribution(Distribution):
     def __init__(self, path):

importlib_metadata/_compat.py

@@ -1,4 +1,4 @@
-from __future__ import absolute_import
+from __future__ import absolute_import, unicode_literals
 
 import io
 import abc
@@ -9,21 +9,27 @@
 if sys.version_info > (3,):  # pragma: nocover
     import builtins
     from configparser import ConfigParser
-    from contextlib import suppress
+    import contextlib
     FileNotFoundError = builtins.FileNotFoundError
     IsADirectoryError = builtins.IsADirectoryError
     NotADirectoryError = builtins.NotADirectoryError
     PermissionError = builtins.PermissionError
     map = builtins.map
+    from itertools import filterfalse
 else:  # pragma: nocover
     from backports.configparser import ConfigParser
     from itertools import imap as map  # type: ignore
-    from contextlib2 import suppress  # noqa
+    from itertools import ifilterfalse as filterfalse
+    import contextlib2 as contextlib
     FileNotFoundError = IOError, OSError
     IsADirectoryError = IOError, OSError
     NotADirectoryError = IOError, OSError
     PermissionError = IOError, OSError
 
+str = type('')
+
+suppress = contextlib.suppress
+
 if sys.version_info > (3, 5):  # pragma: nocover
     import pathlib
 else:  # pragma: nocover
@@ -73,7 +79,7 @@ def disable_stdlib_finder():
     """
     def matches(finder):
         return (
-            finder.__module__ == '_frozen_importlib_external'
+            getattr(finder, '__module__', None) == '_frozen_importlib_external'
             and hasattr(finder, 'find_distributions')
             )
     for finder in filter(matches, sys.meta_path):  # pragma: nocover
@@ -111,9 +117,6 @@ def py2_message_from_string(text):  # nocoverpy3
     email.message_from_string
     )
 
-# https://bitbucket.org/pypy/pypy/issues/3021/ioopen-directory-leaks-a-file-descriptor
-PYPY_OPEN_BUG = getattr(sys, 'pypy_version_info', (9, 9, 9))[:3] <= (7, 1, 1)
-
 
 class PyPy_repr:
     """
@@ -132,3 +135,18 @@ def make_param(name):
     if affected:  # pragma: nocover
         __repr__ = __compat_repr__
     del affected
+
+
+# from itertools recipes
+def unique_everseen(iterable):  # pragma: nocover
+    "List unique elements, preserving order. Remember all elements ever seen."
+    seen = set()
+    seen_add = seen.add
+
+    for element in filterfalse(seen.__contains__, iterable):
+        seen_add(element)
+        yield element
+
+
+unique_ordered = (
+    unique_everseen if sys.version_info < (3, 7) else dict.fromkeys)

importlib_metadata/docs/changelog.rst

@@ -2,6 +2,51 @@
  importlib_metadata NEWS
 =========================
 
+v1.6.1
+======
+
+* Ensure inputs to FastPath are Unicode. Closes #121.
+* Tests now rely on ``importlib.resources.files`` (and
+  backport) instead of the older ``path`` function.
+
+v1.6.0
+======
+
+* Added ``module`` and ``attr`` attributes to ``EntryPoint``
+
+v1.5.2
+======
+
+* Fix redundant entries from ``FastPath.zip_children``.
+  Closes #117.
+
+v1.5.1
+======
+
+* Improve reliability and consistency of compatibility
+  imports for contextlib and pathlib when running tests.
+  Closes #116.
+
+v1.5.0
+======
+
+* Additional performance optimizations in FastPath now
+  saves an additional 20% on a typical call.
+* Correct for issue where PyOxidizer finder has no
+  ``__module__`` attribute. Closes #110.
+
+v1.4.0
+======
+
+* Through careful optimization, ``distribution()`` is
+  3-4x faster. Thanks to Antony Lee for the
+  contribution. Closes #95.
+
+* When searching through ``sys.path``, if any error
+  occurs attempting to list a path entry, that entry
+  is skipped, making the system much more lenient
+  to errors. Closes #94.
+
 v1.3.0
 ======
 

importlib_metadata/docs/using.rst

@@ -70,7 +70,9 @@ Entry points
 The ``entry_points()`` function returns a dictionary of all entry points,
 keyed by group.  Entry points are represented by ``EntryPoint`` instances;
 each ``EntryPoint`` has a ``.name``, ``.group``, and ``.value`` attributes and
-a ``.load()`` method to resolve the value::
+a ``.load()`` method to resolve the value.  There are also ``.module``,
+``.attr``, and ``.extras`` attributes for getting the components of the
+``.value`` attribute::
 
     >>> eps = entry_points()
     >>> list(eps)
@@ -79,6 +81,12 @@ a ``.load()`` method to resolve the value::
     >>> wheel = [ep for ep in scripts if ep.name == 'wheel'][0]
     >>> wheel
     EntryPoint(name='wheel', value='wheel.cli:main', group='console_scripts')
+    >>> wheel.module
+    'wheel.cli'
+    >>> wheel.attr
+    'main'
+    >>> wheel.extras
+    []
     >>> main = wheel.load()
     >>> main
     <function main at 0x103528488>
@@ -87,7 +95,7 @@ The ``group`` and ``name`` are arbitrary values defined by the package author
 and usually a client will wish to resolve all entry points for a particular
 group.  Read `the setuptools docs
 <https://setuptools.readthedocs.io/en/latest/setuptools.html#dynamic-discovery-of-services-and-plugins>`_
-for more information on entrypoints, their definition, and usage.
+for more information on entry points, their definition, and usage.
 
 
 .. _metadata: