Merge pull request #437 from jherland/egg-metadata-work

Fix .files and inferred packages_distributions for .egg-info packages

Author: jaraco

CHANGES.rst

@@ -1,3 +1,9 @@
+v6.3.0
+======
+
+* #115: Support ``installed-files.txt`` for ``Distribution.files``
+  when present.
+
 v6.2.1
 ======
 

importlib_metadata/__init__.py

@@ -12,6 +12,7 @@
 import functools
 import itertools
 import posixpath
+import contextlib
 import collections
 import inspect
 
@@ -461,8 +462,8 @@ def files(self):
         :return: List of PackagePath for this distribution or None
 
         Result is `None` if the metadata file that enumerates files
-        (i.e. RECORD for dist-info or SOURCES.txt for egg-info) is
-        missing.
+        (i.e. RECORD for dist-info, or installed-files.txt or
+        SOURCES.txt for egg-info) is missing.
         Result may be empty if the metadata exists but is empty.
         """
 
@@ -475,9 +476,19 @@ def make_file(name, hash=None, size_str=None):
 
         @pass_none
         def make_files(lines):
-            return list(starmap(make_file, csv.reader(lines)))
+            return starmap(make_file, csv.reader(lines))
 
-        return make_files(self._read_files_distinfo() or self._read_files_egginfo())
+        @pass_none
+        def skip_missing_files(package_paths):
+            return list(filter(lambda path: path.locate().exists(), package_paths))
+
+        return skip_missing_files(
+            make_files(
+                self._read_files_distinfo()
+                or self._read_files_egginfo_installed()
+                or self._read_files_egginfo_sources()
+            )
+        )
 
     def _read_files_distinfo(self):
         """
@@ -486,10 +497,43 @@ def _read_files_distinfo(self):
         text = self.read_text('RECORD')
         return text and text.splitlines()
 
-    def _read_files_egginfo(self):
+    def _read_files_egginfo_installed(self):
         """
-        SOURCES.txt might contain literal commas, so wrap each line
-        in quotes.
+        Read installed-files.txt and return lines in a similar
+        CSV-parsable format as RECORD: each file must be placed
+        relative to the site-packages directory, and must also be
+        quoted (since file names can contain literal commas).
+
+        This file is written when the package is installed by pip,
+        but it might not be written for other installation methods.
+        Hence, even if we can assume that this file is accurate
+        when it exists, we cannot assume that it always exists.
+        """
+        text = self.read_text('installed-files.txt')
+        # We need to prepend the .egg-info/ subdir to the lines in this file.
+        # But this subdir is only available in the PathDistribution's self._path
+        # which is not easily accessible from this base class...
+        subdir = getattr(self, '_path', None)
+        if not text or not subdir:
+            return
+        with contextlib.suppress(Exception):
+            ret = [
+                str((subdir / line).resolve().relative_to(self.locate_file('')))
+                for line in text.splitlines()
+            ]
+            return map('"{}"'.format, ret)
+
+    def _read_files_egginfo_sources(self):
+        """
+        Read SOURCES.txt and return lines in a similar CSV-parsable
+        format as RECORD: each file name must be quoted (since it
+        might contain literal commas).
+
+        Note that SOURCES.txt is not a reliable source for what
+        files are installed by a package. This file is generated
+        for a source archive, and the files that are present
+        there (e.g. setup.py) may not correctly reflect the files
+        that are present after the package has been installed.
         """
         text = self.read_text('SOURCES.txt')
         return text and map('"{}"'.format, text.splitlines())

tests/fixtures.py

@@ -86,8 +86,10 @@ def setUp(self):
 
 
 # Except for python/mypy#731, prefer to define
-# FilesDef = Dict[str, Union['FilesDef', str]]
-FilesDef = Dict[str, Union[Dict[str, Union[Dict[str, str], str]], str]]
+# FilesDef = Dict[str, Union['FilesDef', str, bytes]]
+FilesDef = Dict[
+    str, Union[Dict[str, Union[Dict[str, Union[str, bytes]], str, bytes]], str, bytes]
+]
 
 
 class DistInfoPkg(OnSysPath, SiteDir):
@@ -214,6 +216,97 @@ def setUp(self):
         build_files(EggInfoPkg.files, prefix=self.site_dir)
 
 
+class EggInfoPkgPipInstalledNoToplevel(OnSysPath, SiteDir):
+    files: FilesDef = {
+        "egg_with_module_pkg.egg-info": {
+            "PKG-INFO": "Name: egg_with_module-pkg",
+            # SOURCES.txt is made from the source archive, and contains files
+            # (setup.py) that are not present after installation.
+            "SOURCES.txt": """
+                egg_with_module.py
+                setup.py
+                egg_with_module_pkg.egg-info/PKG-INFO
+                egg_with_module_pkg.egg-info/SOURCES.txt
+                egg_with_module_pkg.egg-info/top_level.txt
+            """,
+            # installed-files.txt is written by pip, and is a strictly more
+            # accurate source than SOURCES.txt as to the installed contents of
+            # the package.
+            "installed-files.txt": """
+                ../egg_with_module.py
+                PKG-INFO
+                SOURCES.txt
+                top_level.txt
+            """,
+            # missing top_level.txt (to trigger fallback to installed-files.txt)
+        },
+        "egg_with_module.py": """
+            def main():
+                print("hello world")
+            """,
+    }
+
+    def setUp(self):
+        super().setUp()
+        build_files(EggInfoPkgPipInstalledNoToplevel.files, prefix=self.site_dir)
+
+
+class EggInfoPkgPipInstalledNoModules(OnSysPath, SiteDir):
+    files: FilesDef = {
+        "egg_with_no_modules_pkg.egg-info": {
+            "PKG-INFO": "Name: egg_with_no_modules-pkg",
+            # SOURCES.txt is made from the source archive, and contains files
+            # (setup.py) that are not present after installation.
+            "SOURCES.txt": """
+                setup.py
+                egg_with_no_modules_pkg.egg-info/PKG-INFO
+                egg_with_no_modules_pkg.egg-info/SOURCES.txt
+                egg_with_no_modules_pkg.egg-info/top_level.txt
+            """,
+            # installed-files.txt is written by pip, and is a strictly more
+            # accurate source than SOURCES.txt as to the installed contents of
+            # the package.
+            "installed-files.txt": """
+                PKG-INFO
+                SOURCES.txt
+                top_level.txt
+            """,
+            # top_level.txt correctly reflects that no modules are installed
+            "top_level.txt": b"\n",
+        },
+    }
+
+    def setUp(self):
+        super().setUp()
+        build_files(EggInfoPkgPipInstalledNoModules.files, prefix=self.site_dir)
+
+
+class EggInfoPkgSourcesFallback(OnSysPath, SiteDir):
+    files: FilesDef = {
+        "sources_fallback_pkg.egg-info": {
+            "PKG-INFO": "Name: sources_fallback-pkg",
+            # SOURCES.txt is made from the source archive, and contains files
+            # (setup.py) that are not present after installation.
+            "SOURCES.txt": """
+                sources_fallback.py
+                setup.py
+                sources_fallback_pkg.egg-info/PKG-INFO
+                sources_fallback_pkg.egg-info/SOURCES.txt
+            """,
+            # missing installed-files.txt (i.e. not installed by pip) and
+            # missing top_level.txt (to trigger fallback to SOURCES.txt)
+        },
+        "sources_fallback.py": """
+            def main():
+                print("hello world")
+            """,
+    }
+
+    def setUp(self):
+        super().setUp()
+        build_files(EggInfoPkgSourcesFallback.files, prefix=self.site_dir)
+
+
 class EggInfoFile(OnSysPath, SiteDir):
     files: FilesDef = {
         "egginfo_file.egg-info": """

tests/test_api.py

@@ -27,6 +27,9 @@ def suppress_known_deprecation():
 
 class APITests(
     fixtures.EggInfoPkg,
+    fixtures.EggInfoPkgPipInstalledNoToplevel,
+    fixtures.EggInfoPkgPipInstalledNoModules,
+    fixtures.EggInfoPkgSourcesFallback,
     fixtures.DistInfoPkg,
     fixtures.DistInfoPkgWithDot,
     fixtures.EggInfoFile,
@@ -62,15 +65,28 @@ def test_prefix_not_matched(self):
                     distribution(prefix)
 
     def test_for_top_level(self):
-        self.assertEqual(
-            distribution('egginfo-pkg').read_text('top_level.txt').strip(), 'mod'
-        )
+        tests = [
+            ('egginfo-pkg', 'mod'),
+            ('egg_with_no_modules-pkg', ''),
+        ]
+        for pkg_name, expect_content in tests:
+            with self.subTest(pkg_name):
+                self.assertEqual(
+                    distribution(pkg_name).read_text('top_level.txt').strip(),
+                    expect_content,
+                )
 
     def test_read_text(self):
-        top_level = [
-            path for path in files('egginfo-pkg') if path.name == 'top_level.txt'
-        ][0]
-        self.assertEqual(top_level.read_text(), 'mod\n')
+        tests = [
+            ('egginfo-pkg', 'mod\n'),
+            ('egg_with_no_modules-pkg', '\n'),
+        ]
+        for pkg_name, expect_content in tests:
+            with self.subTest(pkg_name):
+                top_level = [
+                    path for path in files(pkg_name) if path.name == 'top_level.txt'
+                ][0]
+                self.assertEqual(top_level.read_text(), expect_content)
 
     def test_entry_points(self):
         eps = entry_points()
@@ -184,6 +200,9 @@ def test_files_dist_info(self):
 
     def test_files_egg_info(self):
         self._test_files(files('egginfo-pkg'))
+        self._test_files(files('egg_with_module-pkg'))
+        self._test_files(files('egg_with_no_modules-pkg'))
+        self._test_files(files('sources_fallback-pkg'))
 
     def test_version_egg_info_file(self):
         self.assertEqual(version('egginfo-file'), '0.1')

tests/test_main.py

@@ -171,11 +171,21 @@ def test_metadata_loads_egg_info(self):
         assert meta['Description'] == 'pôrˈtend'
 
 
-class DiscoveryTests(fixtures.EggInfoPkg, fixtures.DistInfoPkg, unittest.TestCase):
+class DiscoveryTests(
+    fixtures.EggInfoPkg,
+    fixtures.EggInfoPkgPipInstalledNoToplevel,
+    fixtures.EggInfoPkgPipInstalledNoModules,
+    fixtures.EggInfoPkgSourcesFallback,
+    fixtures.DistInfoPkg,
+    unittest.TestCase,
+):
     def test_package_discovery(self):
         dists = list(distributions())
         assert all(isinstance(dist, Distribution) for dist in dists)
         assert any(dist.metadata['Name'] == 'egginfo-pkg' for dist in dists)
+        assert any(dist.metadata['Name'] == 'egg_with_module-pkg' for dist in dists)
+        assert any(dist.metadata['Name'] == 'egg_with_no_modules-pkg' for dist in dists)
+        assert any(dist.metadata['Name'] == 'sources_fallback-pkg' for dist in dists)
         assert any(dist.metadata['Name'] == 'distinfo-pkg' for dist in dists)
 
     def test_invalid_usage(self):
@@ -362,3 +372,40 @@ def test_packages_distributions_all_module_types(self):
             assert distributions[f'in_package_{i}'] == ['all_distributions']
 
         assert not any(name.endswith('.dist-info') for name in distributions)
+
+
+class PackagesDistributionsEggTest(
+    fixtures.EggInfoPkg,
+    fixtures.EggInfoPkgPipInstalledNoToplevel,
+    fixtures.EggInfoPkgPipInstalledNoModules,
+    fixtures.EggInfoPkgSourcesFallback,
+    unittest.TestCase,
+):
+    def test_packages_distributions_on_eggs(self):
+        """
+        Test old-style egg packages with a variation of 'top_level.txt',
+        'SOURCES.txt', and 'installed-files.txt', available.
+        """
+        distributions = packages_distributions()
+
+        def import_names_from_package(package_name):
+            return {
+                import_name
+                for import_name, package_names in distributions.items()
+                if package_name in package_names
+            }
+
+        # egginfo-pkg declares one import ('mod') via top_level.txt
+        assert import_names_from_package('egginfo-pkg') == {'mod'}
+
+        # egg_with_module-pkg has one import ('egg_with_module') inferred from
+        # installed-files.txt (top_level.txt is missing)
+        assert import_names_from_package('egg_with_module-pkg') == {'egg_with_module'}
+
+        # egg_with_no_modules-pkg should not be associated with any import names
+        # (top_level.txt is empty, and installed-files.txt has no .py files)
+        assert import_names_from_package('egg_with_no_modules-pkg') == set()
+
+        # sources_fallback-pkg has one import ('sources_fallback') inferred from
+        # SOURCES.txt (top_level.txt and installed-files.txt is missing)
+        assert import_names_from_package('sources_fallback-pkg') == {'sources_fallback'}