fix: don't measure all third-party packages if source is in third-party location

There is logic to not measure third-party packages inside configured sources. However, when a (i.e. another) configured source was inside a third-party location, this logic was previously disabled completely.

This caused a problem if a virtual env is set up inside a configured source directory and a configured source package gets installed inside the virtual env. Previously in this case, coverage was measured for all files in the virtual env for the reason described in the previous paragraph.

This commit changes the code to collect all configured source directories inside third-party locations and disable coverage for code in third-party locations only if its not in one of these collected source directories.

Author: manueljacob

CHANGES.rst

@@ -20,6 +20,12 @@ development at the same time, such as 4.5.x and 5.0.
 Unreleased
 ----------
 
+- Fix: if a virtualenv was created inside a source directory, and a sourced
+  package was installed inside the virtualenv, then all of the third-party
+  packages inside the virtualenv would be measured.  This was incorrect, but
+  has now been fixed: only the specified packages will be measured, thanks to
+  `Manuel Jacob <pull 1560_>`_.
+
 - Fix: the ``coverage lcov`` command could create a .lcov file with incorrect
   LF (lines found) and LH (lines hit) totals.  This is now fixed, thanks to
   `Ian Moore <pull 1583_>`_.
@@ -28,6 +34,7 @@ Unreleased
   duplicate ``<package>`` elements. This is now fixed, thanks to `Benjamin
   Parzella <pull 1574_>`_, closing `issue 1573`_.
 
+.. _pull 1560: https://github.com/nedbat/coveragepy/pull/1560
 .. _issue 1573: https://github.com/nedbat/coveragepy/issues/1573
 .. _pull 1574: https://github.com/nedbat/coveragepy/pull/1574
 .. _pull 1583: https://github.com/nedbat/coveragepy/pull/1583

CONTRIBUTORS.txt

@@ -107,6 +107,7 @@ Leonardo Pistone
 Lex Berezhny
 Loïc Dachary
 Lorenzo Micò
+Manuel Jacob
 Marc Abramowitz
 Marc Legendre
 Marcelo Trylesinski

coverage/inorout.py

@@ -262,7 +262,7 @@ def _debug(msg: str) -> None:
         # Check if the source we want to measure has been installed as a
         # third-party package.
         # Is the source inside a third-party area?
-        self.source_in_third = False
+        self.source_in_third_paths = set()
         with sys_modules_saved():
             for pkg in self.source_pkgs:
                 try:
@@ -274,22 +274,23 @@ def _debug(msg: str) -> None:
                 if modfile:
                     if self.third_match.match(modfile):
                         _debug(
-                            f"Source is in third-party because of source_pkg {pkg!r} at {modfile!r}"
+                            f"Source in third-party: source_pkg {pkg!r} at {modfile!r}"
                         )
-                        self.source_in_third = True
+                        self.source_in_third_paths.add(canonical_path(source_for_file(modfile)))
                 else:
                     for pathdir in path:
                         if self.third_match.match(pathdir):
                             _debug(
-                                f"Source is in third-party because of {pkg!r} path directory " +
-                                f"at {pathdir!r}"
+                                f"Source in third-party: {pkg!r} path directory at {pathdir!r}"
                             )
-                            self.source_in_third = True
+                            self.source_in_third_paths.add(pathdir)
 
         for src in self.source:
             if self.third_match.match(src):
-                _debug(f"Source is in third-party because of source directory {src!r}")
-                self.source_in_third = True
+                _debug(f"Source in third-party: source directory {src!r}")
+                self.source_in_third_paths.add(src)
+        self.source_in_third_match = TreeMatcher(self.source_in_third_paths, "source_in_third")
+        _debug(f"Source in third-party matching: {self.source_in_third_match}")
 
         self.plugins: Plugins
         self.disp_class: Type[TFileDisposition] = FileDisposition
@@ -419,9 +420,8 @@ def check_include_omit_etc(self, filename: str, frame: Optional[FrameType]) -> O
                     ok = True
             if not ok:
                 return extra + "falls outside the --source spec"
-            if not self.source_in_third:
-                if self.third_match.match(filename):
-                    return "inside --source, but is third-party"
+            if self.third_match.match(filename) and not self.source_in_third_match.match(filename):
+                return "inside --source, but is third-party"
         elif self.include_match:
             if not self.include_match.match(filename):
                 return "falls outside the --include trees"
@@ -576,12 +576,13 @@ def sys_info(self) -> Iterable[Tuple[str, Any]]:
             ("coverage_paths", self.cover_paths),
             ("stdlib_paths", self.pylib_paths),
             ("third_party_paths", self.third_paths),
+            ("source_in_third_party_paths", self.source_in_third_paths),
         ]
 
         matcher_names = [
             'source_match', 'source_pkgs_match',
             'include_match', 'omit_match',
-            'cover_match', 'pylib_match', 'third_match',
+            'cover_match', 'pylib_match', 'third_match', 'source_in_third_match',
         ]
 
         for matcher_name in matcher_names:

tests/test_venv.py

@@ -198,8 +198,28 @@ def get_trace_output(self) -> str:
         with open("debug_out.txt") as f:
             return f.read()
 
-    def test_third_party_venv_isnt_measured(self, coverage_command: str) -> None:
-        out = run_in_venv(coverage_command + " run --source=. myproduct.py")
+    @pytest.mark.parametrize('install_source_in_venv', [True, False])
+    def test_third_party_venv_isnt_measured(
+        self, coverage_command: str, install_source_in_venv: bool
+    ) -> None:
+        if install_source_in_venv:
+            make_file("setup.py", """\
+                import setuptools
+                setuptools.setup(
+                    name="myproduct",
+                    py_modules = ["myproduct"],
+                )
+                """)
+            try:
+                run_in_venv("python -m pip install .")
+            finally:
+                shutil.rmtree("build", ignore_errors=True)
+                shutil.rmtree("myproduct.egg-info", ignore_errors=True)
+            # Ensure that coverage doesn't run the non-installed module.
+            os.remove('myproduct.py')
+            out = run_in_venv(coverage_command + " run --source=.,myproduct -m myproduct")
+        else:
+            out = run_in_venv(coverage_command + " run --source=. myproduct.py")
         # In particular, this warning doesn't appear:
         # Already imported a file that will be measured: .../coverage/__main__.py
         assert out == self.expected_stdout
@@ -213,7 +233,7 @@ def test_third_party_venv_isnt_measured(self, coverage_command: str) -> None:
         )
         assert re_lines(r"^Tracing .*\bmyproduct.py", debug_out)
         assert re_lines(
-            r"^Not tracing .*\bcolorsys.py': falls outside the --source spec",
+            r"^Not tracing .*\bcolorsys.py': (module 'colorsys' |)?falls outside the --source spec",
             debug_out,
         )