bpo-31324: Optimize support._match_test() (#4421)

* Rename support._match_test() to support.match_test(): make it
  public
* Remove support.match_tests global variable. It is replaced with a
  new support.set_match_tests() function, so match_test() doesn't
  have to check each time if patterns were modified.
* Rewrite match_test(): use different code paths depending on the
  kind of patterns for best performances.

Co-Authored-By: Serhiy Storchaka <storchaka@gmail.com>

4 files changed, 111 insertions, 15 deletions

diff --git a/Lib/test/libregrtest/main.py b/Lib/test/libregrtest/main.py
index 9871a28..ce01c8c 100644
--- a/Lib/test/libregrtest/main.py
+++ b/Lib/test/libregrtest/main.py
@@ -257,12 +257,12 @@ class Regrtest:
             if isinstance(test, unittest.TestSuite):
                 self._list_cases(test)
             elif isinstance(test, unittest.TestCase):
-                if support._match_test(test):
+                if support.match_test(test):
                     print(test.id())
 
     def list_cases(self):
         support.verbose = False
-        support.match_tests = self.ns.match_tests
+        support.set_match_tests(self.ns.match_tests)
 
         for test in self.selected:
             abstest = get_abs_module(self.ns, test)
diff --git a/Lib/test/libregrtest/runtest.py b/Lib/test/libregrtest/runtest.py
index dbd4634..12bf422 100644
--- a/Lib/test/libregrtest/runtest.py
+++ b/Lib/test/libregrtest/runtest.py
@@ -102,7 +102,7 @@ def runtest(ns, test):
     if use_timeout:
         faulthandler.dump_traceback_later(ns.timeout, exit=True)
     try:
-        support.match_tests = ns.match_tests
+        support.set_match_tests(ns.match_tests)
         # reset the environment_altered flag to detect if a test altered
         # the environment
         support.environment_altered = False
diff --git a/Lib/test/support/__init__.py b/Lib/test/support/__init__.py
index 527cf7f..71d9c2c 100644
--- a/Lib/test/support/__init__.py
+++ b/Lib/test/support/__init__.py
@@ -278,7 +278,6 @@ max_memuse = 0           # Disable bigmem tests (they will still be run with
                          # small sizes, to make sure they work.)
 real_max_memuse = 0
 failfast = False
-match_tests = None
 
 # _original_stdout is meant to hold stdout at the time regrtest began.
 # This may be "the real" stdout, or IDLE's emulation of stdout, or whatever.
@@ -1900,21 +1899,65 @@ def _run_suite(suite):
         raise TestFailed(err)
 
 
-def _match_test(test):
-    global match_tests
+# By default, don't filter tests
+_match_test_func = None
+_match_test_patterns = None
 
-    if match_tests is None:
+
+def match_test(test):
+    # Function used by support.run_unittest() and regrtest --list-cases
+    if _match_test_func is None:
         return True
-    test_id = test.id()
+    else:
+        return _match_test_func(test.id())
 
-    for match_test in match_tests:
-        if fnmatch.fnmatchcase(test_id, match_test):
-            return True
 
-        for name in test_id.split("."):
-            if fnmatch.fnmatchcase(name, match_test):
+def _is_full_match_test(pattern):
+    # If a pattern contains at least one dot, it's considered
+    # as a full test identifier.
+    # Example: 'test.test_os.FileTests.test_access'.
+    #
+    # Reject patterns which contain fnmatch patterns: '*', '?', '[...]'
+    # or '[!...]'. For example, reject 'test_access*'.
+    return ('.' in pattern) and (not re.search(r'[?*\[\]]', pattern))
+
+
+def set_match_tests(patterns):
+    global _match_test_func, _match_test_patterns
+
+    if patterns == _match_test_patterns:
+        # No change: no need to recompile patterns.
+        return
+
+    if not patterns:
+        func = None
+    elif all(map(_is_full_match_test, patterns)):
+        # Simple case: all patterns are full test identifier.
+        # The test.bisect utility only uses such full test identifiers.
+        func = set(patterns).__contains__
+    else:
+        regex = '|'.join(map(fnmatch.translate, patterns))
+        # The search *is* case sensitive on purpose:
+        # don't use flags=re.IGNORECASE
+        regex_match = re.compile(regex).match
+
+        def match_test_regex(test_id):
+            if regex_match(test_id):
+                # The regex matchs the whole identifier like
+                # 'test.test_os.FileTests.test_access'
                 return True
-    return False
+            else:
+                # Try to match parts of the test identifier.
+                # For example, split 'test.test_os.FileTests.test_access'
+                # into: 'test', 'test_os', 'FileTests' and 'test_access'.
+                return any(map(regex_match, test_id.split(".")))
+
+        func = match_test_regex
+
+    # Create a copy since patterns can be mutable and so modified later
+    _match_test_patterns = tuple(patterns)
+    _match_test_func = func
+
 
 
 def run_unittest(*classes):
@@ -1931,7 +1974,7 @@ def run_unittest(*classes):
             suite.addTest(cls)
         else:
             suite.addTest(unittest.makeSuite(cls))
-    _filter_suite(suite, _match_test)
+    _filter_suite(suite, match_test)
     _run_suite(suite)
 
 #=======================================================================
diff --git a/Lib/test/test_support.py b/Lib/test/test_support.py
index 4a577ef..4756def 100644
--- a/Lib/test/test_support.py
+++ b/Lib/test/test_support.py
@@ -483,6 +483,59 @@ class TestSupport(unittest.TestCase):
             with self.subTest(opts=opts):
                 self.check_options(opts, 'optim_args_from_interpreter_flags')
 
+    def test_match_test(self):
+        class Test:
+            def __init__(self, test_id):
+                self.test_id = test_id
+
+            def id(self):
+                return self.test_id
+
+        test_access = Test('test.test_os.FileTests.test_access')
+        test_chdir = Test('test.test_os.Win32ErrorTests.test_chdir')
+
+        with support.swap_attr(support, '_match_test_func', None):
+            # match all
+            support.set_match_tests([])
+            self.assertTrue(support.match_test(test_access))
+            self.assertTrue(support.match_test(test_chdir))
+
+            # match the full test identifier
+            support.set_match_tests([test_access.id()])
+            self.assertTrue(support.match_test(test_access))
+            self.assertFalse(support.match_test(test_chdir))
+
+            # match the module name
+            support.set_match_tests(['test_os'])
+            self.assertTrue(support.match_test(test_access))
+            self.assertTrue(support.match_test(test_chdir))
+
+            # Test '*' pattern
+            support.set_match_tests(['test_*'])
+            self.assertTrue(support.match_test(test_access))
+            self.assertTrue(support.match_test(test_chdir))
+
+            # Test case sensitivity
+            support.set_match_tests(['filetests'])
+            self.assertFalse(support.match_test(test_access))
+            support.set_match_tests(['FileTests'])
+            self.assertTrue(support.match_test(test_access))
+
+            # Test pattern containing '.' and a '*' metacharacter
+            support.set_match_tests(['*test_os.*.test_*'])
+            self.assertTrue(support.match_test(test_access))
+            self.assertTrue(support.match_test(test_chdir))
+
+            # Multiple patterns
+            support.set_match_tests([test_access.id(), test_chdir.id()])
+            self.assertTrue(support.match_test(test_access))
+            self.assertTrue(support.match_test(test_chdir))
+
+            support.set_match_tests(['test_access', 'DONTMATCH'])
+            self.assertTrue(support.match_test(test_access))
+            self.assertFalse(support.match_test(test_chdir))
+
+
     # XXX -follows a list of untested API
     # make_legacy_pyc
     # is_resource_enabled