Change -k EXPRESSION matching to be case-insensitive

src/_pytest/mark/legacy.py

@@ -57,7 +57,15 @@ class KeywordMapping:
         return cls(mapped_names)
 
     def __getitem__(self, subname):
-        for name in self._names:
+        """Return whether subname is included within stored names.
+
+        The string inclusion check is case-insensitive.
+
+        """
+        subname = subname.lower()
+        names = [name.lower() for name in self._names]
+
+        for name in names:
             if subname in name:
                 return True
         return False

testing/test_collection.py

@@ -809,6 +809,40 @@ class TestNodekeywords:
         reprec = testdir.inline_run("-k repr")
         reprec.assertoutcome(passed=1, failed=0)
 
+    def test_keyword_matching_is_case_insensitive_by_default(self, testdir):
+        """Check that selection via -k EXPRESSION is case-insensitive.
+
+        Since markers are also added to the node keywords, they too can
+        be matched without having to think about case sensitivity.
+
+        """
+        testdir.makepyfile(
+            """
+            import pytest
+
+            def test_sPeCiFiCToPiC_1():
+                assert True
+
+            class TestSpecificTopic_2:
+                def test(self):
+                    assert True
+
+            @pytest.mark.sPeCiFiCToPic_3
+            def test():
+                assert True
+
+            @pytest.mark.sPeCiFiCToPic_4
+            class Test:
+                def test(self):
+                    assert True
+
+        """
+        )
+        num_all_tests_passed = 4
+        for expression in ("specifictopic", "SPECIFICTOPIC", "SpecificTopic"):
+            reprec = testdir.inline_run("-k " + expression)
+            reprec.assertoutcome(passed=num_all_tests_passed, failed=0)
+
 
 COLLECTION_ERROR_PY_FILES = dict(
     test_01_failure="""