[libc] Implement str{,n}casecmp

[sivachandra]

libc/src/string/strcasecmp.cpp

19

This add a symbol level dependence on tolower. To avoid it, add a tolower implementation here: https://github.com/llvm/llvm-project/blob/main/libc/src/__support/ctype_utils.h.

libc/test/src/string/strcasecmp_test.cpp

29

Does comparing with the exact difference tightly couple this test with implementation details?

[abrachet]

libc/test/src/string/strcasecmp_test.cpp

29

Does comparing with the exact difference tightly couple this test with implementation details?

Yes, FWIW I copied these verbatim from the strcmp tests. But I've changed the test here to use LT or GT relative to 0.

libc/config/baremetal/entrypoints.txt

set(TARGET_LIBC_ENTRYPOINTS
     libc.src.string.memcmp
     libc.src.string.memcpy
     libc.src.string.memmove
     libc.src.string.mempcpy
     libc.src.string.memrchr
     libc.src.string.memset
     libc.src.string.stpcpy
     libc.src.string.stpncpy
+    libc.src.string.strcasecmp
     libc.src.string.strcat
     libc.src.string.strchr
     libc.src.string.strcmp
     libc.src.string.strcpy
     libc.src.string.strcspn
     libc.src.string.strlcat
     libc.src.string.strlcpy
     libc.src.string.strlen
+    libc.src.string.strncasecmp
     libc.src.string.strncat
     libc.src.string.strncmp
     libc.src.string.strncpy
     libc.src.string.strnlen
     libc.src.string.strpbrk
     libc.src.string.strrchr
     libc.src.string.strspn
     libc.src.string.strstr

libc/config/darwin/arm/entrypoints.txt

set(TARGET_LIBC_ENTRYPOINTS
     libc.src.string.memcmp
     libc.src.string.memcpy
     libc.src.string.memmove
     libc.src.string.mempcpy
     libc.src.string.memrchr
     libc.src.string.memset
     libc.src.string.stpcpy
     libc.src.string.stpncpy
+    libc.src.string.strcasecmp
     libc.src.string.strcat
     libc.src.string.strchr
     libc.src.string.strcmp
     libc.src.string.strcpy
     libc.src.string.strcspn
     libc.src.string.strlcat
     libc.src.string.strlcpy
     libc.src.string.strlen
+    libc.src.string.strncasecmp
     libc.src.string.strncat
     libc.src.string.strncmp
     libc.src.string.strncpy
     libc.src.string.strnlen
     libc.src.string.strpbrk
     libc.src.string.strrchr
     libc.src.string.strspn
     libc.src.string.strstr

libc/config/gpu/entrypoints.txt

set(TARGET_LIBC_ENTRYPOINTS
     libc.src.string.memcmp
     libc.src.string.memcpy
     libc.src.string.memmove
     libc.src.string.mempcpy
     libc.src.string.memrchr
     libc.src.string.memset
     libc.src.string.stpcpy
     libc.src.string.stpncpy
+    libc.src.string.strcasecmp
     libc.src.string.strcat
     libc.src.string.strchr
     libc.src.string.strcmp
     libc.src.string.strcpy
     libc.src.string.strcspn
     libc.src.string.strlcat
     libc.src.string.strlcpy
     libc.src.string.strlen
+    libc.src.string.strncasecmp
     libc.src.string.strncat
     libc.src.string.strncmp
     libc.src.string.strncpy
     libc.src.string.strnlen
     libc.src.string.strpbrk
     libc.src.string.strrchr
     libc.src.string.strspn
     libc.src.string.strstr
     libc.src.string.strtok
     libc.src.string.strtok_r
 )
 
 set(TARGET_LLVMLIBC_ENTRYPOINTS
   ${TARGET_LIBC_ENTRYPOINTS}
 )
 

libc/config/linux/aarch64/entrypoints.txt

set(TARGET_LIBC_ENTRYPOINTS
     libc.src.string.memcmp
     libc.src.string.memcpy
     libc.src.string.memmove
     libc.src.string.mempcpy
     libc.src.string.memrchr
     libc.src.string.memset
     libc.src.string.stpcpy
     libc.src.string.stpncpy
+    libc.src.string.strcasecmp
     libc.src.string.strcat
     libc.src.string.strchr
     libc.src.string.strcmp
     libc.src.string.strcpy
     libc.src.string.strcspn
     libc.src.string.strdup
     libc.src.string.strerror
     libc.src.string.strerror_r
     libc.src.string.strlcat
     libc.src.string.strlcpy
     libc.src.string.strlen
+    libc.src.string.strncasecmp
     libc.src.string.strncat
     libc.src.string.strncmp
     libc.src.string.strncpy
     libc.src.string.strndup
     libc.src.string.strnlen
     libc.src.string.strpbrk
     libc.src.string.strrchr
     libc.src.string.strsignal

libc/config/linux/arm/entrypoints.txt

set(TARGET_LIBC_ENTRYPOINTS
     libc.src.string.memcmp
     libc.src.string.memcpy
     libc.src.string.memmove
     libc.src.string.mempcpy
     libc.src.string.memrchr
     libc.src.string.memset
     libc.src.string.stpcpy
     libc.src.string.stpncpy
+    libc.src.string.strcasecmp
     libc.src.string.strcat
     libc.src.string.strchr
     libc.src.string.strcmp
     libc.src.string.strcpy
     libc.src.string.strcspn
     libc.src.string.strlcat
     libc.src.string.strlcpy
     libc.src.string.strlen
+    libc.src.string.strncasecmp
     libc.src.string.strncat
     libc.src.string.strncmp
     libc.src.string.strncpy
     libc.src.string.strnlen
     libc.src.string.strpbrk
     libc.src.string.strrchr
     libc.src.string.strspn
     libc.src.string.strstr

libc/config/linux/x86_64/entrypoints.txt

set(TARGET_LIBC_ENTRYPOINTS
     libc.src.string.memcmp
     libc.src.string.memcpy
     libc.src.string.memmove
     libc.src.string.mempcpy
     libc.src.string.memrchr
     libc.src.string.memset
     libc.src.string.stpcpy
     libc.src.string.stpncpy
+    libc.src.string.strcasecmp
     libc.src.string.strcat
     libc.src.string.strchr
     libc.src.string.strcmp
     libc.src.string.strcoll
     libc.src.string.strcpy
     libc.src.string.strcspn
     libc.src.string.strdup
     libc.src.string.strerror
     libc.src.string.strerror_r
     libc.src.string.strlcat
     libc.src.string.strlcpy
     libc.src.string.strlen
+    libc.src.string.strncasecmp
     libc.src.string.strncat
     libc.src.string.strncmp
     libc.src.string.strncpy
     libc.src.string.strndup
     libc.src.string.strnlen
     libc.src.string.strpbrk
     libc.src.string.strrchr
     libc.src.string.strsignal

libc/config/windows/entrypoints.txt

set(TARGET_LIBC_ENTRYPOINTS
     libc.src.string.memcmp
     libc.src.string.memcpy
     libc.src.string.memmove
     libc.src.string.mempcpy
     libc.src.string.memrchr
     libc.src.string.memset
     libc.src.string.stpcpy
     libc.src.string.stpncpy
+    libc.src.string.strcasecmp
     libc.src.string.strcat
     libc.src.string.strchr
     libc.src.string.strcmp
     libc.src.string.strcpy
     libc.src.string.strcspn
     libc.src.string.strlcat
     libc.src.string.strlcpy
     libc.src.string.strlen
+    libc.src.string.strncasecmp
     libc.src.string.strncat
     libc.src.string.strncmp
     libc.src.string.strncpy
     libc.src.string.strnlen
     libc.src.string.strpbrk
     libc.src.string.strrchr
     libc.src.string.strspn
     libc.src.string.strstr

libc/spec/bsd_ext.td

HeaderSpec String = HeaderSpec<
         FunctionSpec<
             "strlcpy",
             RetValSpec<SizeTType>,
             [ArgSpec<CharPtr>, ArgSpec<ConstCharPtr>, ArgSpec<SizeTType>]
         >,
       ]
   >;
 
+  HeaderSpec Strings = HeaderSpec<
+      "strings.h",
+      [], // Macros
+      [], // Types
+      [], // Enumerations
+      [
+        FunctionSpec<
+            "strcasecmp",
+            RetValSpec<IntType>,
+            [ArgSpec<ConstCharPtr>, ArgSpec<ConstCharPtr>]
+        >,
+        FunctionSpec<
+            "strncasecmp",
+            RetValSpec<IntType>,
+            [ArgSpec<ConstCharPtr>, ArgSpec<ConstCharPtr>, ArgSpec<SizeTType>]
+        >,
+      ]
+  >;
+
   HeaderSpec SysWait = HeaderSpec<
       "sys/wait.h",
       [], // Macros
       [StructRUsage], // Types
       [], // Enumerations
       [
         FunctionSpec<
             "wait4",
             RetValSpec<PidT>,
             [ArgSpec<PidT>, ArgSpec<IntPtr>, ArgSpec<IntType>, ArgSpec<StructRUsagePtr>]
         >
       ]
   >;
 
   let Headers = [
     String,
+    Strings,
     SysWait,
   ];
 }

libc/src/__support/ctype_utils.h

 static constexpr bool islower(unsigned ch) { return (ch - 'a') < 26; }
 
 static constexpr bool isupper(unsigned ch) { return (ch - 'A') < 26; }
 
 static constexpr bool isspace(unsigned ch) {
   return ch == ' ' || (ch - '\t') < 5;
 }
 
+static constexpr int tolower(int ch) {
+  if (isupper(ch))
+    return ch + ('a' - 'A');
+  return ch;
+}
+
 } // namespace internal
 } // namespace __llvm_libc
 
 #endif //  LLVM_LIBC_SRC_SUPPORT_CTYPE_UTILS_H

libc/src/ctype/tolower.cpp

 #include "src/__support/ctype_utils.h"
 
 #include "src/__support/common.h"
 
 namespace __llvm_libc {
 
 // TODO: Currently restricted to default locale.
 // These should be extended using locale information.
-LLVM_LIBC_FUNCTION(int, tolower, (int c)) {
-  if (internal::isupper(c))
-    return c + ('a' - 'A');
-  return c;
-}
+LLVM_LIBC_FUNCTION(int, tolower, (int c)) { return internal::tolower(c); }
 
 } // namespace __llvm_libc

libc/src/string/CMakeLists.txt

SRCS
     strcmp.cpp
   HDRS
     strcmp.h
   DEPENDS
     .memory_utils.strcmp_implementation
 )
 
 add_entrypoint_object(
+  strcasecmp
+  SRCS
+    strcasecmp.cpp
+  HDRS
+    strcasecmp.h
+  DEPENDS
+    .memory_utils.strcmp_implementation
+    libc.src.__support.ctype_utils
+)
+
+add_entrypoint_object(
   strcoll
   SRCS
     strcoll.cpp
   HDRS
     strcoll.h
 )
 
 add_entrypoint_object(
SRCS
     strncmp.cpp
   HDRS
     strncmp.h
   DEPENDS
     .memory_utils.strcmp_implementation
 )
 
 add_entrypoint_object(
+  strncasecmp
+  SRCS
+    strncasecmp.cpp
+  HDRS
+    strncasecmp.h
+  DEPENDS
+    .memory_utils.strcmp_implementation
+    libc.src.__support.ctype_utils
+)
+
+add_entrypoint_object(
   strncpy
   SRCS
     strncpy.cpp
   HDRS
     strncpy.h
 )
 
 add_entrypoint_object(

libc/src/string/strcasecmp.h

+//===-- Implementation header for strcasecmp --------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_LIBC_SRC_STRING_STRCASECMP_H
+#define LLVM_LIBC_SRC_STRING_STRCASECMP_H
+
+namespace __llvm_libc {
+
+int strcasecmp(const char *left, const char *right);
+
+} // namespace __llvm_libc
+
+#endif // LLVM_LIBC_SRC_STRING_STRCASECMP_H

libc/src/string/strcasecmp.cpp

-//===-- Implementation of tolower------------------------------------------===//
+//===-- Implementation of strcasecmp --------------------------------------===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 
-#include "src/ctype/tolower.h"
-#include "src/__support/ctype_utils.h"
+#include "src/string/strcasecmp.h"
 
 #include "src/__support/common.h"
+#include "src/__support/ctype_utils.h"
+#include "src/string/memory_utils/strcmp_implementations.h"
 
 namespace __llvm_libc {
 
-// TODO: Currently restricted to default locale.
-// These should be extended using locale information.
-LLVM_LIBC_FUNCTION(int, tolower, (int c)) {
-  if (internal::isupper(c))
-    return c + ('a' - 'A');
-  return c;
+LLVM_LIBC_FUNCTION(int, strcasecmp, (const char *left, const char *right)) {
+  auto case_cmp = [](char a, char b) {
+    return __llvm_libc::internal::tolower(a) -
+           __llvm_libc::internal::tolower(b);
+  };
+  return strcmp_implementation(left, right, case_cmp);
 }
 
 } // namespace __llvm_libc

libc/src/string/strncasecmp.h

+//===-- Implementation header for strcasecmp --------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_LIBC_SRC_STRING_STRCASECMP_H
+#define LLVM_LIBC_SRC_STRING_STRCASECMP_H
+
+#include <stddef.h>
+
+namespace __llvm_libc {
+
+int strncasecmp(const char *left, const char *right, size_t n);
+
+} // namespace __llvm_libc
+
+#endif // LLVM_LIBC_SRC_STRING_STRCASECMP_H

libc/src/string/strncasecmp.cpp

+//===-- Implementation of strncasecmp -------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "src/string/strncasecmp.h"
+
+#include "src/__support/common.h"
+#include "src/__support/ctype_utils.h"
+#include "src/string/memory_utils/strcmp_implementations.h"
+
+namespace __llvm_libc {
+
+LLVM_LIBC_FUNCTION(int, strncasecmp,
+                   (const char *left, const char *right, size_t n)) {
+  auto case_cmp = [](char a, char b) {
+    return __llvm_libc::internal::tolower(a) -
+           __llvm_libc::internal::tolower(b);
+  };
+  return strncmp_implementation(left, right, n, case_cmp);
+}
+
+} // namespace __llvm_libc

libc/test/src/string/CMakeLists.txt

SUITE
     libc_string_unittests
   SRCS
     strcmp_test.cpp
   DEPENDS
     libc.src.string.strcmp
 )
 
 add_libc_unittest(
+  strcasecmp_test
+  SUITE
+    libc_string_unittests
+  SRCS
+    strcasecmp_test.cpp
+  DEPENDS
+    libc.src.string.strcasecmp
+)
+
+add_libc_unittest(
   strcoll_test
   SUITE
     libc_string_unittests
   SRCS
     strcoll_test.cpp
   DEPENDS
     libc.src.string.strcoll
 )
SUITE
     libc_string_unittests
   SRCS
     strncmp_test.cpp
   DEPENDS
     libc.src.string.strncmp
 )
 
 add_libc_unittest(
+  strncasecmp_test
+  SUITE
+    libc_string_unittests
+  SRCS
+    strncasecmp_test.cpp
+  DEPENDS
+    libc.src.string.strncasecmp
+)
+
+add_libc_unittest(
   strncpy_test
   SUITE
     libc_string_unittests
   SRCS
     strncpy_test.cpp
   DEPENDS
     libc.src.string.strncpy
 )

libc/test/src/string/strcasecmp_test.cpp

+//===-- Unittests for strcasecmp ------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "src/string/strcasecmp.h"
+#include "utils/UnitTest/Test.h"
+
+TEST(LlvmLibcStrCaseCmpTest, EmptyStringsShouldReturnZero) {
+  const char *s1 = "";
+  const char *s2 = "";
+  int result = __llvm_libc::strcasecmp(s1, s2);
+  ASSERT_EQ(result, 0);
+
+  // Verify operands reversed.
+  result = __llvm_libc::strcasecmp(s2, s1);
+  ASSERT_EQ(result, 0);
+}
+
+TEST(LlvmLibcStrCaseCmpTest, EmptyStringShouldNotEqualNonEmptyString) {
+  const char *empty = "";
+  const char *s2 = "abc";
+  int result = __llvm_libc::strcasecmp(empty, s2);
+  // This should be '\0' - 'a' = -97
+  ASSERT_LT(result, 0);
+
+  // Similar case if empty string is second argument.
+  const char *s3 = "123";
+  result = __llvm_libc::strcasecmp(s3, empty);
+  // This should be '1' - '\0' = 49
+  ASSERT_GT(result, 0);
+}
+
+TEST(LlvmLibcStrCaseCmpTest, Case) {
+  const char *s1 = "aB";
+  const char *s2 = "ab";
+  int result = __llvm_libc::strcasecmp(s1, s2);
+  ASSERT_EQ(result, 0);
+
+  // Verify operands reversed.
+  result = __llvm_libc::strcasecmp(s2, s1);
+  ASSERT_EQ(result, 0);
+}

libc/test/src/string/strcmp_test.cpp

TEST(LlvmLibcStrCmpTest, StringArgumentSwapChangesSign) {
   int result = __llvm_libc::strcmp(b, a);
   // 'b' - 'a' = 1.
   ASSERT_EQ(result, 1);
 
   result = __llvm_libc::strcmp(a, b);
   // 'a' - 'b' = -1.
   ASSERT_EQ(result, -1);
 }
+
+TEST(LlvmLibcStrCmpTest, Case) {
+  const char *s1 = "aB";
+  const char *s2 = "ab";
+  int result = __llvm_libc::strcmp(s1, s2);
+  ASSERT_LT(result, 0);
+
+  // Verify operands reversed.
+  result = __llvm_libc::strcmp(s2, s1);
+  ASSERT_GT(result, 0);
+}

libc/test/src/string/strncasecmp_test.cpp

+//===-- Unittests for strncasecmp -----------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "src/string/strncasecmp.h"
+#include "utils/UnitTest/Test.h"
+
+TEST(LlvmLibcStrNCaseCmpTest,
+     EmptyStringsShouldReturnZeroWithSufficientLength) {
+  const char *s1 = "";
+  const char *s2 = "";
+  int result = __llvm_libc::strncasecmp(s1, s2, 1);
+  ASSERT_EQ(result, 0);
+
+  // Verify operands reversed.
+  result = __llvm_libc::strncasecmp(s2, s1, 1);
+  ASSERT_EQ(result, 0);
+}
+
+TEST(LlvmLibcStrNCaseCmpTest,
+     EmptyStringShouldNotEqualNonEmptyStringWithSufficientLength) {
+  const char *empty = "";
+  const char *s2 = "abc";
+  int result = __llvm_libc::strncasecmp(empty, s2, 3);
+  // This should be '\0' - 'a' = -97
+  ASSERT_LT(result, 0);
+
+  // Similar case if empty string is second argument.
+  const char *s3 = "123";
+  result = __llvm_libc::strncasecmp(s3, empty, 3);
+  // This should be '1' - '\0' = 49
+  ASSERT_GT(result, 0);
+}
+
+TEST(LlvmLibcStrNCaseCmpTest, Case) {
+  const char *s1 = "aB";
+  const char *s2 = "ab";
+  int result = __llvm_libc::strncasecmp(s1, s2, 2);
+  ASSERT_EQ(result, 0);
+
+  // Verify operands reversed.
+  result = __llvm_libc::strncasecmp(s2, s1, 2);
+  ASSERT_EQ(result, 0);
+}

libc/test/src/string/strncmp_test.cpp

TEST(LlvmLibcStrNCmpTest, StringComparisonEndsOnNullByteEvenWithLongerLength) {
   const char *s2 = "abc\0abc";
   int result = __llvm_libc::strncmp(s1, s2, 7);
   ASSERT_EQ(result, 0);
 
   // Verify operands reversed.
   result = __llvm_libc::strncmp(s2, s1, 7);
   ASSERT_EQ(result, 0);
 }
+
+TEST(LlvmLibcStrNCmpTest, Case) {
+  const char *s1 = "aB";
+  const char *s2 = "ab";
+  int result = __llvm_libc::strncmp(s1, s2, 2);
+  ASSERT_LT(result, 0);
+
+  // Verify operands reversed.
+  result = __llvm_libc::strncmp(s2, s1, 2);
+  ASSERT_GT(result, 0);
+}