[libc] Implemented wcsrchr

[uzairnawaz]

fixes #124347
Implemented wcsrchr and added tests

[llvmbot]

@llvm/pr-subscribers-libc

Author: Uzair Nawaz (uzairnawaz)

Changes

Implemented wcsrchr and added tests

---

Full diff: https://github.com/llvm/llvm-project/pull/142436.diff

7 Files Affected:

• (modified) libc/config/linux/x86_64/entrypoints.txt (+1)
• (modified) libc/include/wchar.yaml (+7)
• (modified) libc/src/wchar/CMakeLists.txt (+11)
• (added) libc/src/wchar/wcsrchr.cpp (+31)
• (added) libc/src/wchar/wcsrchr.h (+21)
• (modified) libc/test/src/wchar/CMakeLists.txt (+10)
• (added) libc/test/src/wchar/wcsrchr_test.cpp (+68)

diff --git a/libc/config/linux/x86_64/entrypoints.txt b/libc/config/linux/x86_64/entrypoints.txt
index 545b9227349fe..e6cbfe0257a3e 100644
--- a/libc/config/linux/x86_64/entrypoints.txt
+++ b/libc/config/linux/x86_64/entrypoints.txt
@@ -367,6 +367,7 @@ set(TARGET_LIBC_ENTRYPOINTS
     libc.src.wchar.wmemset
     libc.src.wchar.wcschr
     libc.src.wchar.wcspbrk
+    libc.src.wchar.wcsrchr
     libc.src.wchar.wcsspn
     libc.src.wchar.wmemcmp
     libc.src.wchar.wmemcpy
diff --git a/libc/include/wchar.yaml b/libc/include/wchar.yaml
index bfd9a10342019..39732feb62fcd 100644
--- a/libc/include/wchar.yaml
+++ b/libc/include/wchar.yaml
@@ -49,6 +49,13 @@ functions:
     arguments:
       - type: const wchar_t *
       - type: const wchar_t *
+  - name: wcsrchr
+    standards:
+      - stdc
+    return_type: const wchar_t *
+    arguments:
+      - type: const wchar_t *
+      - type: wchar_t
   - name: wcsspn
     standards:
       - stdc
diff --git a/libc/src/wchar/CMakeLists.txt b/libc/src/wchar/CMakeLists.txt
index 9db121762348b..dd597550e06d8 100644
--- a/libc/src/wchar/CMakeLists.txt
+++ b/libc/src/wchar/CMakeLists.txt
@@ -68,6 +68,17 @@ add_entrypoint_object(
     libc.src.__support.wctype_utils
 )
 
+add_entrypoint_object(
+  wcsrchr
+  SRCS
+    wcsrchr.cpp
+  HDRS
+    wcsrchr.h
+  DEPENDS
+    libc.hdr.wchar_macros
+    libc.src.__support.wctype_utils
+)
+
 add_entrypoint_object(
   wcsspn
   SRCS
diff --git a/libc/src/wchar/wcsrchr.cpp b/libc/src/wchar/wcsrchr.cpp
new file mode 100644
index 0000000000000..bb4e373bb1ce1
--- /dev/null
+++ b/libc/src/wchar/wcsrchr.cpp
@@ -0,0 +1,31 @@
+//===-- Implementation of wcsrchr -----------------------------------------===//
+//
+// 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/wchar/wcsrchr.h"
+
+#include "hdr/types/wchar_t.h"
+#include "src/__support/common.h"
+#include "src/__support/macros/config.h"
+#include "src/__support/macros/null_check.h"
+
+namespace LIBC_NAMESPACE_DECL {
+
+LLVM_LIBC_FUNCTION(const wchar_t *, wcsrchr, (const wchar_t *s, wchar_t c)) {
+  LIBC_CRASH_ON_NULLPTR(s);
+
+  const wchar_t *last_occurrence = nullptr;
+  while (true) {
+    if (*s == c)
+      last_occurrence = s;
+    if (*s == L'\0')
+      return last_occurrence;
+    ++s;
+  }
+}
+
+} // namespace LIBC_NAMESPACE_DECL
diff --git a/libc/src/wchar/wcsrchr.h b/libc/src/wchar/wcsrchr.h
new file mode 100644
index 0000000000000..5e9d8c3c2db92
--- /dev/null
+++ b/libc/src/wchar/wcsrchr.h
@@ -0,0 +1,21 @@
+//===-- Implementation header for wcsrchr ---------------------------------===//
+//
+// 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_WCHAR_WCSRCHR_H
+#define LLVM_LIBC_SRC_WCHAR_WCSRCHR_H
+
+#include "hdr/types/wchar_t.h"
+#include "src/__support/macros/config.h"
+
+namespace LIBC_NAMESPACE_DECL {
+
+const wchar_t *wcsrchr(const wchar_t *s, wchar_t c);
+
+} // namespace LIBC_NAMESPACE_DECL
+
+#endif // LLVM_LIBC_SRC_WCHAR_WCSRCHR_H
diff --git a/libc/test/src/wchar/CMakeLists.txt b/libc/test/src/wchar/CMakeLists.txt
index 9bc230e0bddf3..6a212a620f853 100644
--- a/libc/test/src/wchar/CMakeLists.txt
+++ b/libc/test/src/wchar/CMakeLists.txt
@@ -65,6 +65,16 @@ add_libc_test(
     libc.src.wchar.wcspbrk
 )
 
+add_libc_test(
+  wcsrchr_test
+  SUITE
+    libc_wchar_unittests
+  SRCS
+    wcsrchr_test.cpp
+  DEPENDS
+    libc.src.wchar.wcsrchr
+)
+
 add_libc_test(
   wcsspn_test
   SUITE
diff --git a/libc/test/src/wchar/wcsrchr_test.cpp b/libc/test/src/wchar/wcsrchr_test.cpp
new file mode 100644
index 0000000000000..707dfb6756ee9
--- /dev/null
+++ b/libc/test/src/wchar/wcsrchr_test.cpp
@@ -0,0 +1,68 @@
+//===-- Unittests for wcsrchr ---------------------------------------------===//
+//
+// 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 "hdr/types/wchar_t.h"
+#include "src/wchar/wcsrchr.h"
+#include "test/UnitTest/Test.h"
+
+TEST(LlvmLibcWCSRChrTest, FindsFirstCharacter) {
+  // Should return pointer to original string since 'a' is the first character.
+  const wchar_t *src = L"abcde";
+  ASSERT_EQ(LIBC_NAMESPACE::wcsrchr(src, L'a'), src);
+}
+
+TEST(LlvmLibcWCSRChrTest, FindsMiddleCharacter) {
+  // Should return pointer to 'c'.
+  const wchar_t *src = L"abcde";
+  ASSERT_EQ(LIBC_NAMESPACE::wcsrchr(src, L'c'), (src + 2));
+}
+
+TEST(LlvmLibcWCSRChrTest, FindsLastCharacterThatIsNotNullTerminator) {
+  // Should return pointer to 'e'.
+  const wchar_t *src = L"abcde";
+  ASSERT_EQ(LIBC_NAMESPACE::wcsrchr(src, L'e'), (src + 4));
+}
+
+TEST(LlvmLibcWCSRChrTest, FindsNullTerminator) {
+  // Should return pointer to null terminator.
+  const wchar_t *src = L"abcde";
+  ASSERT_EQ(LIBC_NAMESPACE::wcsrchr(src, L'\0'), (src + 5));
+}
+
+TEST(LlvmLibcWCSRChrTest, CharacterNotWithinStringShouldReturnNullptr) {
+  // Since 'z' is not within the string, should return nullptr.
+  const wchar_t *src = L"abcde";
+  ASSERT_EQ(LIBC_NAMESPACE::wcsrchr(src, L'z'), nullptr);
+}
+
+TEST(LlvmLibcWCSRChrTest, ShouldFindLastOfDuplicates) {
+  // Should return pointer to the last '1'.
+  const wchar_t *src = L"abc1def1ghi";
+  ASSERT_EQ((int)(LIBC_NAMESPACE::wcsrchr(src, L'1') - src), 7);
+
+  // Should return pointer to the last 'X'
+  const wchar_t *dups = L"XXXXX";
+  ASSERT_EQ(LIBC_NAMESPACE::wcsrchr(dups, L'X'), dups + 4);
+}
+
+TEST(LlvmLibcWCSRChrTest, EmptyStringShouldOnlyMatchNullTerminator) {
+  // Null terminator should match
+  const wchar_t *src = L"";
+  ASSERT_EQ(src, LIBC_NAMESPACE::wcsrchr(src, L'\0'));
+  // All other characters should not match
+  ASSERT_EQ(LIBC_NAMESPACE::wcsrchr(src, L'Z'), nullptr);
+  ASSERT_EQ(LIBC_NAMESPACE::wcsrchr(src, L'3'), nullptr);
+  ASSERT_EQ(LIBC_NAMESPACE::wcsrchr(src, L'*'), nullptr);
+}
+
+#if defined(LIBC_ADD_NULL_CHECKS) && !defined(LIBC_HAS_SANITIZER)
+TEST(LlvmLibcWCSRChrTest, NullptrCrash) {
+  // Passing in a nullptr should crash the program.
+  EXPECT_DEATH([] { LIBC_NAMESPACE::wcsrchr(nullptr, L'a'); }, WITH_SIGNAL(-1));
+}
+#endif // LIBC_HAS_ADDRESS_SANITIZER

[michaelrj-google]

just a build fix, otherwise LGTM

[michaelrj-google]

don't need wctype_utils

[michaelrj-google]

please add "fixes #124347" to the description