[libc] Implemented wcspbrk (#142040)

Implemented wcspbrk and added tests

Author: uzairnawaz

libc/config/linux/x86_64/entrypoints.txt

@@ -366,6 +366,7 @@ set(TARGET_LIBC_ENTRYPOINTS
     libc.src.wchar.wctob
     libc.src.wchar.wmemset
     libc.src.wchar.wcschr
+    libc.src.wchar.wcspbrk
     libc.src.wchar.wcsspn
     libc.src.wchar.wmemcmp
 

libc/include/wchar.yaml

@@ -42,6 +42,13 @@ functions:
     arguments: 
       - type: const wchar_t *
       - type: wchar_t
+  - name: wcspbrk
+    standards:
+      - stdc
+    return_type: const wchar_t *
+    arguments:
+      - type: const wchar_t *
+      - type: const wchar_t *
   - name: wcsspn
     standards:
       - stdc

libc/src/wchar/CMakeLists.txt

@@ -57,6 +57,17 @@ add_entrypoint_object(
     libc.src.__support.wctype_utils
 )
 
+add_entrypoint_object(
+  wcspbrk
+  SRCS
+    wcspbrk.cpp
+  HDRS
+    wcspbrk.h
+  DEPENDS
+    libc.hdr.wchar_macros
+    libc.src.__support.wctype_utils
+)
+
 add_entrypoint_object(
   wcsspn
   SRCS

libc/src/wchar/wcspbrk.cpp

@@ -0,0 +1,34 @@
+//===-- Implementation of wcspbrk -----------------------------------------===//
+//
+// 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/wcspbrk.h"
+
+#include "hdr/types/wchar_t.h"
+#include "src/__support/common.h"
+
+namespace LIBC_NAMESPACE_DECL {
+
+bool contains_char(const wchar_t *str, wchar_t target) {
+  for (; *str != L'\0'; str++)
+    if (*str == target)
+      return true;
+
+  return false;
+}
+
+LLVM_LIBC_FUNCTION(const wchar_t *, wcspbrk,
+                   (const wchar_t *src, const wchar_t *breakset)) {
+  // currently O(n * m), can be further optimized to O(n + m) with a hash set
+  for (int src_idx = 0; src[src_idx] != 0; src_idx++)
+    if (contains_char(breakset, src[src_idx]))
+      return src + src_idx;
+
+  return nullptr;
+}
+
+} // namespace LIBC_NAMESPACE_DECL

libc/src/wchar/wcspbrk.h

@@ -0,0 +1,21 @@
+//===-- Implementation header for wcspbrk ---------------------------------===//
+//
+// 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_WCSPBRK_H
+#define LLVM_LIBC_SRC_WCHAR_WCSPBRK_H
+
+#include "hdr/types/wchar_t.h"
+#include "src/__support/macros/config.h"
+
+namespace LIBC_NAMESPACE_DECL {
+
+const wchar_t *wcspbrk(const wchar_t *src, const wchar_t *breakset);
+
+} // namespace LIBC_NAMESPACE_DECL
+
+#endif // LLVM_LIBC_SRC_WCHAR_WCSPBRK_H

libc/test/src/wchar/CMakeLists.txt

@@ -55,6 +55,16 @@ add_libc_test(
     libc.src.wchar.wcschr
 )
 
+add_libc_test(
+  wcspbrk_test
+  SUITE
+    libc_wchar_unittests
+  SRCS
+    wcspbrk_test.cpp
+  DEPENDS
+    libc.src.wchar.wcspbrk
+)
+
 add_libc_test(
   wcsspn_test
   SUITE

libc/test/src/wchar/wcspbrk_test.cpp

@@ -0,0 +1,62 @@
+//===-- Unittests for wcspbrk ---------------------------------------------===//
+//
+// 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/wcspbrk.h"
+#include "test/UnitTest/Test.h"
+
+TEST(LlvmLibcWCSPBrkTest, EmptyStringShouldReturnNullptr) {
+  // The search should not include the null terminator.
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(L"", L""), nullptr);
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(L"_", L""), nullptr);
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(L"", L"_"), nullptr);
+}
+
+TEST(LlvmLibcWCSPBrkTest, ShouldNotFindAnythingAfterNullTerminator) {
+  const wchar_t src[4] = {'a', 'b', '\0', 'c'};
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(src, L"c"), nullptr);
+}
+
+TEST(LlvmLibcWCSPBrkTest, ShouldReturnNullptrIfNoCharactersFound) {
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(L"12345", L"abcdef"), nullptr);
+}
+
+TEST(LlvmLibcWCSPBrkTest, FindsFirstCharacter) {
+  const wchar_t *src = L"12345";
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(src, L"1"), src);
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(src, L"-1"), src);
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(src, L"1_"), src);
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(src, L"f1_"), src);
+}
+
+TEST(LlvmLibcWCSPBrkTest, FindsMiddleCharacter) {
+  const wchar_t *src = L"12345";
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(src, L"3"), src + 2);
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(src, L"?3"), src + 2);
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(src, L"3F"), src + 2);
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(src, L"z3_"), src + 2);
+}
+
+TEST(LlvmLibcWCSPBrkTest, FindsLastCharacter) {
+  const wchar_t *src = L"12345";
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(src, L"5"), src + 4);
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(src, L"r5"), src + 4);
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(src, L"59"), src + 4);
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(src, L"n5_"), src + 4);
+}
+
+TEST(LlvmLibcWCSPBrkTest, FindsFirstOfRepeated) {
+  const wchar_t *src = L"A,B,C,D";
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(src, L","), src + 1);
+}
+
+TEST(LlvmLibcWCSPBrkTest, FindsFirstInBreakset) {
+  const wchar_t *src = L"12345";
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(src, L"34"), src + 2);
+  EXPECT_EQ(LIBC_NAMESPACE::wcspbrk(src, L"43"), src + 2);
+}