[libcxx-commits] [libcxx] 9d7ca6c - [regex] fix uncaught exception when string is like "\\_" (#129348)

[via libcxx-commits]

Author: Austin
Date: 2025-03-11T09:12:14+08:00
New Revision: 9d7ca6cdf038dab24fbc5c22a91be36604decb67

URL: https://github.com/llvm/llvm-project/commit/9d7ca6cdf038dab24fbc5c22a91be36604decb67
DIFF: https://github.com/llvm/llvm-project/commit/9d7ca6cdf038dab24fbc5c22a91be36604decb67.diff

LOG: [regex] fix uncaught exception when string is like "\\_" (#129348)

`\_ `is a valid identity escape in regular expressions, but previously
libc++ incorrectly treated it as invalid. So I deleted this judgment
fixes #129062

Added: 
    

Modified: 
    libcxx/include/regex
    libcxx/test/std/re/re.alg/re.alg.match/ecma.pass.cpp

Removed: 
    


################################################################################
diff  --git a/libcxx/include/regex b/libcxx/include/regex
index eebd68af54297..96229c6a6ee42 100644
--- a/libcxx/include/regex
+++ b/libcxx/include/regex
@@ -3955,7 +3955,7 @@ _ForwardIterator basic_regex<_CharT, _Traits>::__parse_character_escape(
       ++__first;
       break;
     default:
-      if (*__first != '_' && !__traits_.isctype(*__first, ctype_base::alnum)) {
+      if (!__traits_.isctype(*__first, ctype_base::alnum)) {
         if (__str)
           *__str = *__first;
         else

diff  --git a/libcxx/test/std/re/re.alg/re.alg.match/ecma.pass.cpp b/libcxx/test/std/re/re.alg/re.alg.match/ecma.pass.cpp
index 4face8ba02bbe..a0699634c2bc9 100644
--- a/libcxx/test/std/re/re.alg/re.alg.match/ecma.pass.cpp
+++ b/libcxx/test/std/re/re.alg/re.alg.match/ecma.pass.cpp
@@ -7,6 +7,8 @@
 //===----------------------------------------------------------------------===//
 //
 
+// XFAIL: FROZEN-CXX03-HEADERS-FIXME
+
 // <regex>
 
 // template <class BidirectionalIterator, class Allocator, class charT, class traits>
@@ -669,6 +671,21 @@ int main(int, char**)
         assert(m.position(0) == 0);
         assert(m.str(0) == s);
     }
+    {
+      std::cmatch m;
+      const char s[] = "$_se";
+      assert(std::regex_match(s, m, std::regex("\\$\\_se")));
+      assert(m.size() == 1);
+      assert(!m.prefix().matched);
+      assert(m.prefix().first == s);
+      assert(m.prefix().second == m[0].first);
+      assert(!m.suffix().matched);
+      assert(m.suffix().first == m[0].second);
+      assert(m.suffix().second == s + std::char_traits<char>::length(s));
+      assert(m.length(0) >= 0 && static_cast<size_t>(m.length(0)) == std::char_traits<char>::length(s));
+      assert(m.position(0) == 0);
+      assert(m.str(0) == s);
+    }
 
 #ifndef TEST_HAS_NO_WIDE_CHARACTERS
     {
@@ -1305,6 +1322,21 @@ int main(int, char**)
         assert(m.position(0) == 0);
         assert(m.str(0) == s);
     }
+    {
+      std::wcmatch m;
+      const wchar_t s[] = L"$_se";
+      assert(std::regex_match(s, m, std::wregex(L"\\$\\_se")));
+      assert(m.size() == 1);
+      assert(!m.prefix().matched);
+      assert(m.prefix().first == s);
+      assert(m.prefix().second == m[0].first);
+      assert(!m.suffix().matched);
+      assert(m.suffix().first == m[0].second);
+      assert(m.suffix().second == s + std::char_traits<wchar_t>::length(s));
+      assert(m.length(0) >= 0 && static_cast<std::size_t>(m.length(0)) == std::char_traits<wchar_t>::length(s));
+      assert(m.position(0) == 0);
+      assert(m.str(0) == s);
+    }
 #endif // TEST_HAS_NO_WIDE_CHARACTERS
 
   return 0;