From 3d75d315c72facb214ae80ab390ba5137c10c190 Mon Sep 17 00:00:00 2001
From: mohammed arib <arib@bugqore.com>
Date: Sat, 4 Jul 2026 15:46:43 +0530
Subject: [PATCH] fix bad-UTF-8 check in ParseUnicodeGroup

---
 re2/parse.cc              |  2 +-
 re2/testing/parse_test.cc | 15 +++++++++++++++
 2 files changed, 16 insertions(+), 1 deletion(-)

diff --git a/re2/parse.cc b/re2/parse.cc
index 545b965e9..0a7678963 100644
--- a/re2/parse.cc
+++ b/re2/parse.cc
@@ -1787,7 +1787,7 @@ ParseStatus ParseUnicodeGroup(absl::string_view* s,
   absl::string_view name;  // Han or L
   s->remove_prefix(2);  // '\\', 'p'
 
-  if (!StringViewToRune(&c, s, status))
+  if (StringViewToRune(&c, s, status) < 0)
     return kParseError;
   if (c != '{') {
     // Name is the bit of string we just skipped over for c.
diff --git a/re2/testing/parse_test.cc b/re2/testing/parse_test.cc
index 53ef24ec7..639de681e 100644
--- a/re2/testing/parse_test.cc
+++ b/re2/testing/parse_test.cc
@@ -583,4 +583,19 @@ TEST(LookAround, ErrorArgs) {
   EXPECT_EQ(status.error_arg(), "(?<!");
 }
 
+// \p or \P followed by invalid UTF-8 (rather than a group name) must be
+// reported as bad UTF-8, matching the \p{...} form.
+TEST(UnicodeGroups, BadUTF8) {
+  RegexpStatus status;
+  Regexp* re;
+
+  re = Regexp::Parse(std::string("\\p\xff"), Regexp::LikePerl, &status);
+  EXPECT_TRUE(re == NULL);
+  EXPECT_EQ(status.code(), kRegexpBadUTF8);
+
+  re = Regexp::Parse(std::string("\\P\xff"), Regexp::LikePerl, &status);
+  EXPECT_TRUE(re == NULL);
+  EXPECT_EQ(status.code(), kRegexpBadUTF8);
+}
+
 }  // namespace re2