From 3c869816ace0ae4dcce6bab78411cccaa5fb3dbc Mon Sep 17 00:00:00 2001
From: "Philip.Hazel" <Philip.Hazel@gmail.com>
Date: Sat, 16 Nov 2019 17:30:07 +0000
Subject: [PATCH] Fix sometimes failing caseless non-ASCII matching in
 assertion.

---
 ChangeLog            |  5 +++++
 src/pcre2_compile.c  | 13 +++++++++++++
 testdata/testinput4  |  8 ++++++++
 testdata/testoutput4 | 12 ++++++++++++
 4 files changed, 38 insertions(+)

diff --git a/ChangeLog b/ChangeLog
index dd52203..27033ae 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -177,6 +177,11 @@ sanitizer complaint (regexec is supposed to be thread safe).
 37. Add NEON vectorization to JIT to speed up matching of first character and
 pairs of characters on ARM64 CPUs.
 
+38. If a non-ASCII character was the first in a starting assertion in a
+caseless match, the "first code unit" optimization did not get the casing
+right, and the assertion failed to match a character in the other case if it
+did not start with the same code unit.
+
 
 Version 10.33 16-April-2019
 ---------------------------
diff --git a/src/pcre2_compile.c b/src/pcre2_compile.c
index 3204973..800b61b 100644
--- a/src/pcre2_compile.c
+++ b/src/pcre2_compile.c
@@ -8741,6 +8741,19 @@ do {
      case OP_MINPLUSI:
      case OP_POSPLUSI:
      if (inassert == 0) return 0;
+
+     /* If the character is more than one code unit long, we cannot set its
+     first code unit when matching caselessly. Later scanning may pick up 
+     multiple code units. */
+     
+#ifdef SUPPORT_UNICODE
+#if PCRE2_CODE_UNIT_WIDTH == 8
+     if (scode[1] >= 0x80) return 0;
+#elif PCRE2_CODE_UNIT_WIDTH == 16
+     if (scode[1] >= 0xd800 && scode[1] <= 0xdfff) return 0;
+#endif
+#endif
+
      if (cflags < 0) { c = scode[1]; cflags = REQ_CASELESS; }
        else if (c != scode[1]) return 0;
      break;
diff --git a/testdata/testinput4 b/testdata/testinput4
index f3d498c..0871835 100644
--- a/testdata/testinput4
+++ b/testdata/testinput4
@@ -2483,4 +2483,12 @@
 /\X*/
     \xF3aaa\xE4\xEA\xEB\xFEa
 
+/Я/i,utf
+    \x{42f} 
+    \x{44f} 
+
+/(?=Я)/i,utf
+    \x{42f} 
+    \x{44f} 
+
 # End of testinput4
diff --git a/testdata/testoutput4 b/testdata/testoutput4
index 53926ed..2c8037b 100644
--- a/testdata/testoutput4
+++ b/testdata/testoutput4
@@ -4016,4 +4016,16 @@ No match
     \xF3aaa\xE4\xEA\xEB\xFEa
  0: \xf3aaa\xe4\xea\xeb\xfea
 
+/Я/i,utf
+    \x{42f} 
+ 0: \x{42f}
+    \x{44f} 
+ 0: \x{44f}
+
+/(?=Я)/i,utf
+    \x{42f} 
+ 0: 
+    \x{44f} 
+ 0: 
+
 # End of testinput4