From 54396a364d4189dc801b3dffa81d875d51a7d40d Mon Sep 17 00:00:00 2001
From: "Philip.Hazel" <Philip.Hazel@gmail.com>
Date: Fri, 6 Mar 2015 11:57:06 +0000
Subject: [PATCH] Fix bad compile of patterns like /[A-`]/i,utf where the range
 contains characters with multiple other cases and the ranges adjoin.

---
 ChangeLog            |  5 +++++
 src/pcre2_compile.c  |  6 +++++-
 testdata/testinput4  |  3 +++
 testdata/testinput5  |  3 +++
 testdata/testoutput4 |  4 ++++
 testdata/testoutput5 | 10 ++++++++++
 6 files changed, 30 insertions(+), 1 deletion(-)

diff --git a/ChangeLog b/ChangeLog
index 34e1051..e9555b2 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -115,6 +115,11 @@ of various issues, and new ones are still appear unfortunately. To fix
 existing and future issues, size computation is eliminated from the code,
 and replaced by on-demand memory allocation.
 
+25. A pattern such as /(?i)[A-`]/, where characters in the other case are
+adjacent to the end of the range, and the range contained characters with more
+than one other case, caused incorrect behaviour when compiled in UTF mode. In
+that example, the range a-j was left out of the class.
+
 
 Version 10.00 05-January-2015
 -----------------------------
diff --git a/src/pcre2_compile.c b/src/pcre2_compile.c
index 280c3d0..a8defd1 100644
--- a/src/pcre2_compile.c
+++ b/src/pcre2_compile.c
@@ -2864,7 +2864,11 @@ if ((options & PCRE2_CASELESS) != 0)
       range. Otherwise, use a recursive call to add the additional range. */
 
       else if (oc < start && od >= start - 1) start = oc; /* Extend downwards */
-      else if (od > end && oc <= end + 1) end = od;       /* Extend upwards */
+      else if (od > end && oc <= end + 1)
+        {
+        end = od;       /* Extend upwards */
+        if (end > classbits_end) classbits_end = (end <= 0xff ? end : 0xff);
+        }
       else n8 += add_to_class(classbits, uchardptr, options, cb, oc, od);
       }
     }
diff --git a/testdata/testinput4 b/testdata/testinput4
index e121058..7f7d4ec 100644
--- a/testdata/testinput4
+++ b/testdata/testinput4
@@ -2216,4 +2216,7 @@
     \x{23a}\x{2c65}\x{2c65}\x{2c65}
     \x{23a}\x{23a}\x{2c65}\x{23a}
 
+/[A-`]/i,utf
+    abcdefghijklmno
+
 # End of testinput4
diff --git a/testdata/testinput5 b/testdata/testinput5
index 4c8dfeb..9c3771b 100644
--- a/testdata/testinput5
+++ b/testdata/testinput5
@@ -1638,4 +1638,7 @@
 
 /[^\xff]((?1))/utf,debug
 
+/[A-`]/iB,utf
+    abcdefghijklmno
+
 # End of testinput5 
diff --git a/testdata/testoutput4 b/testdata/testoutput4
index 774594c..80b14c6 100644
--- a/testdata/testoutput4
+++ b/testdata/testoutput4
@@ -3735,4 +3735,8 @@ No match
     \x{23a}\x{23a}\x{2c65}\x{23a}
 No match
 
+/[A-`]/i,utf
+    abcdefghijklmno
+ 0: a
+
 # End of testinput4
diff --git a/testdata/testoutput5 b/testdata/testoutput5
index 0a331e1..46e66c5 100644
--- a/testdata/testoutput5
+++ b/testdata/testoutput5
@@ -4009,4 +4009,14 @@ Subject length lower bound = 1
 /[^\xff]((?1))/utf,debug
 Failed: error 140 at offset 11: recursion could loop indefinitely
 
+/[A-`]/iB,utf
+------------------------------------------------------------------
+        Bra
+        [A-z\x{212a}\x{17f}]
+        Ket
+        End
+------------------------------------------------------------------
+    abcdefghijklmno
+ 0: a
+
 # End of testinput5