Fix 32-bit quantifier following a character larger than the maximum UTF character.

Author: PhilipHazel

ChangeLog

@@ -155,6 +155,13 @@ consistency with OP_VREVERSE.
 40. In some legacy environments with a pre C99 snprintf, pcre2_regerror could
 return an incorrect value when the provided buffer was too small.
 
+41. Applied pull request #342 which adds sanity checks for ctype functions and 
+locks out any accidental sign-extension.
+
+42. In the 32-bit library, in non-UTF mode, a quantifier that followed a
+literal character with a value greater than or equal to 0x80000000u caused 
+undefined behaviour.
+
 
 Version 10.42 11-December-2022
 ------------------------------

src/pcre2_compile.c

@@ -2781,6 +2781,7 @@ uint32_t *verbstartptr = NULL;
 uint32_t *previous_callout = NULL;
 uint32_t *parsed_pattern = cb->parsed_pattern;
 uint32_t *parsed_pattern_end = cb->parsed_pattern_end;
+uint32_t *this_parsed_item = NULL;
 uint32_t meta_quantifier = 0;
 uint32_t add_after_mark = 0;
 uint32_t xoptions = cb->cx->extra_options;
@@ -2866,10 +2867,11 @@ while (ptr < ptrend)
   uint32_t xset, xunset, *xoptset;
   uint32_t terminator;
   uint32_t prev_meta_quantifier;
+  uint32_t *prev_parsed_item = this_parsed_item; 
   BOOL prev_okquantifier;
   PCRE2_SPTR tempptr;
   PCRE2_SIZE offset;
-
+  
   if (parsed_pattern >= parsed_pattern_end)
     {
     errorcode = ERR63;  /* Internal error (parsed pattern overflow) */
@@ -2881,6 +2883,10 @@ while (ptr < ptrend)
     errorcode = ERR19;
     goto FAILED;        /* Parentheses too deeply nested */
     }
+    
+  /* Remember where this item started */
+
+  this_parsed_item = parsed_pattern;
 
   /* Get next input character, save its position for callout handling. */
 
@@ -3173,7 +3179,6 @@ while (ptr < ptrend)
     continue;  /* Next character in pattern */
     }
 
-
   /* Process the next item in the main part of a pattern. */
 
   switch(c)
@@ -3450,7 +3455,7 @@ while (ptr < ptrend)
     wrapping it in non-capturing brackets, but we have to allow for a preceding
     (*MARK) for when (*ACCEPT) has an argument. */
 
-    if (parsed_pattern[-1] == META_ACCEPT)
+    if (*prev_parsed_item == META_ACCEPT)
       {
       uint32_t *p;
       for (p = parsed_pattern - 1; p >= verbstartptr; p--) p[1] = p[0];

testdata/testinput12

@@ -560,4 +560,10 @@
 
 # ---------------------------------------------------- 
 
+# Quantifier after a literal that has the value of META_ACCEPT (not UTF). This
+# fails in 16-bit mode, but is OK for 32-bit.
+
+/\x{802a0000}*/
+    \x{802a0000}\x{802a0000}
+
 # End of testinput12

testdata/testoutput12-16

@@ -1803,4 +1803,11 @@ No match
 
 # ---------------------------------------------------- 
 
+# Quantifier after a literal that has the value of META_ACCEPT (not UTF). This
+# fails in 16-bit mode, but is OK for 32-bit.
+
+/\x{802a0000}*/
+Failed: error 134 at offset 11: character code point value in \x{} or \o{} is too large
+    \x{802a0000}\x{802a0000}
+
 # End of testinput12

testdata/testoutput12-32

@@ -1801,4 +1801,11 @@ No match
 
 # ---------------------------------------------------- 
 
+# Quantifier after a literal that has the value of META_ACCEPT (not UTF). This
+# fails in 16-bit mode, but is OK for 32-bit.
+
+/\x{802a0000}*/
+    \x{802a0000}\x{802a0000}
+ 0: \x{802a0000}\x{802a0000}
+
 # End of testinput12