Add support for $+ to pcre2_substitute (#817)

Author: NWilson

doc/html/pcre2api.html

@@ -3957,6 +3957,7 @@ <h2><a name="SEC38" href="#TOC1">CREATING A NEW STRING WITH SUBSTITUTIONS</a></h
   $`                  insert the substring that precedes the match
   $'                  insert the substring that follows the match
   $_                  insert the entire input string
+  $+                  insert the highest-numbered capture group which matched
   $*MARK or ${*MARK}  insert a control verb name
 </pre>
 Either a group number or a group name can be given for <i>n</i>, for example $2

doc/html/pcre2syntax.html

@@ -722,6 +722,7 @@ <h2><a name="SEC32" href="#TOC1">REPLACEMENT STRINGS</a></h2>
   $`                  insert the substring that precedes the match
   $'                  insert the substring that follows the match
   $_                  insert the entire input string
+  $+                  insert the highest-numbered capture group which matched
   $*MARK or ${*MARK}  insert a control verb name
 </pre>
 For ${n}, n can be a name or a number. If PCRE2_SUBSTITUTE_EXTENDED is set,

doc/pcre2.txt

@@ -3956,6 +3956,7 @@ recognized:
   $`                  insert the substring that precedes the match
   $'                  insert the substring that follows the match
   $_                  insert the entire input string
+  $+                  insert the highest-numbered capture group which matched
   $*MARK or ${*MARK}  insert a control verb name
 .sp
 Either a group number or a group name can be given for \fIn\fP, for example $2

doc/pcre2api.3

@@ -703,6 +703,7 @@ special character is the dollar character in one of the following forms:
   $`                  insert the substring that precedes the match
   $'                  insert the substring that follows the match
   $_                  insert the entire input string
+  $+                  insert the highest-numbered capture group which matched
   $*MARK or ${*MARK}  insert a control verb name
 .sp
 For ${n}, n can be a name or a number. If PCRE2_SUBSTITUTE_EXTENDED is set,

doc/pcre2syntax.3

@@ -1143,6 +1143,44 @@ for (;;)
         subptrend = subject + length;
         goto SUBPTR_SUBSTITUTE;
         }
+      else if (next == CHAR_PLUS &&
+               !(ptr+1 < repend && ptr[1] == CHAR_LEFT_CURLY_BRACKET))
+        {
+        /* Perl supports $+ for "highest captured group" (not the same as $^N
+        which is mainly only useful inside Perl's match callbacks). We also
+        don't accept "$+{..." since that's Perl syntax for our ${name}. */
+        ++ptr;
+        if (code->top_bracket == 0)
+          {
+          /* Treat either as "no such group" or "all groups unset" based on the
+          PCRE2_SUBSTITUTE_UNKNOWN_UNSET option. */
+          if ((suboptions & PCRE2_SUBSTITUTE_UNKNOWN_UNSET) == 0)
+            {
+            rc = PCRE2_ERROR_NOSUBSTRING;
+            goto PTREXIT;
+            }
+          group = 0;
+          }
+        else
+          {
+          /* If we have any capture groups, then the ovector needs to be large
+          enough for all of them, or the result won't be accurate. */
+          if (match_data->oveccount < code->top_bracket + 1)
+            {
+            rc = PCRE2_ERROR_UNAVAILABLE;
+            goto PTREXIT;
+            }
+          for (group = code->top_bracket; group > 0; group--)
+            if (ovector[2*group] != PCRE2_UNSET) break;
+          }
+        if (group == 0)
+          {
+          if ((suboptions & PCRE2_SUBSTITUTE_UNSET_EMPTY) != 0) continue;
+          rc = PCRE2_ERROR_UNSET;
+          goto PTREXIT;
+          }
+        goto GROUP_SUBSTITUTE;
+        }
 
       if (next == CHAR_LEFT_CURLY_BRACKET)
         {

src/pcre2_substitute.c

@@ -5010,6 +5010,34 @@ B)x/alt_verbnames,mark
     LabcR\=replace=>$'<
     LabcR\=replace=>$_<
 
+/A(.)|B(.)/
+    AMZ\=replace=>$+<
+    BMZ\=replace=>$+<
+    AMZ\=replace=>$+
+    AMZ\=replace=>$+{name}
+    BMZ\=ovector=2
+    BMZ\=ovector=3
+    BMZ\=ovector=2,replace=>$+<
+    BMZ\=ovector=3,replace=>$+<
+    BMZ\=ovector=2,substitute_matched,replace=>$+<
+    BMZ\=ovector=3,substitute_matched,replace=>$+<
+
+/A|(B)/
+    B\=replace=>$1<
+    A\=replace=>$1<
+    B\=substitute_unset_empty,replace=>$1<
+    A\=substitute_unset_empty,replace=>$1<
+    B\=replace=>$+<
+    A\=replace=>$+<
+    B\=substitute_unset_empty,replace=>$+<
+    A\=substitute_unset_empty,replace=>$+<
+
+/ABC/
+    ABC\=replace=$+z
+    ABC\=substitute_unknown_unset,replace=$+z
+    ABC\=substitute_unset_empty,replace=$+z
+    ABC\=substitute_unknown_unset,substitute_unset_empty,replace=$+z
+
 /^(o(\1{72}{\"{\\{00000059079}\d*){74}}){19}/I
 
 /((p(?'K/

testdata/testinput2

@@ -14154,8 +14154,8 @@ Failed: error 119 at offset 3: parentheses are too deeply nested
 
 /abc/replace=a$++
     123abc
-Failed: error -35 at offset 2 in replacement: invalid replacement string
-        here: a$ |<--| ++
+Failed: error -49 at offset 3 in replacement: unknown substring
+        here: a$+ |<--| +
 
 /abc/replace=a$bad
     123abc
@@ -15980,6 +15980,68 @@ Failed: error -49 at offset 3 in replacement: unknown substring
     LabcR\=replace=>$_<
  1: L>LabcR<R
 
+/A(.)|B(.)/
+    AMZ\=replace=>$+<
+ 1: >M<Z
+    BMZ\=replace=>$+<
+ 1: >M<Z
+    AMZ\=replace=>$+
+ 1: >MZ
+    AMZ\=replace=>$+{name}
+Failed: error -35 at offset 2 in replacement: invalid replacement string
+        here: >$ |<--| +{name}
+    BMZ\=ovector=2
+Matched, but too many substrings
+ 0: BM
+ 1: <unset>
+    BMZ\=ovector=3
+ 0: BM
+ 1: <unset>
+ 2: M
+    BMZ\=ovector=2,replace=>$+<
+Failed: error -54 at offset 3 in replacement: requested value is not available
+        here: >$+ |<--| <
+    BMZ\=ovector=3,replace=>$+<
+ 1: >M<Z
+    BMZ\=ovector=2,substitute_matched,replace=>$+<
+Failed: error -54 at offset 3 in replacement: requested value is not available
+        here: >$+ |<--| <
+    BMZ\=ovector=3,substitute_matched,replace=>$+<
+ 1: >M<Z
+
+/A|(B)/
+    B\=replace=>$1<
+ 1: >B<
+    A\=replace=>$1<
+Failed: error -55 at offset 3 in replacement: requested value is not set
+        here: >$1 |<--| <
+    B\=substitute_unset_empty,replace=>$1<
+ 1: >B<
+    A\=substitute_unset_empty,replace=>$1<
+ 1: ><
+    B\=replace=>$+<
+ 1: >B<
+    A\=replace=>$+<
+Failed: error -55 at offset 3 in replacement: requested value is not set
+        here: >$+ |<--| <
+    B\=substitute_unset_empty,replace=>$+<
+ 1: >B<
+    A\=substitute_unset_empty,replace=>$+<
+ 1: ><
+
+/ABC/
+    ABC\=replace=$+z
+Failed: error -49 at offset 2 in replacement: unknown substring
+        here: $+ |<--| z
+    ABC\=substitute_unknown_unset,replace=$+z
+Failed: error -55 at offset 2 in replacement: requested value is not set
+        here: $+ |<--| z
+    ABC\=substitute_unset_empty,replace=$+z
+Failed: error -49 at offset 2 in replacement: unknown substring
+        here: $+ |<--| z
+    ABC\=substitute_unknown_unset,substitute_unset_empty,replace=$+z
+ 1: z
+
 /^(o(\1{72}{\"{\\{00000059079}\d*){74}}){19}/I
 Capture group count = 2
 Max back reference = 1