regexec.c - teach BRANCH and BRANCHJ nodes to reset capture buffers

In /((a)(b)|(a))+/ we should not end up with $2 and $4 being set at
the same time. When a branch fails it should reset any capture buffers
that might be touched by its branch.

We change BRANCH and BRANCHJ to store the number of parens before the
branch, and the number of parens after the branch was completed. When
a BRANCH operation fails, we clear the buffers it contains before we
continue on.

It is a bit more complex than it should be because we have BRANCHJ
and BRANCH. (One of these days we should merge them together.)

This is also made somewhat more complex because TRIE nodes are actually
branches, and may need to track capture buffers also, at two levels.
The overall TRIE op, and for jump tries especially where we emulate
the behavior of branches. So we have to do the same clearing logic if
a trie branch fails as well.

2 files changed, 21 insertions, 14 deletions

diff --git a/t/re/pat.t b/t/re/pat.t
index b837157c42..95070b2290 100644
--- a/t/re/pat.t
+++ b/t/re/pat.t
@@ -2427,7 +2427,6 @@ SKIP:
         }, 'ok', {}, 'gh20826: test regex save stack overflow');
     }
     {
-        local $::TODO = "Not Yet Implemented";
         my ($x, $y);
         ok( "aaa" =~ /(?:(a)?\1)+/,
             "GH Issue #18865 'aaa' - pattern matches");
@@ -2435,26 +2434,33 @@ SKIP:
         ok( "aaa" =~ /(?:((?{})a)?\1)+/,
             "GH Issue #18865 'aaa' - deoptimized pattern matches");
         $y = "($-[0],$+[0])";
-        is( $y, $x,
-            "GH Issue #18865 'aaa' - test optimization");
-
+        {
+            local $::TODO = "Not Yet Implemented";
+            is( $y, $x,
+                "GH Issue #18865 'aaa' - test optimization");
+        }
         ok( "ababab" =~ /(?:(?:(ab))?\1)+/,
             "GH Issue #18865 'ababab' - pattern matches");
         $x = "($-[0],$+[0])";
         ok( "ababab" =~ /(?:(?:((?{})ab))?\1)+/,
             "GH Issue #18865 'ababab' - deoptimized pattern matches");
         $y = "($-[0],$+[0])";
-        is( $y, $x,
-            "GH Issue #18865 'ababab' - test optimization");
-
+        {
+            local $::TODO = "Not Yet Implemented";
+            is( $y, $x,
+                "GH Issue #18865 'ababab' - test optimization");
+        }
         ok( "XaaXbbXb" =~ /(?:X([ab])?\1)+/,
             "GH Issue #18865 'XaaXbbXb' - pattern matches");
         $x = "($-[0],$+[0])";
         ok( "XaaXbbXb" =~ /(?:X((?{})[ab])?\1)+/,
             "GH Issue #18865 'XaaXbbXb' - deoptimized pattern matches");
         $y = "($-[0],$+[0])";
-        is( $y, $x,
-            "GH Issue #18865 'XaaXbbXb' - test optimization");
+        {
+            local $::TODO = "Not Yet Implemented";
+            is( $y, $x,
+                "GH Issue #18865 'XaaXbbXb' - test optimization");
+        }
     }
 } # End of sub run_tests
 
diff --git a/t/re/re_tests b/t/re/re_tests
index 7379a39787..bfccaa04ed 100644
--- a/t/re/re_tests
+++ b/t/re/re_tests
@@ -478,7 +478,7 @@ a(?:b|c|d)+(.)	acdbcdbe	y	$1	e
 a(?:b|c|d){2}(.)	acdbcdbe	y	$1	b
 a(?:b|c|d){4,5}(.)	acdbcdbe	y	$1	b
 a(?:b|c|d){4,5}?(.)	acdbcdbe	y	$1	d
-((foo)|(bar))*	foobar	Ty	$1-$2-$3	bar--bar		# was bar-foo-bar prior to 5.37.7
+((foo)|(bar))*	foobar	y	$1-$2-$3	bar--bar		# was bar-foo-bar prior to 5.37.7
 :(?:	-	c	-	Sequence (? incomplete
 a(?:b|c|d){6,7}(.)	acdbcdbe	y	$1	e
 a(?:b|c|d){6,7}?(.)	acdbcdbe	y	$1	e
@@ -501,7 +501,7 @@ a(?:b|(c|e){1,2}?|d)+?(.)	ace	y	$1$2	ce
 ((a{4})+)	aaaaaaaaa	y	$1	aaaaaaaa
 (((aa){2})+)	aaaaaaaaaa	y	$1	aaaaaaaa
 (((a{2}){2})+)	aaaaaaaaaa	y	$1	aaaaaaaa
-(?:(f)(o)(o)|(b)(a)(r))*	foobar	Ty	$1:$2:$3:$4:$5:$6	:::b:a:r
+(?:(f)(o)(o)|(b)(a)(r))*	foobar	y	$1:$2:$3:$4:$5:$6	:::b:a:r
 (?<=a)b	ab	y	$&	b
 (?<=af?)b	ab	y	$&	b
 (?<=a)b	cb	n	-	-
@@ -964,8 +964,8 @@ tt+$	xxxtt	y	-	-
 (?i)		y	-	-
 (?a:((?u)\w)\W)	\xC0\xC0	y	$&	\xC0\xC0
 '(?!\A)x'm	a\nxb\n	y	-	-
-^(a(b)?)+$	aba	y	-$1-$2-	-a--		# !normal
-^(aa(bb)?)+$	aabbaa	y	-$1-$2-	-aa--		# !normal
+^(a(b)?)+$	aba	y	-$1-$2-	-a--		
+^(aa(bb)?)+$	aabbaa	y	-$1-$2-	-aa--		
 '^.{9}abc.*\n'm	123\nabcabcabcabc\n	y	-	-
 ^(a)?a$	a	y	-$1-	--
 ^(a)?(?(1)a|b)+$	a	n	-	-
@@ -2139,7 +2139,8 @@ AB\s+\x{100}	AB \x{100}X	y	-	-
 /(([ab]+)|([cd]+)|([ef]+))+/	ace	y	$1-$2-$3-$4=$&	e---e=ace
 /(([ab]+)|([cd]+)|([ef]+))+/	aceb	Ty	$1-$2-$3-$4=$&	b-b--=aceb
 /(([ab]+)|([cd]+)|([ef]+))+/	acebd	Ty	$1-$2-$3-$4=$&	d--d-=acebd
-/(([ab]+)|([cd]+)|([ef]+))+/	acebdf	Ty	$1-$2-$3-$4=$&	f---f=acebdf
+/(([ab]+)|([cd]+)|([ef]+))+/	acebdf	y	$1-$2-$3-$4=$&	f---f=acebdf
+/((a)(b)(c)|(a)(b)|(a))+/	abcaba	y	$1+$2-$3-$4+$5-$6+$7=$&	a+--+-+a=abcaba
 # Keep these lines at the end of the file
 # pat	string	y/n/etc	expr	expected-expr	skip-reason	comment
 # vim: softtabstop=0 noexpandtab