Normalize mb_ereg() return value

mb_ereg()/mb_eregi() currently have an inconsistent return value
based on whether the $matches parameter is passed or not:

> Returns the byte length of the matched string if a match for
> pattern was found in string, or FALSE if no matches were found
> or an error occurred.
>
> If the optional parameter regs was not passed or the length of
> the matched string is 0, this function returns 1.

Coupling this behavior to the $matches parameter doesn't make sense
-- we know the match length either way, there is no technical
reason to distinguish them. However, returning the match length
is not particularly useful either, especially due to the need to
convert 0-length into 1-length to satisfy "truthy" checks. We
could always return 1, which would kind of match the behavior of
preg_match() -- however, preg_match() actually returns the number
of matches, which is 0 or 1 for preg_match(), while false signals
an error. However, mb_ereg() returns false both for no match and
for an error. This would result in an odd 1|false return value.

The patch canonicalizes mb_ereg() to always return a boolean,
where true indicates a match and false indicates no match or error.
This also matches the behavior of the mb_ereg_match() and
mb_ereg_search() functions.

This fixes the default value integrity violation in PHP 8.

Closes GH-6331.

15 files changed, 71 insertions, 77 deletions

diff --git a/ext/mbstring/mbstring.stub.php b/ext/mbstring/mbstring.stub.php
index 14bf595fcf..46a4c59bd5 100644
--- a/ext/mbstring/mbstring.stub.php
+++ b/ext/mbstring/mbstring.stub.php
@@ -93,10 +93,10 @@ function mb_chr(int $codepoint, ?string $encoding = null): string|false {}
 function mb_regex_encoding(?string $encoding = null): string|bool {}
 
 /** @param array $matches */
-function mb_ereg(string $pattern, string $string, &$matches = null): int|false {}
+function mb_ereg(string $pattern, string $string, &$matches = null): bool {}
 
 /** @param array $matches */
-function mb_eregi(string $pattern, string $string, &$matches = null): int|false {}
+function mb_eregi(string $pattern, string $string, &$matches = null): bool {}
 
 function mb_ereg_replace(string $pattern, string $replacement, string $string, ?string $options = null): string|false|null {}
 
diff --git a/ext/mbstring/mbstring_arginfo.h b/ext/mbstring/mbstring_arginfo.h
index ca91c5cd6f..46cc3adf77 100644
--- a/ext/mbstring/mbstring_arginfo.h
+++ b/ext/mbstring/mbstring_arginfo.h
@@ -1,5 +1,5 @@
 /* This is a generated file, edit the .stub.php file instead.
- * Stub hash: 3e5b45cf71fe75bde026062816cb28eceea4aa38 */
+ * Stub hash: 51f0769423c046d612adf81091192165ad265456 */
 
 ZEND_BEGIN_ARG_WITH_RETURN_TYPE_MASK_EX(arginfo_mb_language, 0, 0, MAY_BE_STRING|MAY_BE_BOOL)
 	ZEND_ARG_TYPE_INFO_WITH_DEFAULT_VALUE(0, language, IS_STRING, 1, "null")
@@ -205,7 +205,7 @@ ZEND_END_ARG_INFO()
 #endif
 
 #if defined(HAVE_MBREGEX)
-ZEND_BEGIN_ARG_WITH_RETURN_TYPE_MASK_EX(arginfo_mb_ereg, 0, 2, MAY_BE_LONG|MAY_BE_FALSE)
+ZEND_BEGIN_ARG_WITH_RETURN_TYPE_INFO_EX(arginfo_mb_ereg, 0, 2, _IS_BOOL, 0)
 	ZEND_ARG_TYPE_INFO(0, pattern, IS_STRING, 0)
 	ZEND_ARG_TYPE_INFO(0, string, IS_STRING, 0)
 	ZEND_ARG_INFO_WITH_DEFAULT_VALUE(1, matches, "null")
diff --git a/ext/mbstring/php_mbregex.c b/ext/mbstring/php_mbregex.c
index 45aac7baf2..e87a7c6131 100644
--- a/ext/mbstring/php_mbregex.c
+++ b/ext/mbstring/php_mbregex.c
@@ -891,7 +891,7 @@ static void _php_mb_regex_ereg_exec(INTERNAL_FUNCTION_PARAMETERS, int icase)
 	size_t arg_pattern_len, string_len;
 	php_mb_regex_t *re;
 	OnigRegion *regs = NULL;
-	int i, match_len, beg, end;
+	int i, beg, end;
 	OnigOptionType options;
 	char *str;
 
@@ -938,11 +938,8 @@ static void _php_mb_regex_ereg_exec(INTERNAL_FUNCTION_PARAMETERS, int icase)
 		goto out;
 	}
 
-	match_len = 1;
 	str = string;
 	if (array != NULL) {
-
-		match_len = regs->end[0] - regs->beg[0];
 		for (i = 0; i < regs->num_regs; i++) {
 			beg = regs->beg[i];
 			end = regs->end[i];
@@ -959,10 +956,7 @@ static void _php_mb_regex_ereg_exec(INTERNAL_FUNCTION_PARAMETERS, int icase)
 		}
 	}
 
-	if (match_len == 0) {
-		match_len = 1;
-	}
-	RETVAL_LONG(match_len);
+	RETVAL_TRUE;
 out:
 	if (regs != NULL) {
 		onig_region_free(regs, 1);
diff --git a/ext/mbstring/tests/bug78633.phpt b/ext/mbstring/tests/bug78633.phpt
index a1d6dd5602..3114c6feb2 100644
--- a/ext/mbstring/tests/bug78633.phpt
+++ b/ext/mbstring/tests/bug78633.phpt
@@ -8,7 +8,7 @@ if (!function_exists('mb_eregi')) die('skip mb_eregi function not available');
 --FILE--
 <?php
 $res = mb_eregi(".+Isssǰ", ".+Isssǰ");
-if ($res === 1 || $res === false) {
+if (is_bool($res)) {
     echo "ok\n";
 } else {
     var_dump($res);
diff --git a/ext/mbstring/tests/mb_ereg-compat-02.phpt b/ext/mbstring/tests/mb_ereg-compat-02.phpt
index 2f40422543..e1ec87f644 100644
--- a/ext/mbstring/tests/mb_ereg-compat-02.phpt
+++ b/ext/mbstring/tests/mb_ereg-compat-02.phpt
@@ -19,7 +19,7 @@ function_exists('mb_ereg') or die("skip mb_ereg() is not available in this build
   echo "\n";
 ?>
 --EXPECT--
-32
+1
 This is a nice and simple string
 is
 is
diff --git a/ext/mbstring/tests/mb_ereg.phpt b/ext/mbstring/tests/mb_ereg.phpt
index bc95a5fbaa..91182dd653 100644
--- a/ext/mbstring/tests/mb_ereg.phpt
+++ b/ext/mbstring/tests/mb_ereg.phpt
@@ -31,15 +31,15 @@ output_handler=
     }
 ?>
 --EXPECT--
-(15)6162632064656620676869206a6b6c2064656620676869206a6b6c
-(27)a3e1a3e2a3e320a4a2a4aaa4a420a4aba4b3a4ca20a4efa4f1a4f320a3e1a3e2a3e320a4a2a4aaa4a420a4ab20a4b3a4ca20a4efa4f1a4f3
-(27)a3eda3faa3f8a3e6a3f020a4a6a4aaa4ab20a4ada4ab20a4f2a4f020a3eda3faa3f8a3e6a3f020a4a6a4aaa4ab2020a4ada4ab20a4f2a4f0
-(15)6162632064656620676869206a6b6c2064656620676869206a6b6c
-(27)a3e1a3e2a3e320a4a2a4aaa4a420a4aba4b3a4ca20a4efa4f1a4f320a3e1a3e2a3e320a4a2a4aaa4a420a4ab20a4b3a4ca20a4efa4f1a4f3
-(27)a3eda3faa3f8a3e6a3f020a4a6a4aaa4ab20a4ada4ab20a4f2a4f020a3eda3faa3f8a3e6a3f020a4a6a4aaa4ab2020a4ada4ab20a4f2a4f0
-(15)6162632064656620676869206a6b6c2064656620676869206a6b6c
-(27)a3e1a3e2a3e320a4a2a4aaa4a420a4aba4b3a4ca20a4efa4f1a4f320a3e1a3e2a3e320a4a2a4aaa4a420a4ab20a4b3a4ca20a4efa4f1a4f3
-(27)a3eda3faa3f8a3e6a3f020a4a6a4aaa4ab20a4ada4ab20a4f2a4f020a3eda3faa3f8a3e6a3f020a4a6a4aaa4ab2020a4ada4ab20a4f2a4f0
-(15)6162632064656620676869206a6b6c2064656620676869206a6b6c
-(39)a3e1a3e2a3e320a4a2a4aaa4a420a4aba4b3a4ca20a4efa4f1a4f320a3e1a3e2a3e320a4a2a4aaa4a420a4ab20a4b3a4ca20a4efa4f1a4f3
-(39)a3eda3faa3f8a3e6a3f020a4a6a4aaa4ab20a4ada4ab20a4f2a4f020a3eda3faa3f8a3e6a3f020a4a6a4aaa4ab2020a4ada4ab20a4f2a4f0
+(1)6162632064656620676869206a6b6c2064656620676869206a6b6c
+(1)a3e1a3e2a3e320a4a2a4aaa4a420a4aba4b3a4ca20a4efa4f1a4f320a3e1a3e2a3e320a4a2a4aaa4a420a4ab20a4b3a4ca20a4efa4f1a4f3
+(1)a3eda3faa3f8a3e6a3f020a4a6a4aaa4ab20a4ada4ab20a4f2a4f020a3eda3faa3f8a3e6a3f020a4a6a4aaa4ab2020a4ada4ab20a4f2a4f0
+(1)6162632064656620676869206a6b6c2064656620676869206a6b6c
+(1)a3e1a3e2a3e320a4a2a4aaa4a420a4aba4b3a4ca20a4efa4f1a4f320a3e1a3e2a3e320a4a2a4aaa4a420a4ab20a4b3a4ca20a4efa4f1a4f3
+(1)a3eda3faa3f8a3e6a3f020a4a6a4aaa4ab20a4ada4ab20a4f2a4f020a3eda3faa3f8a3e6a3f020a4a6a4aaa4ab2020a4ada4ab20a4f2a4f0
+(1)6162632064656620676869206a6b6c2064656620676869206a6b6c
+(1)a3e1a3e2a3e320a4a2a4aaa4a420a4aba4b3a4ca20a4efa4f1a4f320a3e1a3e2a3e320a4a2a4aaa4a420a4ab20a4b3a4ca20a4efa4f1a4f3
+(1)a3eda3faa3f8a3e6a3f020a4a6a4aaa4ab20a4ada4ab20a4f2a4f020a3eda3faa3f8a3e6a3f020a4a6a4aaa4ab2020a4ada4ab20a4f2a4f0
+(1)6162632064656620676869206a6b6c2064656620676869206a6b6c
+(1)a3e1a3e2a3e320a4a2a4aaa4a420a4aba4b3a4ca20a4efa4f1a4f320a3e1a3e2a3e320a4a2a4aaa4a420a4ab20a4b3a4ca20a4efa4f1a4f3
+(1)a3eda3faa3f8a3e6a3f020a4a6a4aaa4ab20a4ada4ab20a4f2a4f020a3eda3faa3f8a3e6a3f020a4a6a4aaa4ab2020a4ada4ab20a4f2a4f0
diff --git a/ext/mbstring/tests/mb_ereg_basic.phpt b/ext/mbstring/tests/mb_ereg_basic.phpt
index 9ec2ff754d..7f04855eb0 100644
--- a/ext/mbstring/tests/mb_ereg_basic.phpt
+++ b/ext/mbstring/tests/mb_ereg_basic.phpt
@@ -76,17 +76,17 @@ Regex encoding set to utf-8
 
 **-- ASCII String --**
 -- Without $regs argument--
-int(1)
-int(1)
+bool(true)
+bool(true)
 --With $regs argument --
-int(36)
+bool(true)
 array(2) {
   [0]=>
   string(48) "VGhpcyBpcyBhbiBFbmdsaXNoIHN0cmluZy4gMDEyMzQ1Njc4"
   [1]=>
   string(24) "VGhpcyBpcyBhbiBFbmdsaXM="
 }
-int(17)
+bool(true)
 array(1) {
   [0]=>
   string(24) "VGhpcyBpcyBhbiBFbmdsaXM="
@@ -94,10 +94,10 @@ array(1) {
 
 **-- Multibyte String --**
 -- Without $regs argument --
-int(1)
+bool(true)
 bool(false)
 -- With $regs argument --
-int(35)
+bool(true)
 array(3) {
   [0]=>
   string(48) "5pel5pys6Kqe44OG44Kt44K544OI44Gn44GZ44CCMDEyMzQ="
diff --git a/ext/mbstring/tests/mb_ereg_variation3.phpt b/ext/mbstring/tests/mb_ereg_variation3.phpt
index f667d64bdd..6867165ec2 100644
--- a/ext/mbstring/tests/mb_ereg_variation3.phpt
+++ b/ext/mbstring/tests/mb_ereg_variation3.phpt
@@ -69,84 +69,84 @@ echo "Done";
 *** Testing mb_ereg() : variation ***
 
 -- Iteration 1 --
-int(3)
+bool(true)
 array(1) {
   [0]=>
   string(4) "YUIx"
 }
 
 -- Iteration 2 --
-int(4)
+bool(true)
 array(1) {
   [0]=>
   string(8) "YUJjRA=="
 }
 
 -- Iteration 3 --
-int(4)
+bool(true)
 array(1) {
   [0]=>
   string(8) "YWIvPQ=="
 }
 
 -- Iteration 4 --
-int(2)
+bool(true)
 array(1) {
   [0]=>
   string(4) "IAk="
 }
 
 -- Iteration 5 --
-int(3)
+bool(true)
 array(1) {
   [0]=>
   string(4) "MjM0"
 }
 
 -- Iteration 6 --
-int(9)
+bool(true)
 array(1) {
   [0]=>
   string(12) "5pel5pys6Kqe"
 }
 
 -- Iteration 7 --
-int(4)
+bool(true)
 array(1) {
   [0]=>
   string(8) "Zmpkcw=="
 }
 
 -- Iteration 8 --
-int(9)
+bool(true)
 array(1) {
   [0]=>
   string(12) "5pel5pys6Kqe"
 }
 
 -- Iteration 9 --
-int(5)
+bool(true)
 array(1) {
   [0]=>
   string(8) "LiEiKkA="
 }
 
 -- Iteration 10 --
-int(1)
+bool(true)
 array(1) {
   [0]=>
   string(4) "CQ=="
 }
 
 -- Iteration 11 --
-int(5)
+bool(true)
 array(1) {
   [0]=>
   string(8) "SURTSlY="
 }
 
 -- Iteration 12 --
-int(4)
+bool(true)
 array(1) {
   [0]=>
   string(8) "M2I1RA=="
diff --git a/ext/mbstring/tests/mb_ereg_variation4.phpt b/ext/mbstring/tests/mb_ereg_variation4.phpt
index 07c61971dd..7883361ce9 100644
--- a/ext/mbstring/tests/mb_ereg_variation4.phpt
+++ b/ext/mbstring/tests/mb_ereg_variation4.phpt
@@ -72,21 +72,21 @@ echo "Done";
 *** Testing mb_ereg() : usage variations ***
 
 -- Iteration 1 --
-int(47)
+bool(true)
 array(1) {
   [0]=>
   string(64) "5pel5pys6Kqe44OG44Kt44K544OI44Gn44GZMDEyMzTvvJXvvJbvvJfvvJjvvJk="
 }
 
 -- Iteration 2 --
-int(27)
+bool(true)
 array(1) {
   [0]=>
   string(36) "5pel5pys6Kqe44OG44Kt44K544OI44Gn44GZ"
 }
 
 -- Iteration 3 --
-int(5)
+bool(true)
 array(1) {
   [0]=>
   string(8) "MDEyMzQ="
@@ -99,14 +99,14 @@ bool(false)
 bool(false)
 
 -- Iteration 6 --
-int(20)
+bool(true)
 array(1) {
   [0]=>
   string(28) "MDEyMzTvvJXvvJbvvJfvvJjvvJk="
 }
 
 -- Iteration 7 --
-int(50)
+bool(true)
 array(1) {
   [0]=>
   string(68) "5pel5pys6Kqe44OG44Kt44K544OI44Gn44GZMDEyMzTvvJXvvJbvvJfvvJjvvJnjgII="
@@ -116,14 +116,14 @@ array(1) {
 bool(false)
 
 -- Iteration 9 --
-int(50)
+bool(true)
 array(1) {
   [0]=>
   string(68) "5pel5pys6Kqe44OG44Kt44K544OI44Gn44GZMDEyMzTvvJXvvJbvvJfvvJjvvJnjgII="
 }
 
 -- Iteration 10 --
-int(3)
+bool(true)
 array(1) {
   [0]=>
   string(4) "44CC"
@@ -136,7 +136,7 @@ bool(false)
 bool(false)
 
 -- Iteration 13 --
-int(5)
+bool(true)
 array(1) {
   [0]=>
   string(8) "MDEyMzQ="
diff --git a/ext/mbstring/tests/mb_ereg_variation5.phpt b/ext/mbstring/tests/mb_ereg_variation5.phpt
index 31fd50d97c..fd08e56993 100644
--- a/ext/mbstring/tests/mb_ereg_variation5.phpt
+++ b/ext/mbstring/tests/mb_ereg_variation5.phpt
@@ -59,17 +59,17 @@ function base64_encode_var_dump($regs) {
 --EXPECT--
 *** Testing mb_ereg() : usage variations ***
 
-ASCII String without $regs arg:		int(1)
+ASCII String without $regs arg:		bool(true)
 ASCII String with $regs arg:
-int(38)
+bool(true)
 array(1) {
   [0]=>
   string(52) "VGhpcyBpcyBhbiBFbmdsaXNoIHN0cmluZy4gMDEyMzQ1Njc4OS4="
 }
 
-Multibyte String without $regs arg:	int(1)
+Multibyte String without $regs arg:	bool(true)
 Multubyte String with $regs arg:
-int(53)
+bool(true)
 array(1) {
   [0]=>
   string(72) "5pel5pys6Kqe44OG44Kt44K544OI44Gn44GZ44CCMDEyMzTvvJXvvJbvvJfvvJjvvJnjgII="
diff --git a/ext/mbstring/tests/mb_ereg_variation6.phpt b/ext/mbstring/tests/mb_ereg_variation6.phpt
index dc91aef209..ec1a70c536 100644
--- a/ext/mbstring/tests/mb_ereg_variation6.phpt
+++ b/ext/mbstring/tests/mb_ereg_variation6.phpt
@@ -78,13 +78,13 @@ Regex encoding set to utf-8
 
 --** Pattern is: \w+ **--
 -- ASCII String: --
-int(4)
+bool(true)
 array(1) {
   [0]=>
   string(8) "VGhpcw=="
 }
 -- Multibyte String: --
-int(27)
+bool(true)
 array(1) {
   [0]=>
   string(36) "5pel5pys6Kqe44OG44Kt44K544OI44Gn44GZ"
@@ -92,13 +92,13 @@ array(1) {
 
 --** Pattern is: \W+ **--
 -- ASCII String: --
-int(1)
+bool(true)
 array(1) {
   [0]=>
   string(4) "IA=="
 }
 -- Multibyte String: --
-int(3)
+bool(true)
 array(1) {
   [0]=>
   string(4) "44CC"
@@ -106,7 +106,7 @@ array(1) {
 
 --** Pattern is: \s+ **--
 -- ASCII String: --
-int(1)
+bool(true)
 array(1) {
   [0]=>
   string(4) "IA=="
@@ -117,13 +117,13 @@ NULL
 
 --** Pattern is: \S+ **--
 -- ASCII String: --
-int(4)
+bool(true)
 array(1) {
   [0]=>
   string(8) "VGhpcw=="
 }
 -- Multibyte String: --
-int(53)
+bool(true)
 array(1) {
   [0]=>
   string(72) "5pel5pys6Kqe44OG44Kt44K544OI44Gn44GZ44CCMDEyMzTvvJXvvJbvvJfvvJjvvJnjgII="
@@ -131,13 +131,13 @@ array(1) {
 
 --** Pattern is: \d+ **--
 -- ASCII String: --
-int(10)
+bool(true)
 array(1) {
   [0]=>
   string(16) "MDEyMzQ1Njc4OQ=="
 }
 -- Multibyte String: --
-int(20)
+bool(true)
 array(1) {
   [0]=>
   string(28) "MDEyMzTvvJXvvJbvvJfvvJjvvJk="
@@ -145,13 +145,13 @@ array(1) {
 
 --** Pattern is: \D+ **--
 -- ASCII String: --
-int(27)
+bool(true)
 array(1) {
   [0]=>
   string(36) "VGhpcyBpcyBhbiBFbmdsaXNoIHN0cmluZy4g"
 }
 -- Multibyte String: --
-int(30)
+bool(true)
 array(1) {
   [0]=>
   string(40) "5pel5pys6Kqe44OG44Kt44K544OI44Gn44GZ44CC"
@@ -159,13 +159,13 @@ array(1) {
 
 --** Pattern is: \b **--
 -- ASCII String: --
-int(1)
+bool(true)
 array(1) {
   [0]=>
   bool(false)
 }
 -- Multibyte String: --
-int(1)
+bool(true)
 array(1) {
   [0]=>
   bool(false)
@@ -173,13 +173,13 @@ array(1) {
 
 --** Pattern is: \B **--
 -- ASCII String: --
-int(1)
+bool(true)
 array(1) {
   [0]=>
   bool(false)
 }
 -- Multibyte String: --
-int(1)
+bool(true)
 array(1) {
   [0]=>
   bool(false)
diff --git a/ext/mbstring/tests/mb_ereg_variation7.phpt b/ext/mbstring/tests/mb_ereg_variation7.phpt
index 76e6dce08f..d54011f9f8 100644
--- a/ext/mbstring/tests/mb_ereg_variation7.phpt
+++ b/ext/mbstring/tests/mb_ereg_variation7.phpt
@@ -60,7 +60,7 @@ echo "Done";
 --EXPECT--
 *** Testing mb_ereg() : usage variations ***
 Regex encoding set to utf-8
-int(38)
+bool(true)
 array(5) {
   [0]=>
   string(52) "VGhpcyBpcyBhbiBFbmdsaXNoIHN0cmluZy4gMDEyMzQ1Njc4OS4="
@@ -73,7 +73,7 @@ array(5) {
   [4]=>
   string(4) "ODk="
 }
-int(64)
+bool(true)
 array(5) {
   [0]=>
   string(88) "zpHPhc+Ez4wgzrXOr869zrHOuSDOtc67zrvOt869zrnOus+MIM66zrXOr868zrXOvc6/LiAwMTIzNDU2Nzg5Lg=="
diff --git a/ext/mbstring/tests/mb_eregi.phpt b/ext/mbstring/tests/mb_eregi.phpt
index ced6fadcdf..a12ab70116 100644
--- a/ext/mbstring/tests/mb_eregi.phpt
+++ b/ext/mbstring/tests/mb_eregi.phpt
@@ -15,6 +15,6 @@ var_dump(mb_eregi('xyzp', 'XYZ'));
 var_dump(mb_eregi('ö', 'Öäü'));
 ?>
 --EXPECT--
-int(1)
+bool(true)
 bool(false)
-int(1)
+bool(true)
diff --git a/ext/mbstring/tests/mbregex_stack_limit.phpt b/ext/mbstring/tests/mbregex_stack_limit.phpt
index ccd763e2b5..ade2f1e288 100644
--- a/ext/mbstring/tests/mbregex_stack_limit.phpt
+++ b/ext/mbstring/tests/mbregex_stack_limit.phpt
@@ -26,5 +26,5 @@ echo 'OK';
 --EXPECT--
 bool(false)
 bool(false)
-int(1)
+bool(true)
 OK
diff --git a/ext/mbstring/tests/retry_limit.phpt b/ext/mbstring/tests/retry_limit.phpt
index 2da9010f8e..251e5111ef 100644
--- a/ext/mbstring/tests/retry_limit.phpt
+++ b/ext/mbstring/tests/retry_limit.phpt
@@ -19,5 +19,5 @@ var_dump(mb_ereg($regex, $str));
 
 ?>
 --EXPECT--
-int(1)
+bool(true)
 bool(false)