Optimize strcspn (#12594)

Author: ndossche

UPGRADING

@@ -48,6 +48,8 @@ PHP 8.4 UPGRADE NOTES
   . round() now validates the value of the $mode parameter and throws a ValueError
     for invalid modes. Previously invalid modes would have been interpreted as
     PHP_ROUND_HALF_UP.
+  . strcspn() with empty $characters now returns the length of the string instead
+    of incorrectly stopping at the first NUL character. See GH-12592.
 
 - XML:
   . The xml_set_*_handler() functions now declare and check for an effective

ext/standard/string.c

@@ -1596,16 +1596,15 @@ PHPAPI char *php_stristr(char *s, char *t, size_t s_len, size_t t_len)
 }
 /* }}} */
 
-/* {{{ php_strspn */
-PHPAPI size_t php_strspn(const char *haystack, const char *characters, const char *haystack_end, const char *characters_end)
+static size_t php_strspn_strcspn_common(const char *haystack, const char *characters, const char *haystack_end, const char *characters_end, bool must_match)
 {
 	/* Fast path for short strings.
 	 * The table lookup cannot be faster in this case because we not only have to compare, but also build the table.
 	 * We only compare in this case.
 	 * Empirically tested that the table lookup approach is only beneficial if characters is longer than 1 character. */
 	if (characters_end - characters == 1) {
 		const char *ptr = haystack;
-		while (ptr < haystack_end && *ptr == *characters) {
+		while (ptr < haystack_end && (*ptr == *characters) == must_match) {
 			ptr++;
 		}
 		return ptr - haystack;
@@ -1626,30 +1625,24 @@ PHPAPI size_t php_strspn(const char *haystack, const char *characters, const cha
 	}
 
 	const char *ptr = haystack;
-	while (ptr < haystack_end && table[(unsigned char) *ptr]) {
+	while (ptr < haystack_end && table[(unsigned char) *ptr] == must_match) {
 		ptr++;
 	}
 
 	return ptr - haystack;
 }
+
+/* {{{ php_strspn */
+PHPAPI size_t php_strspn(const char *haystack, const char *characters, const char *haystack_end, const char *characters_end)
+{
+	return php_strspn_strcspn_common(haystack, characters, haystack_end, characters_end, true);
+}
 /* }}} */
 
 /* {{{ php_strcspn */
-PHPAPI size_t php_strcspn(const char *s1, const char *s2, const char *s1_end, const char *s2_end)
+PHPAPI size_t php_strcspn(const char *haystack, const char *characters, const char *haystack_end, const char *characters_end)
 {
-	const char *p, *spanp;
-	char c = *s1;
-
-	for (p = s1;;) {
-		spanp = s2;
-		do {
-			if (*spanp == c || p == s1_end) {
-				return p - s1;
-			}
-		} while (spanp++ < (s2_end - 1));
-		c = *++p;
-	}
-	/* NOTREACHED */
+	return php_strspn_strcspn_common(haystack, characters, haystack_end, characters_end, false);
 }
 /* }}} */
 

ext/standard/tests/strings/bug39032.phpt

@@ -12,7 +12,7 @@ echo "Done\n";
 ?>
 --EXPECT--
 int(1)
-int(0)
+int(1)
 int(1)
 int(1)
 Done

ext/standard/tests/strings/strcspn_variation10.phpt

@@ -156,8 +156,8 @@ int(21)
 int(16)
 
 -- Itearation 9 --
-int(5)
-int(5)
+int(12)
+int(12)
 int(2)
 int(2)
 int(12)
@@ -180,8 +180,8 @@ int(16)
 int(5)
 
 -- Itearation 11 --
-int(0)
-int(0)
+int(2)
+int(2)
 int(2)
 int(2)
 int(2)
@@ -192,8 +192,8 @@ int(2)
 int(2)
 
 -- Itearation 12 --
-int(0)
-int(0)
+int(13)
+int(13)
 int(3)
 int(3)
 int(13)
@@ -204,8 +204,8 @@ int(13)
 int(13)
 
 -- Itearation 13 --
-int(0)
-int(0)
+int(14)
+int(14)
 int(3)
 int(3)
 int(14)
@@ -216,8 +216,8 @@ int(14)
 int(6)
 
 -- Itearation 14 --
-int(5)
-int(5)
+int(11)
+int(11)
 int(2)
 int(2)
 int(11)
@@ -228,8 +228,8 @@ int(11)
 int(11)
 
 -- Itearation 15 --
-int(5)
-int(5)
+int(11)
+int(11)
 int(2)
 int(2)
 int(11)
@@ -240,8 +240,8 @@ int(11)
 int(11)
 
 -- Itearation 16 --
-int(5)
-int(5)
+int(14)
+int(14)
 int(2)
 int(2)
 int(14)

ext/standard/tests/strings/strcspn_variation11.phpt

@@ -650,20 +650,20 @@ int(0)
 int(16)
 
 -- Iteration 9 --
-int(5)
-int(4)
-int(3)
+int(12)
+int(11)
+int(10)
 int(1)
 int(2)
 int(0)
-int(5)
-int(5)
-int(4)
-int(3)
+int(12)
+int(12)
+int(11)
+int(10)
 int(1)
 int(2)
 int(0)
-int(5)
+int(12)
 int(2)
 int(1)
 int(0)
@@ -794,20 +794,20 @@ int(0)
 int(5)
 
 -- Iteration 11 --
+int(2)
+int(1)
 int(0)
+int(1)
+int(2)
 int(0)
+int(2)
+int(2)
+int(1)
 int(0)
+int(1)
+int(2)
 int(0)
-int(0)
-int(0)
-int(0)
-int(0)
-int(0)
-int(0)
-int(0)
-int(0)
-int(0)
-int(0)
+int(2)
 int(2)
 int(1)
 int(0)
@@ -866,20 +866,20 @@ int(0)
 int(2)
 
 -- Iteration 12 --
-int(0)
-int(5)
-int(4)
-int(0)
+int(13)
+int(12)
+int(11)
 int(1)
+int(2)
 int(0)
-int(0)
-int(0)
-int(5)
-int(4)
-int(0)
+int(13)
+int(13)
+int(12)
+int(11)
 int(1)
+int(2)
 int(0)
-int(0)
+int(13)
 int(3)
 int(2)
 int(1)
@@ -938,20 +938,20 @@ int(0)
 int(13)
 
 -- Iteration 13 --
-int(0)
+int(14)
+int(13)
 int(12)
-int(11)
-int(0)
 int(1)
+int(2)
 int(0)
-int(0)
-int(0)
+int(14)
+int(14)
+int(13)
 int(12)
-int(11)
-int(0)
 int(1)
+int(2)
 int(0)
-int(0)
+int(14)
 int(3)
 int(2)
 int(1)
@@ -1010,20 +1010,20 @@ int(0)
 int(6)
 
 -- Iteration 14 --
-int(5)
-int(4)
-int(3)
+int(11)
+int(10)
+int(9)
 int(1)
 int(2)
 int(0)
-int(5)
-int(5)
-int(4)
-int(3)
+int(11)
+int(11)
+int(10)
+int(9)
 int(1)
 int(2)
 int(0)
-int(5)
+int(11)
 int(2)
 int(1)
 int(0)
@@ -1082,20 +1082,20 @@ int(0)
 int(11)
 
 -- Iteration 15 --
-int(5)
-int(4)
-int(3)
+int(11)
+int(10)
+int(9)
 int(1)
 int(2)
 int(0)
-int(5)
-int(5)
-int(4)
-int(3)
+int(11)
+int(11)
+int(10)
+int(9)
 int(1)
 int(2)
 int(0)
-int(5)
+int(11)
 int(2)
 int(1)
 int(0)
@@ -1154,20 +1154,20 @@ int(0)
 int(11)
 
 -- Iteration 16 --
-int(5)
-int(4)
-int(3)
+int(14)
+int(13)
+int(12)
 int(1)
 int(2)
 int(0)
-int(5)
-int(5)
-int(4)
-int(3)
+int(14)
+int(14)
+int(13)
+int(12)
 int(1)
 int(2)
 int(0)
-int(5)
+int(14)
 int(2)
 int(1)
 int(0)