Fix usearch to ucol

Author: youkidearitai

Diff for: ext/intl/grapheme/grapheme_string.c

@@ -950,7 +950,6 @@ PHP_FUNCTION(grapheme_levenshtein)
 		RETURN_THROWS();
 	}
 
-	zend_long *p1, *p2, *tmp;
 	zend_long c0, c1, c2;
 	zend_long retval;
 	size_t i2;
@@ -962,8 +961,7 @@ PHP_FUNCTION(grapheme_levenshtein)
 	int32_t ustring1_len = 0;
 	int32_t ustring2_len = 0;
 
-	UErrorCode ustatus1 = U_ZERO_ERROR;
-	UErrorCode ustatus2 = U_ZERO_ERROR;
+	UErrorCode ustatus = U_ZERO_ERROR;
 
 	/* When all costs are equal, levenshtein fulfills the requirements of a metric, which means
 	 * that the distance is symmetric. If string1 is shorter than string2 we can save memory (and CPU time)
@@ -977,20 +975,20 @@ PHP_FUNCTION(grapheme_levenshtein)
 	pstr1 = ZSTR_VAL(string1);
 	pstr2 = ZSTR_VAL(string2);
 
-	intl_convert_utf8_to_utf16(&ustring1, &ustring1_len, pstr1, ZSTR_LEN(string1), &ustatus1);
+	intl_convert_utf8_to_utf16(&ustring1, &ustring1_len, pstr1, ZSTR_LEN(string1), &ustatus);
 
-	if (U_FAILURE(ustatus1)) {
-		intl_error_set_code(NULL, ustatus1);
+	if (U_FAILURE(ustatus)) {
+		intl_error_set_code(NULL, ustatus);
 
 		intl_error_set_custom_msg(NULL, "Error converting input string to UTF-16", 0);
 		efree(ustring1);
 		RETURN_FALSE;
 	}
 
-	intl_convert_utf8_to_utf16(&ustring2, &ustring2_len, pstr2, ZSTR_LEN(string2), &ustatus2);
+	intl_convert_utf8_to_utf16(&ustring2, &ustring2_len, pstr2, ZSTR_LEN(string2), &ustatus);
 
-	if (U_FAILURE(ustatus2)) {
-		intl_error_set_code(NULL, ustatus2);
+	if (U_FAILURE(ustatus)) {
+		intl_error_set_code(NULL, ustatus);
 
 		intl_error_set_custom_msg(NULL, "Error converting input string to UTF-16", 0);
 		efree(ustring2);
@@ -1017,30 +1015,64 @@ PHP_FUNCTION(grapheme_levenshtein)
 
 	unsigned char u_break_iterator_buffer1[U_BRK_SAFECLONE_BUFFERSIZE];
 	unsigned char u_break_iterator_buffer2[U_BRK_SAFECLONE_BUFFERSIZE];
-	bi1 = grapheme_get_break_iterator((void*)u_break_iterator_buffer1, &ustatus1);
-	bi2 = grapheme_get_break_iterator((void*)u_break_iterator_buffer2, &ustatus2);
+	bi1 = grapheme_get_break_iterator((void*)u_break_iterator_buffer1, &ustatus);
+	if (U_FAILURE(ustatus)) {
+		intl_error_set_code(NULL, ustatus);
+		intl_error_set_custom_msg(NULL, "Error on grapheme_get_break_iterator for argument #1 ($string1)", 0);
+		efree(ustring2);
+		efree(ustring1);
+		ubrk_close(bi1);
+		RETURN_FALSE;
+	}
 
-	ubrk_setText(bi1, ustring1, ustring1_len, &ustatus1);
+	bi2 = grapheme_get_break_iterator(u_break_iterator_buffer2, &ustatus);
+	if (U_FAILURE(ustatus)) {
+		intl_error_set_code(NULL, ustatus);
+		intl_error_set_custom_msg(NULL, "Error on grapheme_get_break_iterator for argument #2 ($string2)", 0);
+		efree(ustring2);
+		efree(ustring1);
+		ubrk_close(bi2);
+		ubrk_close(bi1);
+		RETURN_FALSE;
+	}
+	ubrk_setText(bi1, ustring1, ustring1_len, &ustatus);
 
-	if (U_FAILURE(ustatus1)) {
-		intl_error_set_code(NULL, ustatus1);
+	if (U_FAILURE(ustatus)) {
+		intl_error_set_code(NULL, ustatus);
 
-		intl_error_set_custom_msg(NULL, "Error on ubrk_setText on ustring1", 0);
+		intl_error_set_custom_msg(NULL, "Error on ubrk_setText for argument #1 ($string1)", 0);
 		efree(ustring2);
 		efree(ustring1);
+		ubrk_close(bi2);
+		ubrk_close(bi1);
 		RETURN_FALSE;
 	}
 
-	ubrk_setText(bi2, ustring2, ustring2_len, &ustatus2);
-	if (U_FAILURE(ustatus2)) {
-		intl_error_set_code(NULL, ustatus2);
+	ubrk_setText(bi2, ustring2, ustring2_len, &ustatus);
+	if (U_FAILURE(ustatus)) {
+		intl_error_set_code(NULL, ustatus);
 
-		intl_error_set_custom_msg(NULL, "Error on ubrk_setText on ustring2", 0);
+		intl_error_set_custom_msg(NULL, "Error on ubrk_setText for argument #2 ($string2)", 0);
 		efree(ustring2);
 		efree(ustring1);
+		ubrk_close(bi2);
+		ubrk_close(bi1);
 		RETURN_FALSE;
 	}
+	UCollator *collator = ucol_open("", &ustatus);
+	if (U_FAILURE(ustatus)) {
+		intl_error_set_code(NULL, ustatus);
 
+		intl_error_set_custom_msg(NULL, "Error on ucol_open", 0);
+		efree(ustring2);
+		efree(ustring1);
+		ubrk_close(bi2);
+		ubrk_close(bi1);
+		ucol_close(collator);
+		RETURN_FALSE;
+	}
+
+	zend_long *p1, *p2, *tmp;
 	p1 = safe_emalloc(strlen_2 + 1, sizeof(zend_long), 0);
 	p2 = safe_emalloc(strlen_2 + 1, sizeof(zend_long), 0);
 
@@ -1052,7 +1084,6 @@ PHP_FUNCTION(grapheme_levenshtein)
 	int32_t current2 = 0;
 	int32_t pos1 = 0;
 	int32_t pos2 = 0;
-	int32_t usrch_pos = 0;
 
 	while (true) {
 		current1 = ubrk_current(bi1);
@@ -1067,37 +1098,8 @@ PHP_FUNCTION(grapheme_levenshtein)
 			if (pos2 == UBRK_DONE) {
 				break;
 			}
-			UStringSearch *srch = usearch_open(ustring1 + current1, pos1 - current1, ustring2 + current2, pos2 - current2, "", NULL, &ustatus2);
-			if (U_FAILURE(ustatus2)) {
-				intl_error_set_code(NULL, ustatus2);
-				intl_error_set_custom_msg(NULL, "Error usearch_open", 0);
-				ubrk_close(bi1);
-				ubrk_close(bi2);
-
-				efree(ustring1);
-				efree(ustring2);
-
-				efree(p1);
-				efree(p2);
-				RETURN_FALSE;
-			}
-			usrch_pos = usearch_first(srch, &ustatus2);
-			if (U_FAILURE(ustatus2)) {
-				intl_error_set_code(NULL, ustatus2);
-				intl_error_set_custom_msg(NULL, "Error usearch_first", 0);
-				ubrk_close(bi1);
-				ubrk_close(bi2);
-
-				efree(ustring1);
-				efree(ustring2);
-
-				efree(p1);
-				efree(p2);
-				RETURN_FALSE;
-			}
-			usearch_close(srch);
-
-			if (usrch_pos != USEARCH_DONE) {
+			if (ucol_strcoll(collator, ustring1 + current1, pos1 - current1, ustring2 + current2, pos2 - current2) == UCOL_EQUAL) {
+				c0 = p1[i2];
 				c0 = p1[i2];
 			} else {
 				c0 = p1[i2] + cost_rep;
@@ -1118,6 +1120,8 @@ PHP_FUNCTION(grapheme_levenshtein)
 		p2 = tmp;
 	}
 
+	ucol_close(collator);
+
 	ubrk_close(bi1);
 	ubrk_close(bi2);