Upgrade preg_quote() to support Unicode.

diff --git a/ext/pcre/php_pcre.c b/ext/pcre/php_pcre.c
index 8ae28e4f92d40ee56d5f2e7dd73a5a867334dc00..65adf400002da2e4f6d9ae2e413d2ba54b2b4af6 100644 (file)
--- a/ext/pcre/php_pcre.c
+++ b/ext/pcre/php_pcre.c
@@ -1684,13 +1684,13 @@ PHP_FUNCTION(preg_quote)
        char    *out_str,               /* Output string with quoted characters */
                        *p,                             /* Iterator for input string */
                        *q,                             /* Iterator for output string */
-                        delim_char=0,  /* Delimiter character to be quoted */
                         c;                             /* Current character */
+       UChar32  delim_char=0;  /* Delimiter character to be quoted */
        zend_bool quote_delim = 0; /* Whether to quote additional delim char */
        
        /* Get the arguments and check for errors */
-       if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "s|s", &in_str, &in_str_len,
-                                                         &delim, &delim_len) == FAILURE) {
+       if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "s&|s&", &in_str, &in_str_len, UG(utf8_conv),
+                                                         &delim, &delim_len, UG(utf8_conv)) == FAILURE) {
                return;
        }
        
@@ -1698,17 +1698,24 @@ PHP_FUNCTION(preg_quote)
 
        /* Nothing to do if we got an empty string */
        if (in_str == in_str_end) {
-               RETURN_EMPTY_STRING();
+               RETURN_EMPTY_TEXT();
        }
 
        if (delim && *delim) {
-               delim_char = delim[0];
+               if (UG(unicode)) {
+                       U8_GET(delim, 0, 0, delim_len, delim_char);
+               } else {
+                       delim_char = (UChar32)delim[0];
+               }
                quote_delim = 1;
        }
        
        /* Allocate enough memory so that even if each character
-          is quoted, we won't run out of room */
-       out_str = safe_emalloc(4, in_str_len, 1);
+          is quoted, we won't run out of room. In Unicode mode, the longest UTF-8
+          sequence is 4 bytes, so the multiplier is (4+1). In non-Unicode mode, we
+          have to assume that any character can be '\0', which needs 4 chars to
+          be escaped. */
+       out_str = safe_emalloc(UG(unicode)?5:4, in_str_len, 1);
        
        /* Go through the string and quote necessary characters */
        for(p = in_str, q = out_str; p != in_str_end; p++) {
@@ -1745,16 +1752,28 @@ PHP_FUNCTION(preg_quote)
                                break;
 
                        default:
-                               if (quote_delim && c == delim_char)
-                                       *q++ = '\\';
-                               *q++ = c;
+                               if ((UChar32)(unsigned char)c > 0x7f) { /* non-ASCII char */
+                                       int tmp = 0;
+                                       UChar32 cp = 0;
+                                       U8_NEXT(p, tmp, in_str_end-p, cp);
+                                       if (quote_delim && cp == delim_char) {
+                                               *q++ = '\\';
+                                       }
+                                       memcpy(q, p, tmp);
+                                       q += tmp;
+                                       p += tmp-1; /* going to be incremented by the loop */
+                               } else {
+                                       if (quote_delim && c == delim_char)
+                                                       *q++ = '\\';
+                                       *q++ = c;
+                               }
                                break;
                }
        }
        *q = '\0';
        
        /* Reallocate string and return it */
-       RETVAL_STRINGL(erealloc(out_str, q - out_str + 1), q - out_str, 0);
+       RETVAL_UTF8_STRINGL(erealloc(out_str, q - out_str + 1), q - out_str, 0);
 }
 /* }}} */