Unicode support in entities/special chars funcs. See the NOTE at the

top.

diff --git a/ext/standard/html.c b/ext/standard/html.c
index 03b55ef5e8755349fc9a58f4002a0d478a066823..db4de38051006ee2343ec14c25d1126132522fff 100644 (file)
--- a/ext/standard/html.c
+++ b/ext/standard/html.c
@@ -29,6 +29,12 @@
  *
  * http://www.w3.org/TR/2002/REC-xhtml1-20020801/dtds.html#h-A2
  * 
+ * UNICODE NOTE:
+ *  The way Unicode support is implemented (namely, IS_UNICODE support) is by
+ *  converting the IS_UNICODE strings to UTF-8 and handing them off to existing
+ *  implementation. This saves on redoing all the code that encodes and decodes
+ *  entities to support UChar*, but it does result in slight performance loss.
+ *  Whoever wants to do this properly, go ahead.
  */
 
 #include "php.h"
@@ -1191,18 +1197,35 @@ PHPAPI char *php_escape_html_entities(char *orig, int oldlen, int *newlen, int a
  */
 static void php_html_entities(INTERNAL_FUNCTION_PARAMETERS, int all)
 {
-       char *str, *hint_charset = NULL;
+       zstr str;
+       char *hint_charset = NULL;
        int str_len, hint_charset_len = 0;
+       char *str_utf8;
+       int str_utf8_len;
        int len;
        long quote_style = ENT_COMPAT;
+       zend_uchar type;
        char *replaced;
 
-       if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "s|ls", &str, &str_len, &quote_style, &hint_charset, &hint_charset_len) == FAILURE) {
+       if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "t|ls", &str, &str_len, &type, &quote_style, &hint_charset, &hint_charset_len) == FAILURE) {
                return;
        }
 
-       replaced = php_escape_html_entities(str, str_len, &len, all, quote_style, hint_charset TSRMLS_CC);
-       RETVAL_STRINGL(replaced, len, 0);
+       if (type == IS_UNICODE) {
+               zend_unicode_to_string(UG(utf8_conv), &str_utf8, &str_utf8_len, str.u, str_len TSRMLS_CC);
+               str.s = str_utf8;
+               str_len = str_utf8_len;
+               hint_charset = "utf-8";
+       }
+
+       replaced = php_escape_html_entities(str.s, str_len, &len, all, quote_style, hint_charset TSRMLS_CC);
+
+       if (type == IS_UNICODE) {
+               RETVAL_U_STRINGL(UG(utf8_conv), replaced, len, ZSTR_AUTOFREE);
+               efree(str_utf8);
+       } else {
+               RETVAL_STRINGL(replaced, len, 0);
+       }
 }
 /* }}} */
 
@@ -1221,7 +1244,7 @@ void register_html_constants(INIT_FUNC_ARGS)
 }
 /* }}} */
 
-/* {{{ proto string htmlspecialchars(string string [, int quote_style][, string charset])
+/* {{{ proto string htmlspecialchars(string string [, int quote_style][, string charset]) U
    Convert special characters to HTML entities */
 PHP_FUNCTION(htmlspecialchars)
 {
@@ -1229,27 +1252,43 @@ PHP_FUNCTION(htmlspecialchars)
 }
 /* }}} */
 
-/* {{{ proto string htmlspecialchars_decode(string string [, int quote_style])
+/* {{{ proto string htmlspecialchars_decode(string string [, int quote_style]) U
    Convert special HTML entities back to characters */
 PHP_FUNCTION(htmlspecialchars_decode)
 {
-       char *str, *new_str, *e, *p;
+       zstr str;
+       char *str_utf8;
+       int str_utf8_len;
+       zend_uchar type;
+       char *new_str, *e, *p;
        int len, j, i, new_len;
        long quote_style = ENT_COMPAT;
        struct basic_entities_dec basic_entities_dec[8];
 
-       if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "s|l", &str, &len, &quote_style) == FAILURE) {
+       if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "t|l", &str, &len, &type, &quote_style) == FAILURE) {
                return;
        }
 
-       new_str = estrndup(str, len);
-       new_len = len;
-       e = new_str + new_len;
+       if (type == IS_UNICODE) {
+               if (!u_memchr(str.u, 0x26 /*'&'*/, len)) {
+                       RETURN_UNICODEL(str.u, len, 1);
+               }
 
-       if (!(p = memchr(new_str, '&', new_len))) {
-               RETURN_STRINGL(new_str, new_len, 0);
+               zend_unicode_to_string(UG(utf8_conv), &str_utf8, &str_utf8_len, str.u, len TSRMLS_CC);
+               new_str = str_utf8;
+               new_len = str_utf8_len;
+               p = memchr(new_str, '&', new_len);
+       } else {
+               new_str = estrndup(str.s, len);
+               new_len = len;
+
+               if (!(p = memchr(new_str, '&', new_len))) {
+                       RETURN_STRINGL(new_str, new_len, 0);
+               }
        }
 
+       e = new_str + new_len;
+
        for (j = 0, i = 0; basic_entities[i].charcode != 0; i++) {
                if (basic_entities[i].flags && !(quote_style & basic_entities[i].flags)) {
                        continue;
@@ -1291,31 +1330,52 @@ done:
        new_len = e - new_str;
 
        new_str[new_len] = '\0';
-       RETURN_STRINGL(new_str, new_len, 0);
+       if (type == IS_UNICODE) {
+               RETVAL_U_STRINGL(UG(utf8_conv), new_str, new_len, ZSTR_AUTOFREE);
+       } else {
+               RETVAL_STRINGL(new_str, new_len, 0);
+       }
 }
 /* }}} */
 
-/* {{{ proto string html_entity_decode(string string [, int quote_style][, string charset])
+/* {{{ proto string html_entity_decode(string string [, int quote_style][, string charset]) U
    Convert all HTML entities to their applicable characters */
 PHP_FUNCTION(html_entity_decode)
 {
-       char *str, *hint_charset = NULL;
+       zstr str;
+       char *hint_charset = NULL;
        int str_len, hint_charset_len, len;
+       char *str_utf8;
+       int str_utf8_len;
+       zend_uchar type;
        long quote_style = ENT_COMPAT;
        char *replaced;
 
-       if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "s|ls", &str, &str_len,
+       if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_CC, "t|ls", &str, &str_len, &type,
                                                          &quote_style, &hint_charset, &hint_charset_len) == FAILURE) {
                return;
        }
 
-       replaced = php_unescape_html_entities(str, str_len, &len, 1, quote_style, hint_charset TSRMLS_CC);
-       RETVAL_STRINGL(replaced, len, 0);
+       if (type == IS_UNICODE) {
+               zend_unicode_to_string(UG(utf8_conv), &str_utf8, &str_utf8_len, str.u, str_len TSRMLS_CC);
+               str.s = str_utf8;
+               str_len = str_utf8_len;
+               hint_charset = "utf-8";
+       }
+
+       replaced = php_unescape_html_entities(str.s, str_len, &len, 1, quote_style, hint_charset TSRMLS_CC);
+
+       if (type == IS_UNICODE) {
+               RETVAL_U_STRINGL(UG(utf8_conv), replaced, len, ZSTR_AUTOFREE);
+               efree(str_utf8);
+       } else {
+               RETVAL_STRINGL(replaced, len, 0);
+       }
 }
 /* }}} */
 
 
-/* {{{ proto string htmlentities(string string [, int quote_style][, string charset])
+/* {{{ proto string htmlentities(string string [, int quote_style][, string charset]) U
    Convert all applicable characters to HTML entities */
 PHP_FUNCTION(htmlentities)
 {