Unicode upgrade for quotemeta(). Yes, I know it's ugly. You try making

it pretty.

diff --git a/ext/standard/string.c b/ext/standard/string.c
index ff12fa2f96df4fb434ad5e42a9e4d4f0e74c2f95..839a9c50ceb92fb727a41e7c6771a3dfba6b5182 100644 (file)
--- a/ext/standard/string.c
+++ b/ext/standard/string.c
@@ -3406,54 +3406,81 @@ PHP_FUNCTION(substr_replace)
 }
 /* }}} */
 
-/* {{{ proto string quotemeta(string str)
+/* {{{ proto string quotemeta(string str) U
    Quotes meta characters */
 PHP_FUNCTION(quotemeta)
 {
-       zval **arg;
-       char *str, *old;
-       char *old_end;
-       char *p, *q;
+       zstr str, old;
+       zstr old_end;
+       int  old_len;
+       zstr p, q;
        char c;
+       UChar cp;
+       zend_uchar type;
 
-       if (ZEND_NUM_ARGS() != 1 || zend_get_parameters_ex(1, &arg) == FAILURE) {
-               WRONG_PARAM_COUNT;
+       if (zend_parse_parameters(ZEND_NUM_ARGS() TSRMLS_DC, "t", &old, &old_len, &type) == FAILURE) {
+               return;
        }
 
-       convert_to_string_ex(arg);
-
-       old = Z_STRVAL_PP(arg);
-       old_end = Z_STRVAL_PP(arg) + Z_STRLEN_PP(arg);
-
-       if (old == old_end) {
+       if (old_len == 0) {
                RETURN_FALSE;
        }
 
-       str = safe_emalloc(2, Z_STRLEN_PP(arg), 1);
 
-       for (p = old, q = str; p != old_end; p++) {
-               c = *p;
-               switch (c) {
-                       case '.':
-                       case '\\':
-                       case '+':
-                       case '*':
-                       case '?':
-                       case '[':
-                       case '^':
-                       case ']':
-                       case '$':
-                       case '(':
-                       case ')':
-                               *q++ = '\\';
-                               /* break is missing _intentionally_ */
-                       default:
-                               *q++ = c;
+       if (type == IS_UNICODE) {
+               old_end.u = old.u + old_len;
+               str.u = safe_emalloc(2, UBYTES(old_len), 1);
+
+               for (p.u = old.u, q.u = str.u; p.u != old_end.u; p.u++) {
+                       cp = *p.u;
+                       switch (cp) {
+                               case '.':
+                               case '\\':
+                               case '+':
+                               case '*':
+                               case '?':
+                               case '[':
+                               case '^':
+                               case ']':
+                               case '$':
+                               case '(':
+                               case ')':
+                                       *q.u++ = '\\';
+                                       /* break is missing _intentionally_ */
+                               default:
+                                       *q.u++ = cp;
+                       }
+               }
+               *q.u = 0;
+               RETURN_UNICODEL(eurealloc(str.u, q.u - str.u + 1), q.u - str.u, 0);
+       } else {
+               old_end.s = old.s + old_len;
+               str.s = safe_emalloc(2, old_len, 1);
+
+               for (p.s = old.s, q.s = str.s; p.s != old_end.s; p.s++) {
+                       c = *p.s;
+                       switch (c) {
+                               case '.':
+                               case '\\':
+                               case '+':
+                               case '*':
+                               case '?':
+                               case '[':
+                               case '^':
+                               case ']':
+                               case '$':
+                               case '(':
+                               case ')':
+                                       *q.s++ = '\\';
+                                       /* break is missing _intentionally_ */
+                               default:
+                                       *q.s++ = c;
+                       }
                }
+               *q.s = 0;
+               RETURN_STRINGL(erealloc(str.s, q.s - str.s + 1), q.s - str.s, 0);
        }
-       *q = 0;
 
-       RETURN_STRINGL(erealloc(str, q - str + 1), q - str, 0);
 }
 /* }}} */
 

diff --git a/unicode-progress.txt b/unicode-progress.txt
index 3edcd28cf78e3aba09ad5973e6ea53e3d9ced2cd..44e9cfcafe8cc8813ba24fb7c4219ed06cab0b7e 100644 (file)
--- a/unicode-progress.txt
+++ b/unicode-progress.txt
@@ -23,9 +23,6 @@ ext/standard
     parse_str()
         Params API. How do we deal with encoding of the data?
 
-    quotemeta()
-        Params API, IS_UNICODE upgrade
-
     sscanf()
         Params API. Rest - no idea yet.
 
@@ -147,6 +144,7 @@ ext/standard
     nl2br()
     ord()
     pathinfo()
+    quotemeta()
     range()
     similar_text()
     str_pad()