moriyoshi               Thu Oct  2 02:57:22 2003 EDT

  Added files:                 
    /php-src/ext/standard/tests/strings bug25707.phpt 

  Modified files:              
    /php-src/ext/standard       html.c 
  Log:
  Fixed bug #25707 (html_entity_decode over-decodes <)
  
  
Index: php-src/ext/standard/html.c
diff -u php-src/ext/standard/html.c:1.86 php-src/ext/standard/html.c:1.87
--- php-src/ext/standard/html.c:1.86    Tue Jun 10 16:03:37 2003
+++ php-src/ext/standard/html.c Thu Oct  2 02:57:21 2003
@@ -18,7 +18,7 @@
    +----------------------------------------------------------------------+
 */
 
-/* $Id: html.c,v 1.86 2003/06/10 20:03:37 imajes Exp $ */
+/* $Id: html.c,v 1.87 2003/10/02 06:57:21 moriyoshi Exp $ */
 
 /*
  * HTML entity resources:
@@ -417,7 +417,6 @@
        int entitylen;
        int flags;
 } basic_entities[] = {
-       { '&',  "&",        5,      0 },
        { '"',  """,       6,      ENT_HTML_QUOTE_DOUBLE },
        { '\'', "'",       6,      ENT_HTML_QUOTE_SINGLE },
        { '\'', "'",        5,      ENT_HTML_QUOTE_SINGLE },
@@ -917,54 +916,69 @@
                ret = replaced;
        }
 
-       /* replace numeric entities */
+       /* replace numeric entities & "&" */
        lim = ret + retlen;
-       for (p = ret, q = ret; p < lim; p++) {
+       for (p = ret, q = ret; p < lim;) {
                int code;
 
-               if (p < lim - 1 && p[0] == '&' && p[1] == '#') {
-                       code = strtol(p + 2, &next, 10);
-                       if (next != NULL && *next == ';') {
-                               switch (charset) {
-                                       case cs_utf_8:
-                                               q += php_utf32_utf8(q, code);
-                                               break;
+               if (p[0] == '&') {
+                       if (p + 2 < lim) {
+                               if (p[1] == '#') {
+                                       code = strtol(p + 2, &next, 10);
+                                       if (next != NULL && *next == ';') {
+                                               switch (charset) {
+                                                       case cs_utf_8:
+                                                               q += php_utf32_utf8(q, 
code);
+                                                               break;
 
-                                       case cs_8859_1:
-                                       case cs_8859_5:
-                                       case cs_8859_15:
-                                               if (0xa0 <= code && code <= 0xff) {
-                                                       *(q++) = code;
-                                               }
-                                               break;
+                                                       case cs_8859_1:
+                                                       case cs_8859_5:
+                                                       case cs_8859_15:
+                                                               if (0xa0 <= code && 
code <= 0xff) {
+                                                                       *(q++) = code;
+                                                               }
+                                                               break;
 
-                                       case cs_cp1252:
-                                       case cs_cp1251:
-                                       case cs_cp866:
-                                               if (0x80 <= code && code <= 0xff) {
-                                                       *(q++) = code;
-                                               }
-                                               break;
+                                                       case cs_cp1252:
+                                                       case cs_cp1251:
+                                                       case cs_cp866:
+                                                               if (0x80 <= code && 
code <= 0xff) {
+                                                                       *(q++) = code;
+                                                               }
+                                                               break;
 
-                                       case cs_big5:
-                                       case cs_gb2312:
-                                       case cs_big5hkscs:
-                                       case cs_sjis:
-                                       case cs_eucjp:
-                                               if (code <= 0x7f) {
-                                                       *(q++) = code;
-                                               }
-                                               break;
+                                                       case cs_big5:
+                                                       case cs_gb2312:
+                                                       case cs_big5hkscs:
+                                                       case cs_sjis:
+                                                       case cs_eucjp:
+                                                               if (code <= 0x7f) {
+                                                                       *(q++) = code;
+                                                               }
+                                                               break;
 
-                                       default:
-                                               break;
+                                                       default:
+                                                               break;
+                                               }
+                                               p = next + 1;
+                                       } else {
+                                               *(q++) = *(p++);        
+                                               *(q++) = *(p++);        
+                                       }
+                               } else if (p + 4 < lim &&
+                                                       p[1] == 'a' && p[2] == 'm' 
&&p[3] == 'p' &&
+                                                       p[4] == ';') {
+                                       *(q++) = '&';
+                                       p += 5;
+                               } else {
+                                       *(q++) = *(p++);
+                                       *(q++) = *(p++);
                                }
-                               p = next;
                        } else {
-                               *(q++) = *p;
+                               *(q++) = *(p++);        
                        }
                } else {
-                       *(q++) = *p;
+                       *(q++) = *(p++);        
                }
        }
        *q = '\0';

Index: php-src/ext/standard/tests/strings/bug25707.phpt
+++ php-src/ext/standard/tests/strings/bug25707.phpt
--TEST--
Bug #25707 (html_entity_decode over-decodes &amp;lt;)
--FILE--
<?php
var_dump(html_entity_decode("&amp;lt;", ENT_COMPAT, 'ISO-8859-1'));
var_dump(html_entity_decode("&amp;#38;", ENT_COMPAT, 'ISO-8859-1'));
var_dump(html_entity_decode("&amp;#38;lt;", ENT_COMPAT, 'ISO-8859-1'));
?>
--EXPECT--
string(4) "&lt;"
string(5) "&#38;"
string(8) "&#38;lt;"

-- 
PHP CVS Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php

Reply via email to