]> granicus.if.org Git - php/commitdiff
- Take account of surrogate pairs.
authorMoriyoshi Koizumi <moriyoshi@php.net>
Mon, 7 Dec 2009 15:41:43 +0000 (15:41 +0000)
committerMoriyoshi Koizumi <moriyoshi@php.net>
Mon, 7 Dec 2009 15:41:43 +0000 (15:41 +0000)
ext/standard/html.c
ext/standard/tests/strings/bug49785.phpt

index bd24659d764f79e028078c8b09ccd27796482ba6..9fc8c3a01907fe883687297730b30bee8fe9b19d 100644 (file)
@@ -565,6 +565,8 @@ inline static unsigned int get_next_char(enum entity_charset charset,
                                        this_char = ((c & 0x0f) << 12) | ((str[pos + 1] & 0x3f) << 6) | (str[pos + 2] & 0x3f);
                                        if (this_char < 0x800) {
                                                MB_FAILURE(pos);
+                                       } else if (this_char >= 0xd800 && this_char <= 0xdfff) {
+                                               MB_FAILURE(pos);
                                        }
                                        MB_WRITE((unsigned char)c);
                                        MB_WRITE((unsigned char)str[pos + 1]);
index 7b93c8c7168fa65d8030ddc4236293183bcab8b2..fa42dacd6d6fcf1f1fac86cde0a7c3df297756a4 100644 (file)
@@ -42,10 +42,14 @@ var_dump(_bin2hex(htmlentities("\xe0\x80\x80\xe0\xa0\x80", ENT_QUOTES | ENT_IGNO
 var_dump(_bin2hex(htmlentities("\xf0\x80\x80\x80\xf0\x90\x80\x80", ENT_QUOTES | ENT_IGNORE, "UTF-8")));
 
 echo "--\n";
-// UTF-8: alternative (invalid) UTF-8 sequence
+// UTF-8: alternative (invalid) UTF-8 sequence / surrogate pairs
 var_dump(_bin2hex(htmlspecialchars("\xc0\xa6", ENT_QUOTES, 'UTF-8')));
 var_dump(_bin2hex(htmlspecialchars("\xe0\x80\xa6", ENT_QUOTES, 'UTF-8')));
 var_dump(_bin2hex(htmlspecialchars("\xf0\x80\x80\xa6", ENT_QUOTES, 'UTF-8')));
+var_dump(_bin2hex(htmlspecialchars("\xec\xbf\xbf", ENT_QUOTES, 'UTF-8')));
+var_dump(_bin2hex(htmlspecialchars("\xed\xa0\x80", ENT_QUOTES, 'UTF-8')));
+var_dump(_bin2hex(htmlspecialchars("\xed\xbf\xbf", ENT_QUOTES, 'UTF-8')));
+var_dump(_bin2hex(htmlspecialchars("\xee\x80\x80", ENT_QUOTES, 'UTF-8')));
 
 // Shift_JIS: non-lead byte >= 0x80
 var_dump(_bin2hex(htmlspecialchars("\x80", ENT_QUOTES, 'Shift_JIS')));
@@ -168,6 +172,10 @@ string(8) "f0908080"
 string(0) ""
 string(0) ""
 string(0) ""
+string(6) "ecbfbf"
+string(0) ""
+string(0) ""
+string(6) "ee8080"
 string(2) "80"
 string(2) "a0"
 string(2) "a1"