php/archived-php-src
1
0
Fork 0
mirror of https://github.com/php/php-src.git synced 2026-04-24 00:18:23 +02:00
Code Issues Packages Projects Releases Wiki Activity

- Take account of surrogate pairs.

Browse Source
This commit is contained in:
Moriyoshi Koizumi
2009-12-07 15:41:43 +00:00
parent 9f3ec3c5de
commit b1ec98147c
2 changed files with 11 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ext/standard/html.c
+2
View File
@@ -571,6 +571,8 @@ 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]);
ext/standard/tests/strings/bug49785.phpt
+9 -1
View 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"