faeez/mxmbsocket
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Repo. Project Releases Wiki Activity

[WebSocket] UTF-8 checking

Browse Source 
mb_check_encoding first, as it's much faster, fail fast
Only do the other check if message is under 10k characters
Large payloads crawl otherwise
This commit is contained in:
Chris Boden 2012-07-12 13:29:12 -04:00
parent bdca66a27d
commit c15c324424
1 changed files with 10 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
src/Ratchet/WebSocket/Version/RFC6455.php
View File

@ -266,10 +266,6 @@ class RFC6455 implements VersionInterface {
} }
return false; return false;
if (empty($val)) {
return false;
}
} }
/** /**
@ -296,12 +292,19 @@ class RFC6455 implements VersionInterface {
* @return bool * @return bool
*/ */
function isUtf8($str) { function isUtf8($str) {
if (isset($str[100000])) { if (false === mb_check_encoding($str, 'UTF-8')) {
return false;
}
$len = strlen($str);
// The secondary method of checking is painfully slow
// If the message is more than 10kb, skip UTF-8 checks
if ($len > 10000) {
return true; return true;
} }
$state = static::UTF8_ACCEPT; $state = static::UTF8_ACCEPT;
$len = strlen($str);
for ($i = 0; $i < $len; $i++) { for ($i = 0; $i < $len; $i++) {
$state = static::$dfa[256 + ($state << 4) + static::$dfa[ord($str[$i])]]; $state = static::$dfa[256 + ($state << 4) + static::$dfa[ord($str[$i])]];
@ -311,6 +314,6 @@ class RFC6455 implements VersionInterface {
} }
} }
return mb_check_encoding($str, 'UTF-8'); return true;
} }
} }