Pink Floyd wrote: > (I had posted this in the Zend php-general mailing > list > but it did not show up in php.net, so here it is > again) > Greetings, > I am not sure whether this is a bug or a 'feature' > (new to PHP). I have a cookie that is set by our > password server (NON-PHP) for our domain, whose value > is base64 encoded when set. > Now, when I try to retrieve the Cookie from my > PHP/Apache Web server the $_COOKIE['token'] value is > messed up. What I have noticed is the '+' character is > being replaced by ' '. > I have the magic_quotes_gpc set to 'off' and > variables_order set to 'gpcs' in my php.ini. > Currently, I have hacked around by painfully parsing > the $_SERVER['HTTP_COOKIE'] variable and extracting > the 'token' cookie, where its value appears to be > intact as set by our password server. > > I would like to avoid this hack if possible, is there > a way to retrieve the 'raw' cookie value without php > 'cleansing' it? There is a raw_headers variable, but it would be even more to parse than what you've got... It occurs to me, though, that PHP is almost for sure using the same under-lying code as http://php.net/urldecode and you could just take the cookie value you get and pass it through http://php.net/urlencode to "undo" the decode that PHP did. They're inverse functions, so it should be safe to do that for any/all munging that gets done, not just "+" <-> " " but also any %XX that happens to occur in your base64-encoded value. Still a "hack" but far less painful than parsing the raw Cookie stuff yourself, and you're less likely to screw up and miss some kind of "gotcha" with % or + or space or... If there *IS* a way to turn off PHP URL decoding, it's gonna be in php.ini and it's gonna be documented right in there. PHP doesn't have a bunch of undocumented "features" that you rely on to work. :-) PS It's a feature: [Netscape Cookie spec excerpt] "This string is a sequence of characters excluding semi-colon, comma and white space. If there is a need to place such data in the name or value, some encoding method such as URL style %XX encoding is recommended, though no encoding is defined or required." Since data often *DOES* have white space semi-colon or comma (yours doesn't, but that's the exception) then you have to do SOME kind of encoding on it. It's a lot easier to just always urlencode/urldecode it than to document and code every chunk of data that might or might not have those characters, and then only encode/decode the ones that need it. base64 encoding is also a fine choice, of course, but it's also easier to handle untrusted data of Cookies, POST, and GET all the same, and GET has to be urlencoded, so urlencoding is the choice made. So you basically want to do: $token = base64_decode(urlencode($_COOKIES['token'])); and Bob's your uncle. -- Like Music? http://l-i-e.com/artists.htm -- PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php
