Posted by Andy Hassall on 12/31/05 00:49

On 27 Dec 2005 07:27:06 -0800, "ed_spain" <ed.eichman@gmail.com> wrote:

>Before I write the script myself, I'm looking for PHP code that checks
>that a link's content type matches it's contents. I need this for my
>"broken link" checker script, which fails to spot the "broken" links
>http://www.awin1.com/show.php?mid=1014&linkid=36515&aid=49007 and
>http://www.awin1.com/awclick.php?mid=1014&linkid=36515&id=49007. Any
>ideas?
>
>BTW, in case the links get fixed before you read the message, the first
>link is a GIF MIME that returns
>
>[document.write('<a
>href="http://www.affiliatewindow.com/?id=.&banner=no"
>target="_blank"><img
>src="http://www.awin1.com/show.php?mid=3&gid=14&aid=477&clickref=nobanner"
>border="0"></a>')]
>
>which displays as
>
>[The image
>"http://www.awin1.com/show.php?mid=1014&linkid=36515&aid=49007"
>cannot be displayed, because it contains errors.] .
>
>The second is a text/html that returns
>
>[<br />
><b>Warning</b>: mysql_result(): Unable to jump to row 0 on MySQL
>result index 30 in <b>/home/sites/site3/web/awclick.php</b> on line
><b>51</b><br />
><br />

There's MIME-type detection using "magic numbers" available through:
http://uk2.php.net/mime_content_type

This does a quick check against the first few bytes of the file to see if it
matches up with various file formats.

It won't catch the second one, because that's still a valid HTML page - it may
have PHP errors on it, but they're in HTML.

--
Andy Hassall :: andy@andyh.co.uk :: http://www.andyh.co.uk
http://www.andyhsoftware.co.uk/space :: disk and FTP usage analysis tool

[Back to original message]