I agree with Sandy's suggestion. You might consider a Moo-implementation of the almost-a-standard (but not quite) anchor "ping" attribute.
<a href="http://s3.amazonaws.com/bucket/some-file.pdf"; ping="http:// www.mysite.com/LogRequest.php?some-file.pdf">Some File</a> If a browser does not support the ping attribute, just attach an onclick handler to the tag (or better yet, delegate), and use Request to hit the LogRequest.php. Your server logs will still contain what you want. http://www.w3.org/html/wg/wiki/AddedAttributePing http://www.w3.org/TR/html5-diff/ Eric On Jan 17, 7:57 am, "Steve Onnis" <st...@cfcentral.com.au> wrote: > There is no logging script. The file actually exists > > Say i have a link to a pdf file that iswww.mysite.com/some-file.pdf. This > file exists onwww.mysite.combut i am modifying the output of the html to > changewww.mysite.comtohttp://s3.amazonaws.com/bucket/some-file.pdf. > > When the user clicks on the link, i want it to actually go to > thehttp://s3.amazonaws.com/path but i want to log within the webserver log > files forwww.mysite.comthat some-file.pdf has been requested just as it > would normally be logged. > > So what i want to do is make a request towww.mysite.com/some-file.pdfbut > not actually load the full file. I just want to make a request to it and > then cancel the request. >
