I am writing code which interprets html links in C.
My query relates to some html I have found in the big bad world.

If you look at the html source code for the page www.hosting-review.com/ it has a link of <a href="../../preferredsites/main.shtml>.

But the current page is "/". so how can you reference "../../" ???

I checked for a "base" element ,but I couldn't see any in the html source.

I perused the relavant rfc's but from what I can gather relative links should always be referenced in relation to the current document unless a base document has been stated ?!?!

Am I missing something?

Thnks in advance

well these paths are relative to the server, so that will simply go up to a higher level than the documentroot and then back into another directory. it's fairly ugly, but valid. ideally your web server permissions would be configure to protect yourself from external attacks trying to reach outside of the documentroot, but if you want to do otherwise then that's the webmasters perogative i guess.

Thanks 4 reply.
Part of my program is a http\1.1 implementation.

When i do..

the server sends back a http 400 error stating an invalid request.

but if I do
.

I get the http 200 status and the document is served ok.

Should I re-chek my code???.

From what you are saying the "ugly link" should work should it not.

thanks again.

yes it should serve, but maybe your source you're looking at is possibly serving from that depth in the first place.. could it be an iframe coming from a different directory or something?

I don't believe it was an Iframe,maybe I typed the url wrong at the command line ???

Mozilla handles the link ok,so I'm missing soemthing.

My program is a spider so I would want to translate the relative path to its full url so that it can be indexed,but the full url is hidden from the user in this case?!!

I don't know...