>>>> I decided to use the InternetCrackURL and
>>>> InternetCombineURL functions in the WinInet
>>>> library of Windows (even though I didn't
>>>> enjoy the idea of using yet another
>>>> component for the task).

>>> I can understand that. Also, using WinInet
>>> makes your application break the portability
>>> that libcurl otherwise can help you maintain.

>> there is a library available that specializes
>> in this kind of stuff. The home page is at:
>> http://www.nongnu.org/uri/

> Are you suggesting we should offer a subset
> of these in libcurl? If so, which?
> And possibly also, why?

I guess the point I was trying to make is that
even though calling WinAPI functions definitely
would break portability, it is certainly possible
to use equivalent library calls (to liburi) on *NIX
platforms.

The reason this thread sparked my interest is because
I have been playing around with something that I think
is similar to Mr.Kayserilioglu's project.

The program feeds libcurl's output into an HTML parser
that canonicalizes all of the HREF= and/or SRC=
tags based on the value of CURLOPT_EFFECTIVE_URL.

Since LIB-URI sits on the other side of the parser
from LIB-CURL, I really don't see where it would be
an advantage to have libcurl provide these functions.

Another caveat is the fact that a <BASE HREF> tag
should take precedence over the effective url.

So determining the actual base url and merging the
relatives is a job for the parser, not the downloader.

Unless someone has a good reason why curl should
handle this?

 - Jeff Pohlmeyer
 

__________________________________________________
Do you Yahoo!?
Faith Hill - Exclusive Performances, Videos & More
http://faith.yahoo.com

-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
Received on 2002-10-09