On Tue, 22 Feb 2011, binziwang_at_gmail.com wrote:

> Our application has been relying on libcurl for URL redirection. We would
> need to rewrite a large portion of the code to handle URL redirection
> manually.

Why? It doesn't make any sense to me. Write your own function called kurl()
that does the job and that returns the error code from the last libcurl call.
You wouldn't have to change any logic in your application.

"kurl" would then just be a loop doing multiple libcurl calls, and as I
already showed you libcurl already does all the grunt work and it provides you
with all the info you asked for.

> Daniel, do you see a problem with extending libcurl in either of the
> approach I suggested?

Yes. Because it's pointless when we already offer several ways to extract the
information you want.

The impression I get is that you reject suggestions presented here because
they don't fit the way you already set out to have this working before you
asked.

> I don't really mean: "I don't want to change my code, so I will just mess
> around with libcurl". It is just that the extension to libcurl seems pretty
> simple. So I suggested it.

Sure, it's simple to add since all the info is already there, but that's not a
very good reason to add yet another feature if you ask me...

You need to produce better reasons and arguments for your case. Or perhaps
find others who also think its a good idea so that I'd find myself in a
minority...

-- 
  / daniel.haxx.se
-------------------------------------------------------------------
List admin: http://cool.haxx.se/list/listinfo/curl-library
Etiquette:  http://curl.haxx.se/mail/etiquette.html