On Fri, Aug 17, 2007 at 11:12:30PM +0200, Daniel Stenberg wrote:
> I really don't think this is a good idea and I would like to remove this
> code. Instead the SIZE and MDTM commands will be done and the server will
> respond, but applications that want the info would either get the server
> response headers or use curl_easy_getinfo() etc to extract the information.
>
> Any objections/thoughts?

As long as there's some way to get at the data that doesn't involve redundant
parsing by the application, I don't mind.

> Yes, this is a minor ABI change but I won't bump the soname anyway since I
> fear I'll be chased down the street by people with pitchforks if I do it
> again this soon... :-)

I'm not so sure about this one. If I were an application author who depended
on this feature, I'm be pretty upset if the soname were *not* bumped.
Upgrading libcurl shouldn't cause a properly-written application to start
failing. If you don't want to bump the soname again (and I agree--it's
really annoying), I would suggest adding this to the TODO file and doing
it once a soname change is unavoidable for other reasons.

However, isn't this HTTP-like header behaviour on ftp something that
was just recently enabled due to an unrelated bug fix? My opinion on
fixing it now may change if that's the case.

>>> Dan

-- 
http://www.MoveAnnouncer.com              The web change of address service
          Let webmasters know that your web site has moved