Thanks for reply!
Yes I agree curl has no way out to detect this situation.
Web application should synchronize access to that url page. Any idea about standard mechanism followed by these guys to update url page keeping site alive?

Regards,
Pankaj.

--- On Wed, 3/18/09, Daniel Stenberg <daniel_at_haxx.se> wrote:

> From: Daniel Stenberg <daniel_at_haxx.se>
> Subject: Re: Recieved Partial file
> To: "libcurl development" <curl-library_at_cool.haxx.se>
> Date: Wednesday, March 18, 2009, 12:08 AM
> On Tue, 17 Mar 2009, pankaj takawale
> wrote:
>
> > Intermittently I'm getting partil file from web
> server. cure_easy_perform returns 0, and
> CURLINFO_RESPONSE_CODE is 200. On next hit, after a sec, I
> get complete file. Looks like I'm hiting site while some
> application on web server is in the middle of writing that
> file.
> >
> > How come curl is not detecting it (return code 18)?
> Any way out to trap this situation?
>
> Because the web server hands a complete data and says it
> went fine. How is libcurl supposed to know that you think
> the data is too small?
>
> --
> / daniel.haxx.se
>

      
Received on 2009-03-17