You are right. I apologize for getting it wrong. My question should ask if
there is another way to use the HTTP Range header other than through the
CURLOPT_HTTPHEADER option.

I'm trying to grab large files from a web server in chunks so if the
connection is broken or of the transfer is canceled, I can resume the
transfer. I won't get into the reasons for grabbing the file in chunks
initially, but this is something that I need to be able to do.

The only way I can figure out how to do it is by setting the Range header
using the CURLOPT_HTTPHEADER option, then calling curl_easy_perform(),
rinse, repeat until the entire file is retrieved.

I don't have much experience with libcurl and I'm hopping there is an easier
way to do this. Am I missing something?

On 3/31/07, Jeff Pohlmeyer <yetanothergeek_at_gmail.com> wrote:
>
> On 3/31/07, Scott Hansen <hansensv_at_gmail.com> wrote:
> > Is there another way to use an HTTP Accept-range header other than using
> the
> > CURLOPT_HTTPHEADER option?
>
> If I understand the RFC correctly "Accept-ranges" is intended
> to be a response header, not a request header?
>
> Or do I miss the question completely?
>
> - Jeff
>
Received on 2007-03-31