On Fri, 22 Mar 2002, Phillip Stevens wrote:

> I was just using curl and the byte range option (-r). I only wish to
> download the first 20KB of a document, however not all webservers are
> acknowledging this request, and are sending the entire file. Is this due to
> the lack of HTTP 1.1? or because they simply refuse to send only part of a
> document?

It could be either of these reasons. It really is up to the server. I don't
know why some servers are setup to not support this.

> How can I get curl to download the first 20KB and then simply abort? I am
> using an automated process and don't want to get stuck on a 1MB+ file.

When getting simple ranges (with only one start offset and one stop offset)
that start with byte 1, then we could indeed make this possible.

It currently doesn't do this though.

As I hope is evident, I get so many mails with so many good suggestions and
ideas that I am incapable of making even half of them come true. I need your
help.

-- 
    Daniel Stenberg -- curl groks URLs -- http://curl.haxx.se/