On Thu, 9 Jan 2003 12:37:13 +0100 (MET)
Daniel Stenberg <daniel_at_haxx.se> wrote:

> On Thu, 9 Jan 2003, Philippe Raoult wrote:
> I don't think it is worth doing. The function may very well produce binary
> results, so any user that assumes a zero byte as the end of data is doing bad
> assumptions.
>
> The code that currently use this function use the returned length.
>
> (As we currently don't export this function to the world.)

oh really ? :)
i use it anyway in my code but i agree it's not really worth it.
 
>
> I think we should add documentation for this, and not attempt to fix this
> within libcurl. I'll happily apply any patches to the docs that clarifies
> this issue.

i'll try to do this.
BTW i've downloaded the pre4 archive and it seems to miss share.[ch] in the lib/
directory.

Regards,
Philippe

-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
Received on 2003-01-09