On Tuesday 20 June 2006 03:12, Daniel Stenberg wrote:
> On Thu, 15 Jun 2006, Peter wrote:
> > patch promised in previous letter.
>
> Looks fine, thanks! Please make sure that you use the proper indent level
> (== 2 spaces) and line lengths (< 80 columns) in future submissions - I've
> edited it now. I do however miss the documentation for these new options.
> Can you write up a little patch for docs/libcurl/curl_easy_setopt.3 please?
>
Great! attached.

> Also, when thinking about it I think we should rather make the functions
> accept a 'curl_off_t' rather than a plain long to allow apps to specify
> really huge amounts on systems that support such. Can you think of any
> downsides with that?
>

nope. that's a good thing.

> To prevent the option names from getting overly long then, I suggest we
> call them
>
> CURLOPT_MAX_SEND_SPEED_LARGE and
> CURLOPT_MAX_RECV_SPEED_LARGE
>

I like the SPEED.. If there is only one entry point (and it uses curl_off_t),
then LARGE probably is not needed, and option can be shortened.

I guess _LARGE when you want to support curl_off_t as well as long.
In a few years LARGE will just seem odd (everything will be 64 bit...)
how about CURLOPT_MAX_SEND_SPEED ?

> Oh, and if you provide me with your full real name I'll make sure to give
> you proper credit for this change.
>
see from...

> In case anyone cares, the attached file is the current state of the
> upcoming change.