Hi,

your patch looks like the right solution, it's simple and I just tested
it (git trunk + patches) successfully with a broken service...

I have previously also successfully tested the workaround with the
callback function, but the new option (obviously) has my preference...

    Mischa

On Mon, Feb 06, 2012 at 10:36:18PM +0100, Daniel Stenberg wrote:
> On Sun, 5 Feb 2012, Alessandro Ghedini wrote:
>
> >That is the same solution I have thought in response to Debian bug
> >#658276. I was not sure about the option naming, but
> >CURLOPT_SSL_NO_EMPTY_FRAGMENTS (and maybe
> >--ssl-no-empty-fragments) seems the best to me.
>
> Hi,
>
> I'm proposing changes as the attached patches show. I've not
> actually tried this against a live troublesome server so if you
> have/know any, please have a go and tell us how this behaves!
>
> Of course other opinions and feedback would be welcome as well.
>
> --
>
> / daniel.haxx.se

-- 
Nikhef                      Room  H155
Science Park 105            Tel.  +31-20-592 5102
1098 XG Amsterdam           Fax   +31-20-592 5155
The Netherlands             Email msalle_at_nikhef.nl
  __ .. ... _._. .... ._  ... ._ ._.. ._.. .._..

-------------------------------------------------------------------
List admin: http://cool.haxx.se/list/listinfo/curl-library
Etiquette: http://curl.haxx.se/mail/etiquette.html