[zeromq-dev] Strict aliasing problem

[Pieter Hintjens]

Lourens,

Shoot us a pull request then so people can enjoy your lovely work :-)

-Pieter

On Tue, May 22, 2012 at 3:54 PM, Lourens Naudé
<lourens at methodmissing.com> wrote:
> Hi Guys,
>
> Updated again with feedback thus far :
>
> https://github.com/methodmissing/libzmq/compare/monitor-regressions
>
> - Lourens
>
> On Mon, May 21, 2012 at 9:58 PM, Pieter Hintjens <ph at imatix.com> wrote:
>>
>> Lourens,
>>
>> Looks great. Note that the API for the context got cleaned up and made
>> more consistent in 3.1. Check the zmq man page. We moved to using
>> zmq_ctx_ and zmq_msg_ as prefixes for the context and message classes.
>>
>> So perhaps zmq_ctx_set_monitor () as the method name.
>>
>> -Pieter
>>
>>
>> On Mon, May 21, 2012 at 2:53 PM, Lourens Naudé
>> <lourens at methodmissing.com> wrote:
>> > Hi Guys,
>> >
>> > Still work in progress, but this moves the event callback to the context
>> > level and introduces a formal zmq_monitor() API for registration.
>> >
>> > https://github.com/methodmissing/libzmq/compare/monitor-regressions
>> >
>> > Todo :
>> >
>> > * Doc updates
>> > * Some monitor tests are still failing ( va. args related - looking into
>> > removing that cruft )
>> > * HWM ( send and receive ) events
>> >
>> > Anyways, just wanted to run it by so long for feedback.
>> >
>> > - Lourens
>> >
>> > On Thu, May 17, 2012 at 3:33 PM, Pieter Hintjens <ph at imatix.com> wrote:
>> >>
>> >> On Thu, May 17, 2012 at 9:31 AM, Lourens Naudé
>> >> <lourens at methodmissing.com> wrote:
>> >>
>> >> > I can try to take care of this later today / tomorrow morning.
>> >> > Apologies
>> >> > for
>> >> > the sockopt regressions with the ZMQ_MONITOR patch.
>> >>
>> >> No problem, we have time for this to settle down and it's much more
>> >> fun to patch things into shape than try to make them perfect up front.
>> >>
>> >> > I thought about the callback per context before, however not all
>> >> > sockets
>> >> > would always handle an event in the same way. Thoughts ?
>> >>
>> >> Sure, provide a token in the per-socket call that gets provided back
>> >> to the application handler.
>> >>
>> >> -Pieter
>> >>
>> >> _______________________________________________
>> >> zeromq-dev mailing list
>> >> zeromq-dev at lists.zeromq.org
>> >> http://lists.zeromq.org/mailman/listinfo/zeromq-dev
>> >
>> >
>> >
>> > _______________________________________________
>> > zeromq-dev mailing list
>> > zeromq-dev at lists.zeromq.org
>> > http://lists.zeromq.org/mailman/listinfo/zeromq-dev
>> >
>> _______________________________________________
>> zeromq-dev mailing list
>> zeromq-dev at lists.zeromq.org
>> http://lists.zeromq.org/mailman/listinfo/zeromq-dev
>
>
>
> _______________________________________________
> zeromq-dev mailing list
> zeromq-dev at lists.zeromq.org
> http://lists.zeromq.org/mailman/listinfo/zeromq-dev
>