[zeromq-dev] Assertion failure in pull.cpp

[Jon Dyte]

Pieter Hintjens wrote:
> On Sat, Aug 21, 2010 at 8:47 AM, Jon Dyte <jon at totient.co.uk> wrote:
>
>   
>>> *GRIN* Honestly - I keep running into the same thing every time I start
>>> a new up/downstream pair. Caveat: Because I was part of the discussion
>>> on the naming, I actively think "don't get this the wrong way around"
>>> and proceed to do exactly that :)
>>>       
>> I made exactly the same error doing that test program for Oliver!!!
>>     
>
> The direction of the pipeline is now pretty clear, using PUSH/PULL.
> That's fine.  But we all make mistakes when we write code...
>
>   
too true.
> The main lesson here seems to be that 0MQ will have to do pattern
> verification so that it catches errors like this.
>
>   
yes it was suggested earlier and martin has confirmed it's doable . the 
zmq_init_t  is the place for it, as far I can see.
I'd have to check if the inproc stuff makes use of that though.

> I.e. in the protocol, the connecting socket has to tell the other peer
> what type it is, so the endpoint can reject illegal connections and
> the caller can be told about it.
>
>   
do you we have a definitive list of which sockets are compatible peers 
with each other?

I am just recalling earlier discussions on the esoteric device usage and 
I'm stuff minrk/brian gave
some strange pairings and then with clever use of the msg_more et al 
made them participate as though they
were another type?

Jon