We can enter a bug stating that ZMQ 2.1 will crash when connected with 2.2/3.0 sockets. Couldn't we consider forward incompatibility as a bug ?
Otherwise, we could wait and see if the change really break 2.1. I don't think any 2.1 sockets would break on new flags, and if we can introduce the new control command without breaking the old sockets pattern, that would be even better. Fabien -----Message d'origine----- De : [email protected] [mailto:[email protected]] De la part de Dirkjan Ochtman Envoyé : 16 mai 2011 11:37 À : ZeroMQ development list Objet : Re: [zeromq-dev] ZMTP/1.1 - proposal On Mon, May 16, 2011 at 17:22, Pieter Hintjens <[email protected]> wrote: >> Strictly taken it seems like adding a version header check would go >> against the bugfix-only rule for 2.1 inclusion (changing the wire >> protocol and semantics). > > Yes, we don't want new protocol support or semantics in 2.1, it's > dangerous and not needed. But I'd like to make 2.1 apps resistant > (i.e. not crash) when hit with modified ZMTP protocols. I understand, I'm just stating that this would go against the release policy outlined on the wiki, so that either (a) the release policy maybe should be changed, (b) the non-compliant change should be pointed out exhaustively in the changelog, (c) the new version header should be delayed to 2.3/3.0, with 2.2 the first version checking the flag. Cheers, Dirkjan _______________________________________________ zeromq-dev mailing list [email protected] http://lists.zeromq.org/mailman/listinfo/zeromq-dev _______________________________________________ zeromq-dev mailing list [email protected] http://lists.zeromq.org/mailman/listinfo/zeromq-dev
