control: tags -1 + patch pending Hi everyone,
> When I'm trying to start a OTR session from pidgin I get "We received a > malformed data message from myaccountname@myserver" in an endless loop on the > pidgin client. I have just added limited support of instance tags into Psi+ OTR plugin: https://github.com/psi-plus/plugins/commit/eb1dfff It should fix your problem. > For some reason, psi+ (before the workaround that removes OTRv3 support) > was sending messages with a sender instance of 0. Yes, this was the root of problem. Thanks for a link. > To be clear, it *could* still be a bug in libotr. But since the only > client we've seen that has this "sender instance 0" issue is psi+, > ... Just for a note: LeechCraft project has the same problem with libotr. I am going to prepare similar patch for them. > What's *supposed* to happen is that when a message is received, and > otrl_message_receving is called by psi+, then inside libotr, the > appropriate ConnContext is found, and if there's no sender instance yet > set on it, the client's create_instag callback is called. If there is > no such callback defined (which, looking at psi+, I don't see one), then > libotr will just create a random instance tag. (This is the > populate_context_instag function in libotr:src/message.c) Unfortunately, I have no time to debug libotr now, but I have checked one thing: if all create_instag and otrl_instag_read are commented in patch above, Psi+ OTR plugin sends malformed data. Something is really wrong, but I am not sure where. Best regards, Boris -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
