XSF Discussion - 2022-10-02

What about this https://xmpp.org/extensions/xep-0421.html

Does anyone implement this, can we advance this?

lovetox: yes, Prosody does

MattJ, one minor nit: please consider using 'true' instead of '1' in examples for boolean attribute values. It makes the examples more readable

I'd wager that leads to developers only supporting 'true' in implementations 🙂

Something something examples aren't normative, same as schemas :x

MattJ, the same could be said with '1'

Sure, but not of mixing it up between examples

https://xmpp.org/extensions/xep-0463.html#errors I was re-reading this spec and I'm thinking this should indicate that the error is about affiliations

or it @type='auth' enough? :/

hmm, or maybe not. It's not like it can happen at any time, can it? It'll happen before presence broadcast at join and.. that's it?

(And there it should be pretty obvious that it's about affiliations?)

https://xmpp.org/extensions/xep-0463.html#self « There can be no empty diff but there can be empty full reponses (no affiliations). » What did I mean here again..

> I'd wager that leads to developers only supporting 'true' in implementations 🙂 Bit of JavaScript that parses the examples and renders booleans differently each time? :)

file:///home/ppjet/Src/xeps/build/xep-0463.html#errors should an error here prevent the join even?

oops, link

https://xmpp.org/extensions/xep-0463.html#errors

Wasn't there something about errors from the MUC also kicking users out?

Anybody implementing this spec btw? Or planning to?

From 0045: > If an occupant wants to send an IQ stanza to another user in a semi-anonymous room, the sender can direct the stanza to the recipient's occupant JID and the service SHOULD forward the stanza to the recipient's real JID. What resource should the IQ be sent to? Or is it sent to the bare JID that is joined? ✏

singpolyma, that part of xep45 was written before multi session nicks where a thing. I could imagine that modernxmpp has something about this. the IQ should never be send to the bare JID ✏

flow: well, since probably the most common iq would be for avatar fetch then to bare JID is what I would guess is being done in practise. But I'm not sure

From jdev@ two days ago: MattJ> pep.: I think Prosody stores all [presences of a participant], one is randomly selected as the session leader, and a new leader is selected if it leaves

That is, the iq is directed to the session leader

We were talking about presences but I guess this "leader" is also used for iq

avatar was the main use case for this, a much better solution is to send the iq to translate the avatar request into a pubsub request to the bare jid

no client can depend on getting the correct info if its info that is different per resource

you cant address a specific resource of the user

and if you get something back it would be very random and could change any minute

pep.: Ok, so at least in prosody the iq is sent to one of the resources, you don't know which one

The reason I am asking is because I am implementing bits of binary and was thinking through what would even happen in a MUC use

So I guess with only one resource joined it would work, but otherwise might just randomly fail

While in 0045: « A MUC service MUST include the MUC extensions even if the client did not send an empty <x/> element qualified by the 'http://jabber.org/protocol/muc' namespace on entering the room; naturally, a client MUST ignore such information if it does not understand it (in accordance with RFC 6120). », remnants of GC1? https://xmpp.org/extensions/xep-0045.html#bizrules-presence ✏

Yes

Is there such a thing is no affiliations? Can there be no owner at all? Removing the initial creator?

« A service MUST NOT allow an owner to revoke his or her own owner status if there are no other owners; if an owner attempts to do this, the service MUST return a <conflict/> error to the owner. However, a service SHOULD allow an owner to revoke his or her own owner status if there are other owners. »

So there will always be at least one owner

Yes otherwise that'd leave an orphaned room

Are the rfcxxxx entities usable in xeps autogenerated or do I have to add new RFCs manually?

see xep.ent

thilo.molitor ↑

it's a file in the top level directory of the xeps repo

flow: yes, but are they autogenerated or do I have to submit a pull request to add several RFCs?

thilo.molitor, I would expect that you add them in the same PR where the XEP is added that uses them

Okay, thanks

thilo.molitor, so no, xep.ent is not autogenerated