On 4/17/20 3:10 PM, Matthew Wild wrote:
> On Fri, 17 Apr 2020 at 12:58, Florian Schmaus <[email protected]
> <mailto:[email protected]>> wrote:
> 
>     I think it would also require less code change compare to stuffing
>     <forwarded/> into <iq/>. At least for Smack I think this is likely
>     the case.
> 
> 
> Sorry, but in discussions like this I have to say that Smack always
> surprises me at how hard it is to do relatively simple things :)

There seems to be a misunderstanding here. I never said it was hard with
Smack. I merely said it would require less code change. In fact, I
believe it would require no code change at all. Since Smack had a data
form based MAM API from the very beginning, you could query using a well
defined form field for a list of message ID with all Smack versions
supporting MAM. Whereas Smack does not expect a <forwared/> in the <iq/>
result of a MAM query, because the current mam:2 iq results are not
defined to have one. That is something that one would have to add, but
it is not hard to do so.

But even this wouldn't be the case: I think that using the same scheme
everywhere outweighs the optimization of stuffing the message in iq result.

> I don't feel strongly on this issue one way or the other. I'd be curious
> if anyone else does, or especially for any concrete use cases for
> fetching multiple ids at once.

I think message fastening/reference might yield a use case for this. But
yes, that's just the feeling that there is one, like you get information
about the references, archive IDs and their category of a particular
message, and then decide to fetch the all references of a particular
category by archive ID. Not sure if this accounts as concrete use case.

- Florian

Attachment: signature.asc
Description: OpenPGP digital signature

_______________________________________________
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: [email protected]
_______________________________________________

Reply via email to