MUC MAM

[MarcoPolo-PasTonMolo]

MUC MAM

See #660

How to test the functionality

• step 1: Have mam enabled
• step 2: Join random muc scroll up
• step 3: rejoining muc should fetch messages sent since last online with profanity

I ran valgrind when using my new feature

no

---

TODO

• Look up older history
• Look up history when creating muc window
• Look up newer history
• Fetch mam on incoming message (does a new muc even get created on incoming message ?)
• Fetch mam on reconnects
• Scroll history even without mam
• Why loading message sometimes lingers? (when at the end of muc mam maybe?)
• Why is a chatwin opened for each member of muc?
• Last messages appear doubled (maybe this only happens for chatwins?)
• wrong timestamp for messages through mam on new join of muc
• mentions/triggers show printed date not original date
• respect mam setting
• get all rsm results even from pagination
• Autocompletion has to be correct
• how to handle read indicator with mam
• valgrind
• make loading mam for muc faster if possible
• Don't request history from muc

[jubalh]

Fetch mam on incoming message (does a new muc even get created on incoming message ?)

I'm not sure I understand correctly. But: I think MAM only needs to be fetched for MUCs that are currently joined? Windows are only open for currently joined MUCs.

[MarcoPolo-PasTonMolo]

I'm not sure I understand correctly. But: I think MAM only needs to be fetched for MUCs that are currently joined? Windows are only open for currently joined MUCs.

I remember that chatwins are created/opened when the other party sends a message. I was thinking that maybe that was also happening for MUCs and so I would have to handle MAM in the case that the user is not the one that opens a mucwin by /joining it

[ike08]

@MarcoPolo-PasTonMolo, do you want help with this?

[MarcoPolo-PasTonMolo]

@MarcoPolo-PasTonMolo, do you want help with this?

Yes please. I don't have time to work on this. My code can be pretty garbage at times especially in iq.c so if you have any questions ask me. I think I would respond quicker in xmpp so Ill tag you there.

If I remember correctly MAM for MUCs is pretty much done. I had started implementing MAM for when a user gets reconnected but hit a few road blocks like how do I prevent new messages, that arrive while fetching mam after a reconnect, from getting jumbled up with the old ones from MAM. Or how do I load the actual MAM messages while also giving some feedback to the user so he doesn't think that the ui is stuck (like use pagination and fetch 10 messages at a time etc)

I mapped pagedown to trigger a disconnect so that testing is easier with a /reconnect interval of 5 seconds.

Here's what I had written down as TODO

== Reconnect mam ==

• Fetch more messages if not all fetched
• Make it work
• refactor code to be cleaner and more understandable

= MUC MAM =

• Look up older history
• Look up history when creating muc window
• Look up newer history
• lost connection while loading messages
• Fetch mam on reconnects
• Why loading message sometimes lingers and doesn't disappear? (when at the end of muc mam maybe?)
• how to handle read indicator with mam
• Last messages appear doubled (maybe this only happens for chatwins?) when i get a new message from the contact while window is not open
• Autocompletion has to be correct
• make loading faster (or somehow show progress)
• Fetch mam on incoming message (does a new muc even get created on incoming message ?)
• Scroll history even without mam
• Why is a chatwin opened for each member of muc?
• wrong timestamp for messages through mam on new join of muc
• mentions/triggers show printed date not original date
• respect mam setting
  • not showing history when mam is off,
  • and after first join and fetch of messages
• get all rsm results even from pagination
• valgrind
• see if all added code is neccessary
• split first commit
• does it load multiple (does pagination work)? or did we distroy it with return 0 in rsm handler?
• no nick in front of messages in muc mam
• auto_chars

[jubalh]

Thanks for your work and the summary @MarcoPolo-PasTonMolo !!

[jubalh]

@ike08 @MarcoPolo-PasTonMolo slowly I have more free time to work on profanity again :) I was curious whether the same is true for one of you guys and whether you are still interested in implementing this feature. Cheers!

[MarcoPolo-PasTonMolo]

Not much time sorry

[jubalh]

@MarcoPolo-PasTonMolo and @ike08 are all your changes pushed to this branch?

[ike08]

Yes; I have not done anything on this.