This creates the timeline object
earlier in the Room.getTimeline() method.
This results to that the
Timeline object already starts
to listen on the event stream
while the getTimeline() method
requests users from the database
and tries to decrypt room
events. I assume that this
causes the problem that
on timeline creation new
events get lost because they
come in, while getTimeline() is
not yet completed but the
onEvent stream is not yet
connected at the same time.
This fixes the response
timeout for the initial
sync and sets it to 2
minutes instead of 10
seconds. This should increase
the speed for the initial
sync especially for large
accounts. This change
also adds some
documentation in the
code about what
timeout does what.
This refactors three methods
regarding powerlevels. It
makes them a little bit more
readable and removes
unnecessary checks. it
also changes the canSendDefaultMessage
method in a way that it also
checks the events map in the
powerlevels so that it returns
true if message events are
overwritten there. It also
improves the dart docs and
explains in more detail what it
does.
dynamic.copy returned a type
error so I reverted the previous
change of the copyMap
method to an extension. Also
I found out that we have used
two copyMap methods in the
SDK which did exactly the same.
I deleted the old one and
changed the tests.
Unfortunately we need to copy
all maps. I took the opportunity
and abstracted the copy map
method to utils. I kept the
copyMap method as a
reference in HiveCollections
database to not require a
huge refactoring there as we
are going to ditch it anyway.
BREAKING CHANGE: This replaces the old dehydrated devices
implementation, since there is no way to query what is supported easily
and supporting both would be complicated.
fixes https://github.com/famedly/matrix-dart-sdk/issues/1579
There were several issues here. Key uploads in general failed, because
the await caused a Future<dynamic> to be returned, which failed type
checking.
But also we never unset our future, which was used for the online key
backup uploads, which meant we would very quickly stop uploading any
keys at all.
This changes the behavior
of the markdown method to
only convert linebreaks inside
of p blocks. I found no better
solution yet for the problem
as otherwise also lists
will have linebreaks between
the list items. Unfortunately
the default linebreak syntax
seems not to fulfill our needs.
BREAKING CHANGE: This changes the runInRoot method to not return a
future. As a user, if you need the result of an async computation passed
to runInRoot, please await it directly. Also the KeyVerification start
and a few call methods now return a future.
- store left rooms in archive during sync (as well as they are removed
on join already)
- refactor room archive code
- fix typo
Internal reference: SMC-385
Signed-off-by: The one with the braid <info@braid.business>
Removes two test cases in the markdown test which do not work anymore.
Reason for this is that just parsing a word inside of $$ word $$
katex is not valid anyway because katex is only made for mathematical
things. So the output is undefined behavior anyway.
At least in our CI this throws a null assertion error and since we
explicitly expect this to sometimes be null in our code, we should store
it as such.
If you create DM room, but the other party doesn't join it, they might
be unable to create a new DM using startDirectChat. Since creating a new
DM is pretty much the same as joining the invite, we try to join the
pending DM invite first and fall back to creating a new room if that
fails.
This does make the reset take longer on big accounts, but otherwise
users might sign out before the keys are uploaded, which makes the reset
more destructive than necessary. In the common case of not having any
keys it shouldn't make a difference.
It wasn't quite obvious that invites were always sorted at the top, if
you just looked at the sort function. This makes it more explicit and
also makes invites always sort before favourited rooms.
fix: Also update last event on redaction in store
Closes famedly/company/product-management#990
See merge request famedly/company/frontend/famedlysdk!1298
If we join first, it is possible that our member event is not the invite
event anymore. As such we should fetch that state first, before joining.
But also there is little reason not to mark the room as a DM
immediately. That prevents the room from temporarily becoming a group
room, that might be visible in the UI temporarily.
fixes https://gitlab.com/famedly/company/product-management/-/issues/1006
Otherwise it would be possible, that we haven't loaded account data, so
we return that cross-signing is disabled and then we load it and return
a different result. Might fix the sentry issue for that.
fix: while singleShotStaleCallChecker code seems to be correct we are seeings some random disconnects in group calls, this makes sure we have all the to-device messages we are supposed to have before checking for stale calls
Without this, when new rules are added with new conditions, we would
always match on them. This would mean all messages now notify. It is
better to skip them instead.
currently in the app we start the group call (but do not enter it), then we use the localStreams provided by it to show the setup page (enable/disable media devices). This causes some issues because the user hasn't joined the group call so the member state events don't update and the group call is killed if they just sit on the setup call page. Also creating a group call, sends notifications in our apps but no one has entered it. With this MR, apps can get their own user media, then edit that stream on setup page add it to GroupCall.enter(stream). This way, a group call doesn't have to be started and we get to do the setup page as well.
You can either set them to private on the client or overwrite the
behaviour for that on each readMarker call. This also handles incoming
thread receipts and tries its best to figure out what is the latest
receipt. This however breaks the old m.receipt account data event.
This makes sure that the deleteFromDisk() method is
called AFTER the clear while
the execution order was
random before which could
lead to the problem that
Hive tries to clear a box after
it got deleted.
In Synapse 1.62.0 the default cache duration of initial sync responses
was changed to 2 minutes. That means our archive calls might return
stale data. Since different sync timeouts bust through the cache, we
make use of that to be able to load the new archive immediately.
fixes https://gitlab.com/famedly/company/frontend/famedly-web/-/issues/634
Krille
td
Krille
Nicolas Werner
Karthikeyan S
Nicolas Werner
Mohammad Reza Moradi
krille-chan
td
The one with the braid
Clemens-Toegel
Ray
mErrenst
The one with the braid
Malin Errenst
Reza
Kristian Grønås
Dmitriy Bragin
Philipp Grieshofer
m_kushal
cloudwebrtc