We've always had 'age' in events but it's never really been an
accurate representation of the event's age because we never did
anything with it. This transforms it into a local clock timestamp
when the event arives and when it comes out of the sync store, and
changes getLocalAge() to use it.
react-sdk doesn't appear to use getLocalAge() but any 3rd party apps
that do may notice a slight change in bahaviour.
I think this was attempting to remove the events from _pendingEvents
but a) it wasn't and b) it probably shouldn't be because the retry
itself will re-add them when the decryption attempt starts and remove
them if it succeeds.
Also fix what was presumably a c+p fail.
The bulk of this is just export/import changes, though there's a couple pieces to highlight:
* We no longer use default exports. This is because it's discouraged by the JS community, though not in any official capacity.
* We now use `polyfillSuper` for some prototype inheritance because the tests, and sometimes webpack, break on "cannot call EncryptionAlgorithm without 'new'". It's very much a workaround, and definitely not needed when we use real classes.
There is some import shuffling to help keep the imports clean - this was done by my IDE.
Fixes https://github.com/vector-im/riot-web/issues/10083
Fixes https://github.com/vector-im/riot-web/issues/9197
Fixes https://github.com/vector-im/riot-web/issues/8629
The issue is partially fixed by https://github.com/matrix-org/matrix-react-sdk/pull/3123 in that users would no longer see "Incoming request", but would launch their client to a bunch of "key verification cancelled" dialogs. To work around this, we just don't handle key verification requests which we know are cancelled.
The changes are a bit awkward (flagging the event as cancelled instead of filtering it) because:
* We probably don't want to prevent events getting sent over the EventEmitter because applications may still rely on them.
* The cypto side only has visibility of 1 event at a time, so it needs to have some kind of flag to rely on.
An attempt has been made to generalize the new event flag for possible future cases.
because the RedactionDimensions was trying to redact an event
that was already redacted after it's remote echo had come in
but it's redaction hadn't synced yet.
Germain Souquet