This requires changes in a lot of places because we were assuming that a thread ID is present.
Diff Detail
- Repository
- rCOMM Comm
- Branch
- master12
- Lint
No Lint Coverage - Unit
No Test Coverage
Event Timeline
lib/actions/activity-actions.js | ||
---|---|---|
32β34 | It really feels like we should extract this to a function | |
lib/keyserver-conn/keyserver-call-utils.js | ||
21 | This will throw in keyserverAuth if threadWatcher.getWatchedIDs() includes thick threads. Are we sure this can't happen? | |
native/chat/settings/thread-settings-push-notifs.react.js | ||
179β184 | I feel like this could be done inside of deviceTokenSelector, but not sure | |
web/calendar/entry.react.js | ||
479β481 | Shouldn't this be checking some connection to tunnelbroker? Probably not in this diff, but it feels like we shouldn't just leave this like that. There probably is a reason for checking if we can send the save entry request |
lib/keyserver-conn/keyserver-call-utils.js | ||
---|---|---|
21 | Great catch | |
web/calendar/entry.react.js | ||
479β481 | Looks like the only place this is used is to trigger auto-retries of failed entry updates once we detect the user has come online What @tomek is doing here is basically disabling that automatic retry. Checking Tunnelbroker connection would probably be appropriate, but it might not be necessary if we have a separate automatic retry mechanism, which I think @tomek is planning for this month (but tracked separately) Unrelated: I might suggest collapsing this all into a single const online = useSelector |