Soooo I opened a bug report about the thing I talked about yesterday, regarding broken thread and the «broken thread black hole». I succeeded at reproducing the issue locally so that will be way easier to fix.
Now that I’m able to reproduce the bug locally, it doesn’t seem hard to fix.
I guess the first part of the bug can be fixed by looking at the code that’s run on the static front-end and doing the same thing on the logged-in interface.
The second part is mostly finding where does the crash happen exactly, the fix in itself should be really tiny.
I think it can be a good way to get involved in Mastodon development.
I found a way to fix the rendering issue. That’s probably not a very clean way to deal with it but that’s already something.
https://github.com/tootsuite/mastodon/issues/7899#issuecomment-400857818
@lanodan I mean it doesn’t take into account other types of corruption? I don’t know.
Also now I’m lost about the other issue, do you know where to search?
> it doesn’t take into account other types of corruption
One bug at a time I guess.
And what is the other issue you are referring to?
@lanodan I want the UI to be resilient and it looks like a hack to me but maybe good enough for 99% of cases.
The other problem described in the Github issue is the fact that the thread isn’t loaded entirely, and it’s possible because it’s done in the static frontend.
> I want the UI to be resilient
I think it’s a very big and complicated issue, there is probably a lot of cases like that in the frontend and AFAIK there is no way in React to have it fault-tolerant or something like that.
And I don’t really know how to address this issue and yeah I can reproduce it, here too with a pleroma backend.
Ping @lanodan