- England
- https://timedout.uk
-
Continuwuity maintainer and resident spec pedant.
-
Joined on
2025-04-13 -
-
https://timedout.uk
hi I make computers talk to each other sometimes
see also: https://nexy.blog
idk read my website there's more interesting things on there
Please discuss such large changes with the maintainers ahead of time :/
Using an A tag as a button is not semantically correct
This is not how sync works - the rooms will be included in rooms.join when they have new events, but with full_state=true, they will also be included with the full state before the timeline,…
I don't use docker but these changes look sensible to me
This PR is overall very complicated, poorly documented, has a plethora of codestyle issues, and I can't help but feel like an LLM was used here. Frankly, I think it might be a better idea to break…
nit: references to an MSC should include a link to said MSC, either hyperlinked or referenced
This docstring doesn't actually describe anything about the function (the first sentence of a doc string at the very least should summarise what the function does)
nit: shortstatehash_after_count would be more a intuitive name
This docstring doesn't actually describe anything about the function (the first sentence of a doc string at the very least should summarise what the function does)
These were removed ages ago, I thought you said this was built based on the merged spec proposal?
Given this is only interesting for debug builds, it might be a good idea to feature-gate this behind debug_assertions
This is a complicated calculation that could probably just do with being split up
nit: references to an MSC should include a link to said MSC, either hyperlinked or referenced
this should probably be un-nested because it's hard to follow