Whatever we decide in the final conclusion for the 'decide status' questions
we better _actually_ turn this off when USE_ADMIN=False, this matches expectations
better
See #131
i.e. the code in check_for_thresholds (when the `qs` filters per project)
On SQLite the effect is like so, from:
EXPLAIN QUERY PLAN SELECT COUNT(*) AS "__count" FROM "events_event" WHERE ("events_event"."project_id" = 1 AND "events_event"."digested_at" >= '2025-12-07 15:12:24.479666');
QUERY PLAN
`--SEARCH events_event USING COVERING INDEX events_even_project_ac6fc7_idx (project_id=?)
to
EXPLAIN QUERY PLAN SELECT COUNT(*) AS "__count" FROM "events_event" WHERE ("events_event"."project_id" = 1 AND "events_event"."digested_at" >= '2025-12-07 15:12:24.479666');
QUERY PLAN
`--SEARCH events_event USING COVERING INDEX events_even_project_625413_idx (project_id=? AND digested_at>?)
The key phrase from our codebase was:
> # .save() will be called by the caller of this function
But this wasn't the case for the conditionally called path.
Adds data-fixing migration too.
Fix#292
500 was just fine when I was still calibrating the envelope parsing, but AFAICT it's
correct now, so I'd rather not get notified about 'remaining problems' (which are SDK problems)
Facilitate visual comparison with the 'reference impl' (slack).
The service backends do heavy copy/pasting, good to keep the beast
in check by at least limiting the changes to the bits that actually
diverge
W.r.t. the user-contributed version, this:
* removes commented-out code
* removes channels (not supported at the UI level)
* removes all other things
Said differently: the mattermost Alert Service Backend is simply the Slack version
with edits to make it mattermost-specific (and nothing else).
(In a few places I've edited the slack backend to make comparing easier).
Fix#277
See #253
I had someone run into this on hosted Bugsink; couldn't reproduce it.
Thought I fixed it in e8fb9556f7 (specific to the Chinese translation)
but appararently there's other ways to reach this point.
No matter, just create a version of the yesno filter that's not sensitive
to any future mistranslation.
i.e. running this:
```
python manage.py tailwind update
python manage.py tailwind build
```
prompted by:
https://github.com/advisories/GHSA-5j98-mcp5-4vw2
not (likely to be) affecting us (since this is for local builds only)
but might as well run the update to get rid of the warning the proper
way.
