| Commit message (Collapse) | Author | Age | Lines |
| | |
|
| | |
|
| |
|
|
|
| |
We're no longer going to use a two-channel setup for this event, as we
don't want to split the event community into two, staff and non-staff.
|
| |
|
|
|
|
| |
This helper function was only being used in one spot and did not factor
out any logic. I've removed the helper function to just create the embed
where it's needed.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
I've rewritten the Advent of Code leaderboard logic. Unfortunately,
nearly all of the changes made are interrelated, meaning that they've
ended up in the same commit.
To add a bit of structure to the extension, I've chosen for a subpackage
structure instead of a single file structure.
The biggest changes:
- Whether or not you get a join code for the staff leaderboard will now
be determined by looking for the Helpers-role.
- The Python Discord Leaderboard now includes all boards, including the
staff leaderboard. This is one event.
- Redis is now used to set a cache expiry period. This means that our
code does not have to check for cache staleness; Redis will do that
for us.
- The period "fetching" task has been removed. We now fetch solely when
the data is needed to prevent putting unnecessary stress on the Advent
of Code website.
- The option to display the Global Leaderboard within Discord has been
removed. Rather, we now link to the website. This simplified the code
for now, although we could add it back later.
- An additional command, `.aoc refresh`, has been added to allow Admins
and the Events Lead to force the cache to be invalidated. This should
be done sparingly to not overburden the AoC website.
- I've also made sure that the daily notification task actually pings
the notification role by setting the `allowed_mentions` kwarg.
|
| |
|
|
|
|
| |
As the Advent of Code file was getting massive, I've removed the old
single-file based extension as I'm going to replace it with a partially
rewritten subpackage-based extension.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
I've added a unified approach to setting data for multiple leaderboards
using environmental variables. Instead of setting separate variables for
the three pieces of data we have, hoping that the position of each board
matches up in the three fields, I now set one environmental variable,
AOC_LEADERBOARDS, that holds delimited data.
The data is in the format:
board_id1,session1,join_code1::board_id2,session2,join_code2[::...]
The staff leaderboards should be included as usual in this environmental
variable. Another environment variable, AOC_STAFF_LEADERBOARD_ID, can be
used to designate which leaderboard should be used as the staff board.
I've also made some other constants configurable in this commit and
added the role ID of the Events Lead role to allow the Events Lead to
force a reload of the leaderboard cache.
|
| |
|
| |
Co-authored-by: Joe Banks <[email protected]>
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |\ |
|
| | |
| |
| |
| | |
We no longer have this channel, so this cog serves no purpose.
|
| | |
| |
| |
| |
| | |
I've tried to replace this with generic references where appropriate,
but a lot of the time it just doesn't make a lot of sense to do so.
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | | |
|
| | |
| |
| |
| | |
This commit implements a wonder twins inspired command. This is a purely fun command that uses real transformations from the show to make random new transformations. The yaml is all hand transcribed from the actual show.
|
| |/
|
|
|
|
|
|
|
|
|
| |
We ripped some features out of the bot to make the migration to
kubernetes easier. However, that process left us with some linting
errors that will play up if anyone makes a PR.
I've added "noqa" to the valid TODO lines and commented out lines that
introduced now unused names or tried to use now undefined names. This
should have no influence on the bot, as it does not touch code that is
actually being run at the moment.
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |\ |
|
| | |\ |
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Since we disable both privileged intents, we can just construct an
instance with them already off. As Senjan notes, this will also
future-proof us against any privileged intents added in the future.
Co-authored-by: Senjan21 <[email protected]>
|
| | | | |
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | | |
With 'd.py' 1.5.1, the Member converter will now lazily fetch the
member object.
It is believed that this removes our need for the intent.
|