dev-contrib

[rapid igloo]

haha, that could work for colors in bulma settings

now coming to think of it, there might've been a few places where I had to manually adjust some color variables to be dimmer/brighter, then set in the css (due to limitations of css as compared to scss), though I tried to keep that a minimum.
I could go and look through them when I get the time, but I'm not sure if it'll be too helpful to note these down, because if someone is working on that area specifically they will see it in the css file (where I have added comments)

[crude gyro]

I just figure, if they're changing stuff inside that dictionary, they might think the change will affect everything on the site, and it won't. But the tip you gave about search and replacing the hex is pretty good, I think.

just having that bit of information in a nearby docstring may save some frustration for someone in the future

I wouldn't start listing every single place you needed to make changes

just more of a general heads up

[rapid igloo]

I think the comments in the css is enough

[rapid igloo]

as far as I know in terms of color variables there, it does change everything -- where we've used those classes

[crude gyro]

mm, alright, fair

[rapid igloo]

I could PR something like this soon 👍

[vale ibex]

sir-robin#112 wil make it easier to stop issueing the completionist role this year, and future years, ratehr than having to comment out the line each time

[dusky shore]

GitHub

[sir-robin] #112 Command to toggle issuing AoC completionist

[vale ibex]

There's also a bunch of general improvements in the PR that I did while writing & testing this feature

[thorny obsidian]

I'll review later today

[vale ibex]

Nice, ty

[vale ibex]

@crude gyro where are youlooking for the test output in CI?

I don't think we output a list of tests that ran in the CI anywhere

[crude gyro]

the coverage report shows the files that were run

the file wasn't in that list

[vale ibex]

https://github.com/python-discord/bot/actions/runs/7371282307/job/20058499764?pr=2871#step:7:45

https://github.com/python-discord/bot/actions/runs/7371282307/job/20058499764?pr=2871#step:7:198 for the test file too

we have test coverage on the tests 💪

[crude gyro]

testing the tests

next level

[timid sentinel]

Did you try to find it by ctrl f before by any chance?

[crude gyro]

I did that first, but that didn't work, so I skimmed the list and I probably just missed it

I'm very old, you see

[timid sentinel]

because that's something I hate about github. It lazily loads stuff so you can't just ctrl-f

stupid

[crude gyro]

mfw looking for a test

[vale ibex]

Ah yes

[crude gyro]

hahaha

[timid sentinel]

found it!

[crude gyro]

oh thanks

no wonder I couldn't find it they all look the same!

[vale ibex]

pytest should encode the test number in the rgb values of the dot

[crude gyro]

to make a beautiful rainbow

[vale ibex]

Yup, and since we use pytest-xdist for multiprocessing, they'll all be in a random order too

[royal prawn]

i know you shouldn't, but what if you have 17 million tests? also, should there be buffer zones for easier identification? less totals if so

[vale ibex]

16-bit colour it is then

[royal prawn]

aren't the monitors that can even display 10-bit colors high-end now?

oh, that's 10 bits per channel, 16 is 5+5+6. nevermind

[vale ibex]

yea, and greenshould get the extra bit, for obvious reasons

[crude gyro]

just use ROYGBIV and then rotate back to R

raaainbooow

can we please make the tests a little gayer

I vote we run the whole output through lolcat

[royal prawn]

it might be neat, and possibly convincable, to get different test functions/classes/files/etc as different colors

[vocal wolf]

bot#2580 is in need of reviews. be there or be less there.

[dusky shore]

GitHub

[bot] #2580 Handle adding reactions to messages from users who have blocked the bot

[vocal wolf]

@timid sentinel henlo give review thanks

[royal prawn]

i take that discord.py doesn't have a feature to detect that in advance?

[vocal wolf]

if it does exist, I haven't found it

[wind ruin]

• that is not on topic for this channel
• see #voice-verification

[dim pelican]

.bm

[vocal wolf]

@outer oasis Fix. thanks. sir-lancebot#1404

[dusky shore]

GitHub

[sir-lancebot] #1404 Add a space before "tree" in the "christmas" tree holiday reaction

[vocal wolf]

@tame pebble Yo what's good gamer, can I get uhhhhhhhhhhh sir-lancebot#1197

[dusky shore]

GitHub

[sir-lancebot] #1197 Make deleting a bookmark a context menu command

[covert fog]

i feel like there might be some flaws here

oh

there were but in my testing

[waxen jasper]

hey, i want to contribute, how does the process go?

[rapid igloo]

!contribute

We have a bunch of open source projects mostly written in python. There are detailed guides on how to set up the development environment on our website linked below.
Sir-lancebot is a good place to start, it's a discord.py project with fun commands and events/info related commands.

Contributions happen through opening issues and PRs on the GitHub repos. Feel free to discuss in this channel too.

[stable mountain]

Contribute to Python Discord's open source projects

Looking to contribute to Open Source Projects for the first time? Want to add a feature or fix a bug on the bots on this server? We have on-going projects that people can contribute to, even if you've never contributed to open source before!

Projects to Contribute to

• Sir Lancebot - our fun, beginner-friendly bot
• Python - our utility & moderation bot
• Site - resources, guides, and more

Where to start

1. Read our contribution guide
2. Chat with us in #dev-contrib if you're ready to jump in or have any questions
3. Open an issue or ask to be assigned to an issue to work on

[outer oasis]

I uh...

I actually need help with regex

I'll go poke pygen in a minute

[waxen jasper]

im tryna figure out gitpod but its really confusing, it says i need to have cloned the repo but then gitpod opens a ssh connection

i dont want to use ssh and instead want to do it locally

do i have to use gitpod then?

[dim pelican]

No, you can use Docker instead

[outer oasis]

No, it's just an alternate option
You can setup your enviornment however you want

[waxen jasper]

ah, so a normal venv should be fine?

[outer oasis]

Yeah, with Docker for any supporting services

[timid sentinel]

For lancebot you don't even really need docker.

[waxen jasper]

ah, thats easier, my docker installation isnt working, it crashes for some reason nvm fixed it by running it in administrator

[timid sentinel]

You just need to remember add REDIS_USE_FAKEREDIS=true to your .env for it to work

[tame pebble]

You can, but there's duplicated error handling happening and I want to avoid that, which is why I opened bot-core#198, but unfortunately I didn't get any feedback yet.

[dusky shore]

GitHub

[bot-core] #198 Centralize error handling

[waxen jasper]

why am i getting this?

Installing the current project: sir-lancebot (0.1.0)
The current project could not be installed: No file/folder found for package sir-lancebot
If you do not want to install the current project use --no-root```

poetry install this is the command btw

[vale ibex]

that's fine to ignore

[waxen jasper]

aight!

[vale ibex]

FWIW it's a fairly new thing in poetry, geared towards library maintainers

for projects like our's it doesn't make sense. https://github.com/python-poetry/poetry/issues/1132 and https://github.com/python-poetry/poetry/pull/8650 look to make this better

[outer oasis]

[vocal wolf]

I see. Thank you for update, I will poke at people to get feedback, and take a stare myself.

I have pinged the almighty chris in the issue, as well

[tame pebble]

Thanks, I'll take care of the rest

[timid sentinel]

Does anyone have any idea why on https://www.pythondiscord.com/resources/ in production the titles are in bold

but locally they aren't

I'm pretty sure I'm running on main locally

it seems that the css from bulma.css is being applied in a different order in prod, is cloudflare messing up the css when minifying it??

[vale ibex]

I've enbaled CF cache dev mode so it temporarily doesn't use the cache

and it still seems bold

[jovial barn]

so Chris.

are you software developer?

[vale ibex]

Want me to leave it on for a bit?

ah, I've also just disabled CSS minification

which is separate from the cache

[timid sentinel]

The HTML element controlling the text has the card-header-title and subtitle classes, both of which are defined in bulma.css. I tried prettifying the minified version from prod and comparing it to the version served when I run locally, and it seems the relevant definitions affecting the font weight are in a different order on each:

prod:
line 4822: .subtitle font-weight: 400 
line 10492 .card-header-title font-weight: 700

dev:
line 675 .card-header-title font-weight: 700
line 6218 .subtitle font-weight: 400

suggesting it could be the minification, but that seems like such an obviously breaking transformation...

[vale ibex]

how weird

ntos ure where it's getting minfied either

I've disabled everything in cf

do we have something in site itself that's minifying statically served files?

[timid sentinel]

headers suggest it's still being cached+minified for me

[vale ibex]

I turned it back on

one sec

try now

[timid sentinel]

ok now it's not minified, but it's still different, so it seems the css is different before minification/cloudflare

[vale ibex]

its bold for me lcoall;y too

have you rebuilt your docker image recently?

[timid sentinel]

I'm not using docker, i'll try just remaking my env

[vale ibex]

have you ran collectstatic?

lines 25-34 in the dockerfile

[timid sentinel]

aaaaaah

got it

lol

that was painful

I think it's because django simple bulma uses iterdir here, which affects the order the order of the imports in the scss, which affects the order of the compiled css https://github.com/lemonsaurus/django-simple-bulma/blob/f26264f2155bc30f8afc5153e9245937193c33da/django_simple_bulma/finders.py#L116

[stable mountain]

django_simple_bulma/finders.py line 116

for dirname in self.bulma_submodule_path.iterdir():```

[timid sentinel]

And iterdir returns the contents of a directory in an arbitrary order

[vale ibex]

ah lol

[outer oasis]

Is there a method that garunteees order?

[timid sentinel]

I don't think so, you can sort the result fairly easily though

what i'm not sure about is if there's a "correct" order

other than just being consistent with e.g. alphabetical order

[pseudo gorge]

Can anyone tell me what I'm doing wrong? I'm trying to install a simple program with pthon

[outer oasis]

#❓｜how-to-get-help

[pseudo gorge]

yea..

[timid sentinel]

I opened an issue here https://github.com/lemonsaurus/django-simple-bulma/issues/98 CC @crude gyro

[thorny obsidian]

@vale ibex how did you test the scheduler changes for AoC?

[rapid igloo]

I've also been noticing this issue on resources page, where some icon images (particularly svg, I think) that fails to get styled to the correct width on production site

but they're fine on the netlify preview (and probably local)

I've only observed this to happen on mobile, chrome-based browsers

it may, or may not, have been introduced after dark mode, but I've double-checked that the commit which may affect this does the image styling correctly

[vale ibex]

I used one of the leaderboards I'm in

[vale ibex]

ah I saw your comment now, IU thought the question was about the task changes

I tested the other scheduler stuff by changing the arrow.now(EST) call at the start of the while loop within the tasks to different times last year

[vale ibex]

Also, I lied in github, for some reason sir-robin doesn't hook up to sentry

I'll fix that later

[waxen jasper]

hi @timid sentinel, i noticed u added this comment, i didnt quite understand what u meant by it

pertaining to PR #1440

[vale ibex]

which repo?

[waxen jasper]

sir lance bot

[vale ibex]

Ah yea I see what wookie means

these changes

[waxen jasper]

oh thats cause of #black-formatter

[vale ibex]

yea, we don't use that on our repo

we use ruff

[waxen jasper]

sorry about that!

i run ruff at the end tho

[vale ibex]

since we use ruff, we have ruff line length configured, but not black

[waxen jasper]

alright, ill change that

also another question

[vale ibex]

sure

[waxen jasper]

so for the same command

there are a few class variables

but they are all used in one and only one command

should i convert them to normal variables?

[vale ibex]

do they need to persist across command invocations?

[waxen jasper]

i think the current channel needs to

since it redirects u to the channel

but i dont think self.correct needs to stay

[vale ibex]

alright, if it doesn't need to persist accross different inocations, then yea we can make them normal variables

[waxen jasper]

aight!

[thorny obsidian]

Does the sentry alert propagate anywhere in server? Or would I have to check sentry for it?

[outer oasis]

Speaking of, @vale ibex I've been meaning to talk to you about Sentry
It marks my commits as being included in several releases, even though we release on every merge
It always says "20 commits [in this release]" -- is it always just taking the last 20 commits?

I don't really know where to start looking with this one
But I did notice that we use a custom action: https://github.com/python-discord/sir-lancebot/blob/main/.github/workflows/sentry_release.yaml#L13
Is there a reason for that?
I've only ever used their offical action, and it's always been fine for me: https://github.com/letsbuilda/anubis/blob/main/.github/workflows/sentry-release.yaml#L15-L22

[stable mountain]

.github/workflows/sentry_release.yaml line 13

uses: tclindner/sentry-releases-action@v1.2.0```
`.github/workflows/sentry-release.yaml` lines 15 to 22
```yaml
- name: Create Sentry release
  uses: getsentry/action-release@4744f6a65149f441c5f396d5b0877307c0db52c7 # v1.4.1
  env:
    SENTRY_AUTH_TOKEN: ${{ secrets.SENTRY_AUTH_TOKEN }}
    SENTRY_ORG: ${{ vars.SENTRY_ORG }}
    SENTRY_PROJECT: ${{ vars.SENTRY_PROJECT }}
  with:
    version_prefix: "${{ vars.SENTRY_PROJECT }}@"```

[outer oasis]

Should hit #dev-alerts , at least the first time it occurs

[vale ibex]

#dev-alerts

[vale ibex]

no idea on the history of why we use that action

sir-robin#114 for sentry integration

[dusky shore]

GitHub

[sir-robin] #114 Add sentry integration

[timid sentinel]

IIRC the official one is not really maintained

[outer oasis]

I keep forgetting how much lockfile changes inflate diffs

[timid sentinel]

oh actually it might be good now

[outer oasis]

Would you accept a PR on Lance trying out switching to it?

[vale ibex]

Sure

[outer oasis]

Cool
I'll poke at that later today

[timid sentinel]

Yeah would have to double check the reason we weren't using it but this PR from 3 weeks ago updates it from using a 3 year old sentry CLI version so it seems like it's been revived https://github.com/getsentry/action-release/pull/186

[thorny obsidian]

@vale ibex is the profiles_sample_rate no longer an experiment?

[vale ibex]

nope

but they made it backwards compat, so setting it in experiments still works

[waxen jasper]

first PR :D

[outer oasis]

Poor @humble leaf

Being made an example of

[timid sentinel]

I think that's a different issue. I can recreate it on desktop in firefox and chrome but only when using a mobile user agent. I think it might actually be cloudflare's fault this time. @vale ibex do we have cloudflare mirage enabled (https://developers.cloudflare.com/speed/optimization/images/mirage), if so could you try temporarily disabling it?

[vale ibex]

we did, it's off now

[vale ibex]

Congrats! It was a nice one too :D

[timid sentinel]

Ok nice, I think that does fix the issue

[vale ibex]

fwiw

[timid sentinel]

My guess at what is happening is that there's a bug in cloudflare mirage's lazy loading. The way it's meant to work is that if it detects you're on a mobile or slow device, then it transforms images in the HTML

<img class="dark-image" src="/static/images/resources_dark/sentdex.png" alt style="height: 50px; ">

to something like this:

<img class="dark-image" data-cfsrc="/static/images/resources_dark/sentdex.png" alt data-cfstyle="height: 50px; " style="display:none;visibility:hidden;"><noscript><img class="dark-image" src="/static/images/resources_dark/sentdex.png" alt style="height: 50px; "></noscript>

And injects some JS that detects when the user has scrolled to the image and sets src and style to the values in data-cfsrc and data-cfstyle, deleting the style that made it hidden (thereby lazy loading it and saving bandwidth).

However, I think that the backend HTML transformation doesn't happen for SVG images, but the style attribute is still removed by the JS when the image is viewed, so the height styling is removed.

[vale ibex]

ah right, yea that would make sense

to cloudflare's credit, it is still marked as a beta optimisation, so something like this isn't un heard of

Any chance of a second review on sir-robin#112

[dusky shore]

GitHub

[sir-robin] #112 Command to toggle issuing AoC completionist

[tame pebble]

I'll take a look at it if i make it home early and no one beats me to it

[last patio]

I'm very confused, isn't lazy loading a feature with the stock img tag

[vale ibex]

Nice , thanks

[timid sentinel]

yeah, I didn't know but it looks like you can set loading="lazy" on an img tag to get it to load lazily. I've no clue what benefit/difference the way cloudflare seems to do it would have over that.

[rapid igloo]

that makes sense! thanks for finding the issue wookie.

[rapid igloo]

looking at the docs, I guess it also does some other optimisation things in addition to lazy-loading

and also @timid sentinel, regarding the font awesome thread was just wondering, what specifically do you mean that django fa plugin can help with it?
AIUI lemon suggests we stop using it entirely and host the icons ourselves (imo because we don't want icons to be served only "sometimes" due to the 10k limit).
doing so would:

• drop the js/css/fonts baggage of resources
• prevent inclusion of icons we don't need
  although using that plugin can let us include only css/js files that is needed on the page, it's still under the usage limit right?

btw I'd also include that switching to host our own icons also has these disadvantages

• might need to write some css for fine-tuning the icon styling
• less easy to try-out multiple icons before deciding the one to use during development

[timid sentinel]

I meant https://fontawesome.com/docs/web/use-with/python-django, which which would be a drop in replacement. It just bundles all the js/css/fonts into Django's static files so all that stuff would be served by us and we don't need to worry about kits/tokens/their cdn limits etc. (though the django plugin isn't really necessary for that, it could be better to just copy the files we need into the repo ourself)

Trying to manage each icon we want individually sounds like a bit more of a pain, the current way the icons are packaged using css and fonts is quite convenient, so I'm not sure if it would be worth adding complication just for a bit of a performance increase. I don't know exactly how that would work though so I can't really tell.

[rapid igloo]

ahh I see, my bad 👍

[crude gyro]

I think I like this approach, seems simple enough.

doubt it'll take much of a performance hit.

[waxen jasper]

what would be a good way to figure out where the head of an avatar img is? i can use cv2 but then the cascades will be a problem since it wont work for everything

[vale ibex]

Sounds like a non-trivial problem. Most facial recognition models are built to detect human faces, but I'm going to hazard a guess that they'll be a small minority of profile images

[waxen jasper]

but I'm going to hazard a guess that they'll be a small minority of profile images
but a lot of people use pfps which arent humans

[vale ibex]

Yea, that's my point, you'd need to find a model that's works for those

But most are trained on human faces

[waxen jasper]

oh sorry, i mistinterpreted that 😅

[waxen jasper]

might have to train one myself

but dont really have the processing power

any other ideas?

[rapid igloo]

I agree that the advantage of using fa's framework of css classes outweigh the possibly negligible decrease in performance. And since (I just realized) this plugin lets us host the static files ourselves which doesn't seem to be affected by the usage limits. It's a great idea

[vale ibex]

What is this for?

[waxen jasper]

the christmasify command

[vale ibex]

and what is that command going to do?

[waxen jasper]

its like the easterify and spookify commands, so i thought maybe we add ice/snow and then a santa hat

[vale ibex]

Ahh right

Maybe a wreath around the edge, rather than a santa hat then?

similar effect, with way less complexity

[vale ibex]

Cool, sir-robin redis data migration was a success

Also updated the global namespace to be sir-robin instead of bot to make it easier to see what is what

&aoc toggle

[humble leaf]

:+1: Completionist role issuing is now on.

[vale ibex]

fyi @thorny obsidian

[tame pebble]

This has a one character change
bot-core#203

[dusky shore]

GitHub

[bot-core] #203 Bump version to 1.7.0 in pyproject.toml

[tame pebble]

The version's stuck in the past

[vale ibex]

ah that's annoyiong that I missed that

the tagged version will have 10.6.0 now too

Ah well, this fixes the site at least

[tame pebble]

Yeah, that's how I first saw it

[vale ibex]

will wait for CI to finish and merge

[tame pebble]

[vale ibex]

@thorny obsidian I also just noticed sir-robin#75 do we still want to increase the time of the loop?

[dusky shore]

GitHub

[sir-robin] #75 AoC: Increase task time for Completionist Role & create a stop/start command for Admins/Events Lead

[vale ibex]

fwiw the loop uses the leaderboard cache, so it's not a huge amount of requests actually making it to aoc

[thorny obsidian]

I think it's fine

[vale ibex]

👍 Closed the issue

[atomic ivy]

Do we have a command to convert Discord timestamp format or a "human readable datetime" to a duration understood by !remind or other such commands?

[tame pebble]

Yeah we do

!source remind

[stable mountain]

Command: remind

Commands for managing your reminders.

Source Code

Go to GitHub

bot/exts/utils/reminders.py:277

[tame pebble]

It's in bot.converters

let me double check

!remind "in 2 hours" to go to sleep

[stable mountain]

Bad argument

in 2 hours is not a valid duration string.

Command Help

!remind [mentions]... <expiration> [content]
Can also use: reminder, reminders, remindme

Commands for managing your reminders.

The expiration duration of !remind new supports the following symbols for each unit of time: - years: Y, y, year, years - months: m, month, months - weeks: w, W, week, weeks - days: d, D, day, days - hours: H, h, hour, hours - minutes: M, minute, minutes - seconds: S, s, second, seconds

For example, to set a reminder that expires in 3 days and 1 minute, you can do !remind new 3d1M Do something.

Subcommands:

!remind delete [ids]...
Delete up to (and including) 5 of your active reminders.
!remind edit
Commands for modifying your current reminders.
!remind list
View a paginated embed of all reminders for your user.
!remind new [mentions]... <expiration> [content]
Set yourself a simple reminder.

[tame pebble]

Well

I'm the one who needs to go sleep xD

As I apparently am mistaken

Anyway, I could've sworn I've seen it in action before

[thorny obsidian]

Would be nice if we could have remind accommodate human readable stuff like the epoch command does

[atomic ivy]

It'd be nice to have a converter command for this like .epoch

Because there are times when you need it for something other than !remind

Such as… ModMail

[rapid igloo]

hello, any ideas why the netlify preview workflow failed for my latest commits in site#1210?

8:21:34 PM: Fetching download URL from https://pythondiscord.com/api/github/artifact/python-discord/site/608cd9580be66df87b63edb051d625917177e165/CI/static-build
8:21:34 PM: {"error_type": "NotFoundError", "error": "Could not find a run matching the provided settings in the previous hundred runs.", "requested_resource": "python-discord/site/608cd9580be66df87b63edb051d625917177e165/CI/static-build"}

https://app.netlify.com/sites/pydis-static/deploys/65bcde2f388d7e00088eae56
The error doesn't seem to suggest it was caused by something from my commit changes

[dusky shore]

GitHub

[site] #1210 Redesign of events page

[rapid igloo]

it looks like publish-static-preview somehow did not complete before netlify starts attempting to fetch the artifact it was supposed to publish...

[vale ibex]

it's because the "normal" github CI doesn't run if there are merge conflicts

and it's our CI that builds the static site that netlify uses

if you rebase onto main, resolve conflicts, and force push, it should just work

[outer oasis]

Why can’t Netlify build the site itself?

[vale ibex]

reasons that I can't remember

you can probably find the PR from scaleios that added the static generation to see why

[vale ibex]

ah yea, netlify only has support for Python 3.8 in its build system

well, and 2.7

[outer oasis]

🥴

[rapid igloo]

ahh I see

thanks!

[rapid igloo]

(in reply to chris but the messages further down did not load)

[karmic silo]

can sir lancebot react with something everytime “python” and “slow” are in the same message

[waxen jasper]

someone should make the bot react with "🤓 ☝️ " every time someone says that

[outer oasis]

What would you want to say?

""?

[karmic silo]

that works

[chrome lake]

guys

can someone help me

build

soft aim

[rapid igloo]

re: #advent-of-code message

would it be useful if /aoc join could still let people join the leaderboard for the previous year, when the next event has not started?
I guess the lb is most useful for competing in time to solve, but the number of stars also gives a placing in private leaderboards.

[outer oasis]

(the leaderboard doesn't change from year-to-year)

[rapid igloo]

yes. what I mean was say someone run /aoc join right now, they cant join the leaderboards, because aoc is not currently happening

yknow, in case they want to catch with 2023 puzzles

[outer oasis]

I would add a footnote that says "this event is not currently running; it happens in December"

It wouldn't break anything to let people join
But these people are confused and think that the event is currently ongoing, and letting them join a leaderboard would only add to that confusion

[rapid igloo]

ah that's a good point

[vale ibex]

the aoc join command requires up to date leaderboard cookies, which expire after ~1 month

so we usually only keep on top of those during the event

[outer oasis]

[vale ibex]

it's to make sure we're not giving out join codes to full leaderboards

[rapid igloo]

so I've been looking at this:
https://github.com/python-discord/sir-robin/blob/5d26dcebd1d964c594dae66ed71bd31573c7e2d3/bot/exts/advent_of_code/_cog.py#L212-L215

and in constants for AocConfig.year we have this:
https://github.com/python-discord/sir-robin/blob/5d26dcebd1d964c594dae66ed71bd31573c7e2d3/bot/constants.py#L78

it looks as though we are comparing datetime.now().year == datetime.now().year+1 which should never be true.

is this something unintentional that happens to work, so we'll say "leaderboard for this year not yet available" for any month other than Nov and Dec?

[stable mountain]

bot/exts/advent_of_code/_cog.py lines 212 to 215

if not (
    current_date.month in allowed_months and current_date.year == AocConfig.year
    or current_date.month == Month.JANUARY.value and current_date.year == AocConfig.year + 1
):```
`bot/constants.py` line 78
```py
year: int = datetime.now(tz=UTC).year```

[vale ibex]

the value in bot.constants is the default

we overwrite it using an env var though

https://github.com/python-discord/infra/blob/main/kubernetes/namespaces/default/sir-robin/configmap.yaml#L6

[stable mountain]

kubernetes/namespaces/default/sir-robin/configmap.yaml line 6

AOC_YEAR: '2023'```

[vale ibex]

The idea being that it will work Nov, Dec & Jan

[rapid igloo]

hmmm

ohhhh we're in February!

sorry about that

thanks!

[subtle kraken]

Henlo @tame pebble are you able to finish bot#2839 ? Should I label is as up for grabs and look for someone else to finish it?

[dusky shore]

GitHub

[bot] #2839 Support editing of timeout durations.

[tame pebble]

Hello,

Yes I'll wrap it up

[subtle kraken]

Thanks.

[tame pebble]

Sure!

[subtle kraken]

I'll be cheering

[vale ibex]

Also @lyric temple thank you for the beautifully written issue

[lyric temple]

Thanks for the easy to follow setup for dev environments. Had a bot up and running in minutes for local testing.

[outer oasis]

@lyric temple
"\u2764\ufe0f"
"\u2665\ufe0f"
"\U0001f49c"
"\U0001f49f"
Why do only the first two have 2 \u(s)?

https://stackoverflow.com/questions/38100329/what-does-u-ufe0f-in-an-emoji-mean-is-it-the-same-if-i-delete-it
I see
Do they not all need it?

[lyric temple]

<#bot-commands message>
It's a great question. I'm not sure on the rules of unicode but most didn't.

[outer oasis]

Alright then

[lyric temple]

If you're curious I spot checked encoding errors with: https://gist.github.com/Preocts/d5fd9fb231e5af5f06e3a324ac418b7b

[outer oasis]

I'll never understand Unicode
PyCharm running locally can't see the emojis
But a Codespace in the browser can

[tawdry vapor]

The first two are defined as a pair of an existing character and a combining character called a "variation selector".

The latter two are characters outside the original 16-bit code point range (BMP). These characters are defined as surrogate pairs (which use a reserved section of the BMP) and you'd think they could be represented as \uD83D\uDC9C in Python, but Python does not support encoding surrogate pairs for reasons I don't know. So instead we have to represent the character as a single code point.

[rapid igloo]

hey @thorny obsidian, would you mind taking a look at the "annual" and "other" events list to check whether the order and list of events is correct? just want to make sure the high level structure is there in terms of the names of the events to include, before marking PR as ready for review

https://deploy-preview-1210--pydis-static.netlify.app/events/

(note that only light mode is fully readable for now)

[subtle kraken]

Is our site supposed to advertise last years codejam?

[rapid igloo]

I've included the previous events there for returning users to quickly visit an event they might have participated in, and want to check some info (such as winners, submissions etc)

and also to showcase the sort of events we hosted recently to give newer visitors a sense of what to expect

[rapid igloo]

oh! I just realized you might have meant the hero section on the homepage

[subtle kraken]

Yes

It looks really weird to advertise event that has already occured

I presume it'd be better to either more explicitly indicate that its over and its a previous event (I myself assumed it meant 2024 and it was merely a typo)
Or to start advertising the 2024 one tbh

[thorny obsidian]

Pixels can be bumped down to "Workshops & Other events", the Game Jam can be left off entirely. Otherwise looks great!

[rapid igloo]

Alright, we'll have only two out of three items in the second row for annual events then. Will update accordingly, thanks!

[atomic ivy]

Bug report: @stable mountain suggests the code block language incorrectly, when the original code block language is specified as Python (with P capitalized).
#python-discussion message

Looks like they deleted the message.

    def fn():
        print("hello, world!")```

This doesn't trigger for me, but should for any non-staff, I think

[tawdry vapor]

I suppose Discord made it more lenient now.

[hoary haven]

hallo! this issue is now up for grabs again. i would love to see a PR for it https://github.com/python-discord/bot/issues/2172

[tawdry vapor]

looks like someone really liked that

[hoary haven]

!reminder edit duration 6557 8d12h5M

[stable mountain]

Okay.

That reminder has been edited successfully! It will arrive on <t:1708305299:F>.

ID: 6557

[hoary haven]

woohoo woohoo 🎉

thanks mark!

[tawdry vapor]

Np

[eternal owl]

hey, anyone with M series mac had issues setting up site repo with docker ?

[atomic ivy]

What kind of issues? Snekbox?

Oh wait, snekbox doesn't get used by site, does it?

[eternal owl]

https://paste.pythondiscord.com/4EUA

the entire error is here, its a m3 mac
it works fine on linux though (I am new to mac)

@atomic ivy

[atomic ivy]

Seems like you're missing gcc

You can install it with brew

[tame pebble]

It's not saying command not found, so that shouldn't imply the command is missing, no ?

It's saying that the command failed because there's a missing file/dir required for the installation, but i'm not sure why

[drowsy ravine]

Mina,
Do you know of any other beginner friendly contribs?
Im trying to get more involved and better at git, but I’m scared! :-p

[rapid igloo]

we have plenty on sir-lancebot, a few on site too! :)

[drowsy ravine]

Oh good one. I actually havent looked at lancebot. Will have a look when at work!

[atomic ivy]

Hmmm, I'm on an M1 and have gcc and it doesn't error on poetry install
BUT, I also just realized that that error is coming during docker build, which is… weird

[rapid igloo]

on site:

• https://github.com/python-discord/site/issues/1217
• https://github.com/python-discord/site/issues/1214
  on lancebot, after a brief look
• https://github.com/python-discord/sir-lancebot/issues/707 (it looks like vcokltfre is no longer on it)
• https://github.com/python-discord/sir-lancebot/issues/824
• https://github.com/python-discord/sir-lancebot/issues/945 (depending on what you implement, this might take some more work)

there are also a few issues on sir-robin that looks pretty beginner-friendly, not sure if they're high on priority though but probably nice-to-have:

• https://github.com/python-discord/sir-robin/issues

[atomic ivy]

Wall

Oh, the docker build also fails for me.

[atomic ivy]

Adding a platform: linux/amd64 in the compose for site works.

   web:
+    platform: linux/amd64
     build:

[tame pebble]

try DOCKER_DEFAULT_PLATFORM=linux/amd64 docker compose up --build

I just remembered

Tizzy had a similar issue before

[tame pebble]

And Chris suggested this fix

[atomic ivy]

That also works, yeah

[eternal owl]

cool, lemme try that

works 👍

[tame pebble]

Nice

[drowsy ravine]

At work now, checking

[rapid igloo]

hey @drowsy ravine by the way you didn't join the org on github?

[drowsy ravine]

Eeuehm, did I not? 😛

How can I ?

[rapid igloo]

hmm it looks like you didn't.
but either way, once you do you can just create branches off the main repo, rather than maintaining your own forks :)

[hoary haven]

what's your github username?

[drowsy ravine]

JelmerKorten

[hoary haven]

invite sent https://github.com/python-discord

[drowsy ravine]

You are now a member of Python Discord!

[hoary haven]

belated but better late than never :D

[waxen eagle]

Hello. everyone.

I am looking some one who can support me.

Plz reply to me if you mind .

[vale ibex]

What do you need support with?

[waxen eagle]

Hi. Chris.
Thank you for my reply.
where are you from?

I am Umar Chandra from Japan.

I'd like to discuss something with you.

[opal kite]

do you have some problems in python ?

[waxen eagle]

No..

[opal kite]

what's the problem ?

[waxen eagle]

I am a developer who have rich experience in backend.
I worked at a Corvee company for 3 years, but the company was broken.
And I fired, so I'm finding some work or jobs.

Can you help me?

[vale ibex]

You're better off asking in #career-advice or in one of the off topic channels

This channel is for discussing Python Discord projects

[waxen eagle]

Oh.. I see, sorry for interrupting to your.

[vocal wolf]

@sharp crag thing

bot#2614

[dusky shore]

GitHub

[bot] #2614 Show timestamp in pypi command

[drowsy ravine]

Afternoon all!
Can anyone talk me through, or link me to a document, on the steps to follow on merging the stuff I worked on? (I think the term is "pull request"?)
(contributing noob here)

[rapid igloo]

hey, so you've made your commits locally?

[drowsy ravine]

Yessss 👍

[rapid igloo]

have you pushed to the remote branch?

[drowsy ravine]

Also yes

[rapid igloo]

alright now can you navigate to the repo on github? if you've pushed to a feature branch from the python-discord repo, open the /pulls page on github

it should show you a button to open a pull request

[drowsy ravine]

I pushed to the fork of the repo in my own github, right?

[rapid igloo]

that's fine as well, can you open the github repo for your fork then?

do you see something like a yellow box that prompts you to open a PR?

[drowsy ravine]

I've got a contribute button, it's just grey / github coloured.
If I click the little arrow it says

This branch is 2 commits ahead of python-discord/sir-lancebot:main. Open a pull request to contribute your changes upstream.

[rapid igloo]

yeppers that works too

now you can click on it and it should show you a preview of your changes

[drowsy ravine]

Able to merge. These branches can be automatically merged.

Create pull request?

[rapid igloo]

yep!

[drowsy ravine]

Im scared! 😛
Feel free to refuse if you dont want it hey

[rapid igloo]

put in some descriptive text on your PR and it shall be left to reviewers on what becomes of it :)

and you can always push new commits to the PR, by simply pushing to that same branch on your fork afterwards

[drowsy ravine]

Copied.
But don't I need to get like a branch or something? (or is that just for bigger new features?)

[rapid igloo]

it's not really necessary if you work on your fork, and only work on one PR at a time

by the way: in the future, you can just use a branch from the python-discord repo, since you're in the org

for this PR, since you've already made your commits from your fork, it's fine :)

[drowsy ravine]

Since I'm not too comfortable with these things yet I rather not until someone show me how and why etc ❤️

[rapid igloo]

no worries! you can continue using your fork for as long as you want to, to get used to the process

and btw I just found out that we do have a guide for PRs:
https://www.pythondiscord.com/pages/guides/pydis-guides/contributing/pull-requests/
so in case you need help with it in the feature, it's there :)

[drowsy ravine]

Thanks, i've saved that!

[drowsy ravine]

Where can I get the poetry files needed to get precommit?

[tame pebble]

You mean you want to know how to install precommit ?

[drowsy ravine]

I guess yes.

pre-commit 3.6.0 A framework for managing and maintaining multi-language pre-commit hooks.

[tame pebble]

yes, that's the package, that uses yaml to define the things to run in the hook

Then you need to "install" these rules

which puts them in the precommit hook

To do that, you need to run poetry run task precommit

That'll take care of installing the hook

which will then allow it to run before each commit

[drowsy ravine]

so I did that, and it cant find task

pCommand not found: task
korty.codes@kortys-macbook sir-lancebot % poetry run task lint
Command not found: task```

(nvm the p in front of the Command, I was too fast typing)

[tame pebble]

can you run poetry show taskipy ?

& tell me what it says

[drowsy ravine]

 name         : taskipy                          
 version      : 1.12.2                           
 description  : tasks runner for python projects 

dependencies
 - colorama >=0.4.4,<0.5.0
 - mslex >=1.1.0,<2.0.0
 - psutil >=5.7.2,<6.0.0
 - tomli >=2.0.1,<3.0.0```

I think I fixed all the issues ruff now gave me, but I rather want to do fix myself before it fails again.

[tame pebble]

Strange

What about poetry env list?

I mean, we can try to run the command directly: pre-commit install

But i'm not sure to why poetry isn't recognizing the command if it's installed

[drowsy ravine]

I have now pip install pre-commit
and pre-commit install to install the git command

korty.codes@kortys-macbook sir-lancebot % poetry env list sir-lancebot-k6fhlbY6-py3.11 (Activated)

[tame pebble]

pip install pre-commit
The installation is handled by poetry

which poetry version do you have ?

[drowsy ravine]

poetry --version
>>> Poetry (version 1.7.1)

[tame pebble]

Strange, I can't reproduce it

I've updated to the same version & made a new env

I'm not sure to what this could be related

Can you try making a new environment ?

[drowsy ravine]

If that's via poetry I have no clue how to.

[tame pebble]

poetry env list

copy the name of the one that has "activated" next to it

sir-lancebot-k6fhlbY6-py3.11 in your case

poetry env remove sir-lancebot-k6fhlbY6-py3.11

then poetry shell

& finally poetry install --no-root

[drowsy ravine]

Ok I see how that works.

Thanks for that run through.
Im not sure how it wasnt installed, but in any case I now know how to fix it ❤️

[tame pebble]

Did that fix it ?

[drowsy ravine]

It did! 🙂 Thanks

[tame pebble]

Ha, nice!

You're welcome

[drowsy ravine]

I was just saying I dont know why/how it wasn't installed in the first place

I'm also new to poetry so just a simple "contribute" turned into learning about bots, poetry, setting up the project etc etc

Feels like the classic high school problem:

In class:
John has five oranges, each orange cost 60 cents.
How much is the total costs of the oranges?

In test:
John has 3 apples and a pear.
Calculate the mass of the sun.

[tame pebble]

That's great ! 😄

[drowsy ravine]

And this poetry install --no-root installs from poetry.lock?
Can I see that the same as requirements.txt?

[tame pebble]

Not entirely, requirements.txt doesn't pin transitive dependencies

[drowsy ravine]

Okay I see.
So poetry will lock these dependency versions that those and only those get loaded (if still available of course)
Whereas requirements.txt will download the dependencies but the latest version?

[tame pebble]

At least, not the usual approach

The equivalent would be to do pip freeze > requirements.txt

But not all people do that

[drowsy ravine]

Oh, how else are you making requirements.txt? people typing manually?

Now that we're talking about the pip freeze as well.
If I were to free up disk storage on my computer, and have a few venvs that I don't actively use.
would pip freeze be a good idea, then delete the venv?
If I want to come back to the project I can just create new venv and install the requirements again?

[tame pebble]

Not entirely.

• With the .lock version, poetry will install exactly all the pinned dependencies, and the transitive ones as well.
• requirements.txt can have a listing of the main dependencies, and let pip resolve all the tranitive ones.

Example
package1 dependens on transitivePackage > v1.6
package2 dependens on transitivePackage > v1.2
HEre, pip will resolve into installing a version of transitivePackage that's striclty greater than 1.6.
It won't be always the latest as it depends on the python intepreter's version as well

However, with poetry, we also lock transitivePackage, so it'll only install that one

[drowsy ravine]

Gotcha. Thanks for explaining that.

[tame pebble]

Yes, that could be a case

[vale ibex]

pip freeze also locks transitive deps

jsut not with hashes

[tame pebble]

That's what I'm saying

What i'm saying is not all poeple use pip freeze

but with poetry, it locks them for you automatically

[vale ibex]

true

[outer oasis]

You can --with-hashes
(I forget the specific flag)

[vale ibex]

I didn't think pip ever added that?

I thought they were pointing people to pip tools or pipenv for that

[outer oasis]

Well now that I pull up the docs to reference, I can't find it

[tame pebble]

I think that's the installation one you're referring to

Not the freezing cmd

[vale ibex]

you can generate requirements.txt diles with hashes though

poetry export by default includes hashes

[outer oasis]

https://pip.pypa.io/en/stable/topics/secure-installs/
I see checking, but I don't see generation

[vale ibex]

Yea

[outer oasis]

I know I did.... I know I tried it once
Because it failed
But I... Oh.

[tame pebble]

I think pdm does it as well

you can instruct the format

[vale ibex]

yea pipenv and pip-tools can too

[tame pebble]

I never used those

But good to know

[outer oasis]

pip-compile --generate-hashes
Nevermind I'm dumb

[vale ibex]

.PHONY: lock
lock:
    poetry lock
    poetry export --only main --output requirements.txt
    pre-commit run --files pyproject.toml poetry.lock requirements.txt

this is why I love make files

lance does too

[drowsy ravine]

How do we type hint that something has to be a function ?

[vale ibex]

https://docs.python.org/3/library/typing.html#annotating-callable-objects

[drowsy ravine]

Callable[[str], str]
a function that takes in a string and returns a string?
Is that correct?

[vale ibex]

yup

Just keep in mind it's slightly different for async functions

[drowsy ravine]

the function thats type hinted is not async, but the function it’s going into is. Does it then need to be changed?

async def func(fn: Callable):

[vale ibex]

Nah

[drowsy ravine]

Great!
Is there a reload cog command?

[vale ibex]

.help cog

[dusky shore]

Query "cog" not found.

Did you mean:
c
cogs
colour

[vale ibex]

.help ext

[dusky shore]

Command Help

**```
.extensions

**Can also use:** `c`, `cogs`, `ext`, `exts`

*Load, unload, reload, and list loaded extensions.*

**Subcommands:**
**`extensions list`**
*Get a list of all extensions, including their loaded status.*
**`extensions load [extensions...]`**
*Load extensions given their fully qualified or unqualified names.*
**`extensions reload [extensions...]`**
*Reload extensions given their fully qualified or unqualified names.*
**`extensions unload [extensions...]`**
*Unload currently loaded extensions given their fully qualified or unqualified names.*

[vale ibex]

.src ext

[dusky shore]

Command: extensions

Load, unload, reload, and list loaded extensions.

Source Code

Go to GitHub

bot/exts/core/extensions.py:76

[vale ibex]

that one

it uses the filename that the cog is located in, unless the name kwarg is overwritten when creating the cog class

[drowsy ravine]

Good stuff! 🙂

[vale ibex]

!rule 9

[stable mountain]

Rules

9. Do not offer or ask for paid work of any kind.

[jovial barn]

I am now developing machine learning model from scratch that do interior design automatically.
What kind of models can I use?
How can I make an approach?
I need someone who can collaborate with me.

[vale ibex]

#data-science-and-ml is likely the channel you're looking for, this channel is for discussing internal Python Discord projects

[drowsy ravine]

@vocal wolf
I had a copy open and decided to just implement the suggestion you made to the pypi package regarding last nights comment. I added "to PyPI" to make it extra clear. Not sure if that's required but yeah.
This what you need?

[vocal wolf]

yuh

[drowsy ravine]

I think we should strip the time information though and just provide the date?

[vocal wolf]

yeah we just need date, don't need that much verboseness

[drowsy ravine]

exactly.
brb

[drowsy ravine]

@vocal wolf I think I've done this: Add more commits by pushing to the pypi-timestamp branch on Robin5605/pydis-bot. as suggested in the issue. (not too sure though)

[rapid igloo]

if you're PR-ing to robin's fork, might wanna PR to their pypi-timestamp branch instead (rather than main)

because their original PR uses that branch, so new commits to the branch could then be reflected in robin's PR

[drowsy ravine]

I did!

oh but the merge is into :main

hmm

[rapid igloo]

you can update the base branch by clicking Edit on the pull request.
although, if you're continuing the work in the original PR, maybe robin can just add you as a collaborator on their fork so you don't have to open the PR there (cc @sharp crag)

[drowsy ravine]

must i git push --set-upstream pypi-timestamp mybranch?

[rapid igloo]

no need to push to a different branch, the issue here is with your PR

[drowsy ravine]

Okay, I changed the base.

[rapid igloo]

okay great!

[drowsy ravine]

Thanks!
once again, sorry for all the questions on git! ❤️

[vocal wolf]

ah sorry for the late reply. glad things worked out! I hope to check on the PR tomorrow

[drowsy ravine]

No worries! Thanks for the patience ❤️

[sharp crag]

Let me know if you have any questions on this PR - I started it a while ago but I might've written something stupid as is par for the course :p thanks for taking over!

[drowsy ravine]

All I did was implement the suggestion on the issue. ❤️
tbh, It's an opportunity for me to get involved in open source 🙂
And I must admit, it really helps doing this actively rather than reading how to do it once then never do it and I forget haha.
I'll try to be better to PR it to the pydis GH next time

I like how it looks in any case.
Maybe i'll have a look and see if I can get these different python versions to work.
potentially with an extra command in the !pypi package --versions (or something)
Not that it needs to be implemented, but just to see if I am able to

[final geode]

Discord changed something with nicknames, this may affect .catify and other commands that affect nicknames

Story: someone on another server started complaining about discord changing name length. I'm bored so I started testing and it seems it's only for fancy Unicode stuff, so far I found that 4-byte Unicode takes 2 characters in nicknames now, Cyrillic (2 bytes per character) takes one character in nickname (ie doesn't affect max length).

Testing method : manual.
I padded my nickname on my test server with digits and just started deleting n last characters, pasting test characters, then tried to fill in the padding

E.g. Ascii only/ascii plus Cyrillic is 32 characters in nickname, but "𝔗𝔢𝔰𝔱" in the middle makes it 28.
Edit: 3-byte 'ࠀ' also takes 2 spaces wait, no, it doesn't, it's just rtl and ate a number I gave to it

[subtle kraken]

Why would that affect nickname commands?

[final geode]

.src catify because e.g. catify has a check for 26 characters because it adds 6 characters to the name

[dusky shore]

Command: catify

Convert the provided text into a cat themed sentence by interspercing cats throughout text.

Source Code

Go to GitHub

bot/exts/fun/catify.py:17

[final geode]

Wait, wtf... Bot changed my name successfully - exactly 32, but if I go to edit it, the cat's head doesn't exist. So the weird name limit change is only for users and bot can still do it the old way? O_o

Emoji in my name makes it so that if I changed my name manually, I can only fit 5 ascii characters
But bot can fit 6 characters in there

Also, I was wrong above, 3-byte characters take one space in the name not two (I edited my message above). The script used for the cat itself is therefore safe on its own, it's user's previous name that can be weird

[vapid vapor]

I will use that hahahaha

[tame pebble]

@drowsy ravine 👋
What I've meant to say is, the only change that was supposed to happen is

-def neutralise_string(txt: str) -> str:
+def neutralise_string(txt: str | None) -> str | None:
    """Attempts to neutralise all punctuation and cases and returns a string of lowercase words."""
    # take out punctuation
+   if not txt:
+     return None
    txt = re.sub(r"([^\w\s]|_)", " ", txt)

The clean_text_or_reply already takes care of raising the BadArgument error

[drowsy ravine]

👋
Oh I see. I’ll change it in a bit. Just having dinner quick.
Thanks so far! 🙏🏻

[tame pebble]

No worries !

[drowsy ravine]

Question about ruff.
`- hook id: ruff

• exit code: 1
• files were modified by this hook

Found 1 error (1 fixed, 0 remaining).`
Any way to see what exactly it's done?
(and a way to switch off the auto correct?? I rather manually type it after suggestions so I will actually learn)

[vale ibex]

Any way to see what exactly it's done?
pre commit hooks never stage their changes, so you compare the code in the staging area with the code that isn't staged

git diff --staged will do it in the command line, most IDEs have a way to do it too, but I'm only familiar with vscode

[drowsy ravine]

Okay, so I just check the two files and see whats changed. 👍🏻

[vale ibex]

Do you use vscode?

vscode shows it like this, where the staged changes are at the top, and then if pre-commit makes a change it appears in the changes section

you can click on the file in the changes to see just the diff from your staging area to what precommit did

[drowsy ravine]

Yeah I do. It opens the two files next to each other. Somehow I had struggles to spot the difference between an empty return and a return None 😅😅

[vale ibex]

ahh lol

fwiw, you can also run ruff . manually and it outputs the change

ruff . --fix is what auto fixes

make sure you're inside the poetry shell though

[drowsy ravine]

The type hinting + (i think its) pydantic really makes my code better. Seems a bit much work for quickly testing some things. But its good for me to get used to 🙂

[tame pebble]

@drowsy ravine I just noticed that you're developping straight on main in your fork

It's better to create a separate branch for each feature/bug fix

Since your main tracks the main branch from the pydis repo, when we merge your pr and then you'll sync, you'll run into issues

[tame pebble]

Chris, please be careful with logging :p

[drowsy ravine]

Yeah I'm still in the process of figuring out how to work with git. Let me see if i can fix it now

[tame pebble]

The main (as in the source of truth one, which is usually called main or master) should always have production ready code.
Whenever you want to add something new/fix something, you almost always would want to make a separate branch for it
Then you make your changes, get it reviewed, and reiterare until its approved, then it gets merged into main.

What you'd want to do is

1. Make a new branch out of your main
2. Hard Reset your main onto pydis'
3. Change the Pull request's source branch to the one you created in step 1

[drowsy ravine]

Okay, so with my current code I did git checkout -b fun-cases
now im going to do git push -u origin fun-cases, correct? (this creates the branch on the remote, which should be my forked pydis repo in my account, right?)

[tame pebble]

you can configure get to always push up to the remote branch with the same name as the local one automatically

[tame pebble]

Indeed

But you need to reset your fork's main as well

that would require adding a new remote

[drowsy ravine]

So how do I reset the main?

[tame pebble]

What's the output of git remote -v?

[drowsy ravine]

origin https://github.com/JelmerKorten/sir-lancebot.git (fetch) origin https://github.com/JelmerKorten/sir-lancebot.git (push)

[tame pebble]

you have 2 options, you can either add pydis repo as repo as well, then reset on it
or, you can just locate the last commit hash before you started committing on main, and you git reset --hard {{hash}}

[drowsy ravine]

before you started committing on main
The very first one? L:P

[tame pebble]

the one just before your very first commit

so the one before 90625a1d323e052d8509f55fa49958a2b88bbd8e

once you do that (while you're still checkout out at main), you need to force push

[drowsy ravine]

I hear what you're saying. But I'm not sure what to do with the information.

I can click the sync fork button on github.com , it says this will discard the 11 commits, but I've made that branch now, fun-cases so those same commits are still there? I'd prefer to do it in CLI just to get more experience with that.

[tame pebble]

do what in cli exactly ?

so those same commits are still there?

yes

You also pushed them, so they'll be in your fork as well

[drowsy ravine]

reset main to up-to-date pydis. then if I want to work on something new, I can create a branch for myself?

[tame pebble]

yes, you pull to get latest changes, then branch out

[drowsy ravine]

Then how do I change the PR source?

[drowsy ravine]

Okay, I think im synced, took me a bit to figure out. Now sleeps!
nn ❤️

[tame pebble]

Nice !

[tame pebble]

👋
This is done
Cc @rapid swallow

[rapid swallow]

Thanks

I'll try to check it out tonight

Or tomorrow

[tame pebble]

[tame pebble]

@vale ibex Re bot-core#48, using a FQN for each type seems to do the job.

[dusky shore]

GitHub

[bot-core] #48 Document BotBase instance attributes

[tame pebble]

However, the deploy preview is off, it still shows 1.7.0, and the new attributes aren't there

But the locally built docs are just fine

Forgot I didn't link the PR 😅
bot-core#204

[dusky shore]

GitHub

[bot-core] #204 Document the instance attributes of BotBase

[vale ibex]

Ah nice

will give it a look tomorrow

[tame pebble]

Sure !

[drowsy ravine]

@tame pebble

#dev-log message
Does that mean I did a contribute? 🤩🥹

[tame pebble]

It means you're about to!

You still need other approvals

[tame pebble]

@vale ibex, the new line is too long, do you want me to noqa it or just break it up ?

IT's just that I don't like it like this, and I didn't want to break a sentence before it was finished :p

I mean, like this

[drowsy ravine]

ooooh, so close! 😛

Do I just create a new PR? (I can't do anything to that current PR right?)

[vale ibex]

Yea just break it up, I think elsewhere we only tab in the second line by 4 spaces, rather than all the way like that though

[tame pebble]

You can commit to the same remote branch if you'd like, it's up to you

[vale ibex]

Committing to the same branch will update the PR, I'd suggest you make a new branch for each new feature you work on

[tame pebble]

Done

[drowsy ravine]

I did that! I have a brand spanking new branch after yesterday's finding out 🙂

[vale ibex]

Also, I don't think we need a version bump for this, since it's just docs changes

you can leave the line in the changelog, but remove the version specifier

it just works

[tame pebble]

+ :support:`204` Document the instance attributes of :obj:`pydis_core.BotBase`.

So just like this ?

[vale ibex]

Yea

just delete line 7 basically

(and update pyproject.toml)

[tame pebble]

Yes yes, that part is done

[vale ibex]

Nice

[tame pebble]

Alright, pushed

[vale ibex]

we're great

[tame pebble]

Oooo

Yes we are

I didn't know it worked like that

[drowsy ravine]

Hey!
I have question! (Again, lol)
Had some struggles with my own tiny github project. I'm still trying to get used to the whole github process.
So what I (thought I) did was:

• Create a repo.
• Add some stuff to it and push to remote, say main.
• Then I branched off locally git checkout -b newbranch
• Did some newbranch stuff locally and eventually pushed to remote git push -u origin my-branch-name
• Decided that it's implemented enough and wanted to merge. This is where my confusion came in.
• I went to GH and it popped up: newbranch has recent commits Compare
• Inside the compare screen, it said, there's nothing to compare, these two branches have no commit history in common (or something pretty close along those lines)

What did I do wrong?
And how to prevent it next time?
(I somehow merged it, but surely not the way I was supposed to)

[tame pebble]

Can you reproduce this on a public repo ?

Was this repo a new one ?

Maybe it's better to discuss this in a help post

[drowsy ravine]

Fair point, I've opened a channel

[vale ibex]

fakeredis usage

[drowsy ravine]

@rapid igloo thanks for your feedback on the repo. I will 100% have a look at it and try to implement. I am however not available until next week unfortunately, working away from base. That okay?

[tame pebble]

We don't really have time limits on when something should be implemented. unless it's absolutely critical

So don't worry about it

[thorny obsidian]

@vale ibex how would I go about testing the postgres logic? Can I trust that you tested it and it works? >_>

[vale ibex]

I enabled postgres query logging to get the actual query that was ran, and then played with the logic directly on the cluster

The queries I used are in the pr to show the logic is right

[crystal sonnet]

How can I do this when I click a button because a Select should come up? How does that work? (in py-cord)?

[vale ibex]

I think you're looking for #1035199133436354600 or #discord-bots, we don't use pycord in our projects, which this channel is for

[waxen jasper]

https://github.com/python-discord/sir-lancebot/pull/1460 🙏

[rapid igloo]

hey there, I think we'd much prefer if each PR only included relevant commits. it looks like you didn't sync your PR's base branch from upstream (python-discord:main) before adding your christmasify commit

ideally the PR should only include c4c45a0 and perhaps a merge commit before that to update the branch

(and in the future, try using feature branches instead of main so you can work on each of your PRs separately and sync with upstream independently of each other)

[tame pebble]

Like hedy just said, it's better if you clean up your branch history.
I've discussed the same topic a few days ago here: #dev-contrib message
So you might wanna do the same

[waxen jasper]

oh, i didnt realise everything came under it, im pretty new to using github so didnt rlly understand. So rn seems like i have to first sync the main branch then push the new code and then make a new pull request?

[waxen jasper]

aight! will take a look

[rapid igloo]

yep! technically you can still reuse the same PR by force-pushing after you've fixed your branch history. but feel free to use a new pull request if it's easier

[waxen jasper]

okay so right now

i move the new stuff to a copy of the repo or rather a different folder for now

then hard reset right?

[tame pebble]

not really

create a new branch from your main

it will have the relevant commit

recheckout to main

sync your fork, then hard reset your main on the remote one

checkout to a new feature branch

and cherry pick your commit

i think you only have one commit

[waxen jasper]

sorry, what does recheckout mean?

[tame pebble]

checkout back to main I mean

[waxen jasper]

still confused 😅

just go back to the main branch?

[tame pebble]

yeah

switching branches is called checking out

[waxen jasper]

oooo good to know

[rapid igloo]

(because you use the git checkout command)

[waxen jasper]

ah, makes more sense

how do i figure out the hash

[tame pebble]

It's the one hedy sent you

You can identify the hash based on your commit message

[waxen jasper]

c4c45a0? i thought that was the commit code

[tame pebble]

git log --oneline will list the commits for you

There's no such thing as a commit code, at least not to my knowledge

It's the hash

Just not the entire hash

Since usually the first 8 characters are enough

[rapid igloo]

you can understand the hash as the "ID" for each commit

[waxen jasper]

now its back to my commit :D

no wait

i think i have to go to a previous commit

right?

ive contributed before

[rapid igloo]

you can use HEAD^ to refer to the previous commit

[waxen jasper]

how would i do that?

[tame pebble]

what do you mean ?

What have you done so far ?

[waxen jasper]

first i checked out to a new branch with the old commits

checked out back to main

hard reset up until c4c45a0 (git reset --hard c4c45a0)

and then checked out to a new branch

[tame pebble]

did you sync your fork first ?

and you should hard reset to main's head

git reset --hard origin/main

[waxen jasper]

yes

[tame pebble]

then, make a new branch from the clean main git checkout -b christmasify-avatar

then, cherry pick the relavant commits git cherry-pick hash1 hash2

[waxen jasper]

i ran this command but this is still showing up

and my modifications havent been removed

[tame pebble]

ok, can you run git remote -v

[waxen jasper]

origin  https://github.com/SockYeh/sir-lancebot.git (push)```

[tame pebble]

git remote add pydis https://github.com/python-discord/sir-robin.git

git reset --hard pydis/main

when you're on main

git push -f

[tame pebble]

then, do this

[waxen jasper]

sir-robin?

[tame pebble]

sir-lancebot

xD

[waxen jasper]

git remote add pydis https://github.com/python-discord/sir-lancebot.git

right?

[tame pebble]

i got lost for a second

yes

[waxen jasper]

PS D:\Python\sir-lancebot> git remote add pydis  https://github.com/python-discord/sir-lancebot.git
PS D:\Python\sir-lancebot> git reset --hard pydis/main
fatal: ambiguous argument 'pydis/main': unknown revision or path not in the working tree.
Use '--' to separate paths from revisions, like this:
'git <command> [<revision>...] -- [<file>...]'```

[tame pebble]

try git reset --hard pydis main

[waxen jasper]

fatal: ambiguous argument 'pydis': unknown revision or path not in the working tree.
Use '--' to separate paths from revisions, like this:
'git <command> [<revision>...] -- [<file>...]'```

[tame pebble]

what does git remote -v say ?

[waxen jasper]

origin  https://github.com/SockYeh/sir-lancebot.git (push)
pydis   https://github.com/python-discord/sir-lancebot.git (fetch)
pydis   https://github.com/python-discord/sir-lancebot.git (push)```

👍(•_•)👍

[tame pebble]

Ah

We need to fetch first

git fetch pydis

[waxen jasper]

oh

i really do need to learn git

[tame pebble]

because there are no refs stored

then reset

[waxen jasper]

ive been relying solely on vsc's source control

Use '--' to separate paths from revisions, like this:
'git <command> [<revision>...] -- [<file>...]'```

still the same issue

even after fetch

PS D:\Python\sir-lancebot> git fetch pydis
remote: Enumerating objects: 192, done.
remote: Counting objects: 100% (179/179), done.
remote: Compressing objects: 100% (59/59), done.
remote: Total 192 (delta 135), reused 154 (delta 120), pack-reused 13
Receiving objects: 100% (192/192), 102.71 KiB | 6.04 MiB/s, done.
Resolving deltas: 100% (137/137), completed with 31 local objects.
From https://github.com/python-discord/sir-lancebot
 * [new branch]        dependabot/pip/pre-commit-3.6.2 -> pydis/dependabot/pip/pre-commit-3.6.2
 * [new branch]        dependabot/pip/pydantic-settings-2.2.1 -> pydis/dependabot/pip/pydantic-settings-2.2.1
 * [new branch]        dependabot/pip/ruff-0.2.2 -> pydis/dependabot/pip/ruff-0.2.2
 * [new branch]        dependabot/pip/sentry-sdk-1.40.5 -> pydis/dependabot/pip/sentry-sdk-1.40.5
 * [new branch]        main                  -> pydis/main
 * [new branch]        reorder-hacktoberfest-commands -> pydis/reorder-hacktoberfest-commands
 * [new branch]        revert-1432-dependabot/pip/emoji-2.10.0 -> pydis/revert-1432-dependabot/pip/emoji-2.10.0
 * [new branch]        shenanigansd-patch-1  -> pydis/shenanigansd-patch-1
 * [new branch]        sim_matches           -> pydis/sim_matches
 * [new branch]        use-botcore-command-error-manager -> pydis/use-botcore-command-error-manager
 * [new branch]        use-botcore-paginator -> pydis/use-botcore-paginator```

[tame pebble]

Did you do it like this ?

[waxen jasper]

oh

[tame pebble]

this one's on me :p

[waxen jasper]

yup its up to date now

[tame pebble]

now force push

git push -f

[waxen jasper]

yayy

so now i compare and pull?

[tame pebble]

Indeed

[tame pebble]

Nice

[waxen jasper]

still shows all tho

oh wait i didnt cherry pick

[tame pebble]

on the new branch ?

[waxen jasper]

yea forgot to do that lol

okay now thats done

i push?

no wait i merge

since different branches

[tame pebble]

Wait

You don't want to merge

you want a branch, with just one commit

which branch are you on ? is it a newly created one ?

from the fresh main

[waxen jasper]

yup

[tame pebble]

How come you have all that then ?

[waxen jasper]

i cherry picked the commit too

[waxen jasper]

oh wait this is the wrong one

the new one doesnt even show up

[tame pebble]

did you push it ?

[waxen jasper]

nice

thanks a lot!

[tame pebble]

Sure thing!

[waxen jasper]

okay so now

i leave the branches be?

[tame pebble]

which branches?

[waxen jasper]

lets say the current one

once its merged

do i delete it or nothing like that

[tame pebble]

yeah, it's preferable to delete branches

[waxen jasper]

so everytime i have to add a new feature, i do all this?

[tame pebble]

No

you just need to

1. sync fork
2. Pull main into your local branch
3. Make a new feature branch
4. Develop, commit, then push

We did alot of other steps to reset your main

as you shouldn't work on it directly

[waxen jasper]

nice, i was worried i would have to go through all this every time

also, any recommendations on youtube videos for learning git

[tame pebble]

I like this one https://learngitbranching.js.org/

Also, atlassian have great tutorials

Check them out

( I don't like watching videos to learn )

[waxen jasper]

this looks sick (from a website standpoint)

[tame pebble]

It's quite nice indeed

[waxen jasper]

will go through with it

[waxen jasper]

interesting

yea sometimes videos become too much

[rapid igloo]

what happened with your first chrismasify PR was that, because you worked off main for the riddle command, and without syncing to upstream (pydis remote), you started to work on the chrismasify PR by adding new commits afterwards. this makes your main diverge off of the upstream pydis main branch.

so when you open your first PR, github tries to find the commits from your main that are different/newly-updated from pydis, which includes commits from the riddle PR. what we want instead is only to include the chrismasify commit.

how Flying Whale helped you fix this issue was essentially:

1. save your chrismasify commit elsewhere
2. reset your main to the last commit that was still in sync with pydis
3. sync your main with pydis by pulling in new changes from upstream
4. restore your chrismasify commit

all of this can be avoided if you'd worked on PRs with feature branches, rather than using main, which would skip the steps 1 and 2 from above.

hope this makes sense, if it helps!
(and also correctly me if I'm wrong Bella)

[waxen jasper]

Thanks! Makes complete sense now as to where i went wrong, branches == gud

[tame pebble]

Correct, you should almost never commit to main directly

You should always branch out, develop, then merge changes into main

[tame pebble]

I say almost never because hot fixes can be pushed directly in case of emergencies, etc.

[rapid igloo]

hi @timid sentinel, is there a consensus (perhaps among core devs?) for what we should do for site#1232?

[dusky shore]

GitHub

[site] #1232 Switch to Font Awesome django plugin

[drowsy ravine]

.bm

[rapid swallow]

Hey @quick nymph ! I've removed your post since it is off-topic for this channel.

!ot

[stable mountain]

Off-topic channel

#ot2-never-nester’s-nightmare

Please read our off-topic etiquette before participating in conversations.

[quick nymph]

😢

[timid sentinel]

Added a comment

[rapid igloo]

is it possible to have netlify retry the static preview without pushing another commit?

pr: site#1232
https://app.netlify.com/sites/pydis-static/deploys/65d87c92d20f310008ce0cfd

the gh workflow that runs the build failed at first but worked after I re-ran it.
if anyone with the right perms can trigger such a re-run on netlify's side, (or any other solution) would be appreciated. thanks!

[dusky shore]

GitHub

[site] #1232 Move from Font Awesome Kit to serving our own FA static files

[vale ibex]

Just kicked off another static build

[clever wraith]

Does this server pay you for the emotes you make for it?

[tame pebble]

@timid sentinel I'll be away for quite some time and have no pc access.

I'll just recreate the PR and abandon the changes for now.

To not block the bump

But I just thought it made sense to combine both those changes specifically

[wise marten]

Python eval'd my message twice
#1211041719743676496 message

[subtle kraken]

The bot just restarted so presumably briefly there were 2 instances of it running at once

[wise marten]

I C

that's probably why I can't delete some other eval output, with interactions failling

[subtle kraken]

Yeah the old bot died so its interaction presumably died with it 😔

[tame pebble]

That's because it was deploying a new version, so the old instance wasn't killed.
Overlap happens sometimes

[wise marten]

RIP

[tame pebble]

Oh, i didn't see sen's response. Discord is acting weird lately

[subtle kraken]

Discord wants me gone
Gotcha

Will keep that in mind

[tame pebble]

It's been happening quite a few times lately, where a lot of messages are not loaded

[drowsy ravine]

Regarding !pypi:
Referring to pygame, but I think this issue will be there for any package.

TL;DR:
pypi site is adjusted for local timezone settings and the API is just in Zulu/UTC, hence sometimes there will be a day difference when comparing (upcoming) !pypi output to the website. What do you want to do?

2023-09-17T22:27:49
Is listed as the source distribution upload time, but since I'm +4 UTC atm, it brings it to 2am on the next day (18th) on the pypi site.
But the api upload_time.date() will still return the 17th. Then when I go check the site, there will be 1 day difference. Which is confusing and will break my OCD.
the upload_time is in Zulu / UTC time, and pypi site reflects adjusted for your timezone settings.

[rapid igloo]

I wonder if using a discord timestamp would be an overkill? it could then use the user's local timezone

[drowsy ravine]

To be honest, if that's an option I would go for that.
Seeing 2 different dates is even more confusing than not seeing a date at all.

The problem still arises:
user's local timezone?
What determines what the local timezone is? Because if you use the command, I also get to see the output? Unless that would somehow be different for everyone seeing it with your suggested method?

[rapid igloo]

yes, discord shows the date depending on the user's timezone

you can try it with the .epoch command iirc

.epoch in 2 hours

[dusky shore]

Date and time parsed as: Sun, 25 Feb 2024 14:27:15 +0000

<t:1708871235:t>

[rapid igloo]

<t:1708871235:t> (this is an example that shows the time of course)

[vale ibex]

You can use this util to get a time stamp easily https://github.com/python-discord/bot/blob/main/bot/utils/time.py#L77-L82

[stable mountain]

bot/utils/time.py lines 77 to 82

def discord_timestamp(timestamp: Timestamp, format: TimestampFormats = TimestampFormats.DATE_TIME) -> str:
    """
    Format a timestamp as a Discord-flavored Markdown timestamp.

    `​timestamp`​ can be any type supported by the single-arg `​arrow.get()`​, except for a `​tzinfo`​.
    """```

[vale ibex]

The different formats available are in this Enum https://github.com/python-discord/bot/blob/main/bot/utils/time.py#L41-L46

[stable mountain]

bot/utils/time.py lines 41 to 46

class TimestampFormats(Enum):
    """
    Represents the different formats possible for Discord timestamps.

    Examples are given in epoch time.
    """```

[drowsy ravine]

(Im reading up on this! gimmie a second to wrap my head around it)

Im getting this now xD

Released: <t:1694989669:f>

Which works when I take it out of the embed as we just seen.

Released: <t:1694989669:f>

[vale ibex]

timestamps don't work in embed footers, you'll need ot add a field

[drowsy ravine]

Rip, back to a field then xD

!pypi pygame

[stable mountain]

pygame v2.5.2

Python Game Development

[drowsy ravine]

oh, sorry xD

What do you think?

[rapid igloo]

how's this instead? it looks similar to what we previously had as a footer except it's a different font size (note that the date format depends on locale)

[drowsy ravine]

The only way I know how to do that is by adding stuff to the description:
embed.description += f"\n\nMessage{time}"
If that's good enough..?

[rapid igloo]

yeah, that's how I did it

but I'd also see what others think about it

[drowsy ravine]

Ok, let me push it and await feedback

[clever wraith]

this looks better

[drowsy ravine]

I'm trying to understand how the bots deal with doing commands on a replied text. instead of providing the text to the argument.

[subtle kraken]

It checks whether there was any text supplied, if there was it uses the text to execute the command, if no text was supplied it checks if message invoking the command is a reply to other message and uses that instead
https://github.com/python-discord/sir-lancebot/blob/2e9c12f4f9ebe22890919e0febbccc946c386685/bot/exts/fun/uwu.py#L167-L176 is an example of that

[stable mountain]

bot/exts/fun/uwu.py lines 167 to 176

text = text or getattr(ctx.message.reference, "resolved", None)
if isinstance(text, discord.Message):
    embeds = text.embeds
    text = text.content
else:
    embeds = None

if text is None:
    # If we weren't able to get the content of a replied message
    raise commands.UserInputError("Your message must have content or you must reply to a message.")```

[drowsy ravine]

text = text or getattr(ctx.message.reference, "resolved", None)
does this mean that if text is None the statement right of the or will be evaluated
and we then assign that to text instead? Or still None if no reference was made (im assuming the reference here is the reply..?)

[subtle kraken]

or would assign first "truthy" value or last argument (right side of or iirc)

[drowsy ravine]

Yes. I think that's what I tried to say xD

[subtle kraken]

Anything more that needs explaining?

[drowsy ravine]

No that's all for now, this makes sense 🙂

Thank you.

[cold island]

@vale ibex What happened in https://github.com/python-discord/bot/pull/2913 ? 🤔

[vale ibex]

merged it locally and pushed, but it didn't auto-close the PR due to a rebase

[cold island]

so basically, rebase bad

[vale ibex]

i was very useful because I wanted to merge 3 PRs at the same time without merge commits :P

[cold island]

So can this be closed? https://github.com/python-discord/bot/issues/2912

[vale ibex]

maybe

[cold island]

And this https://github.com/python-discord/bot/issues/2910

[vale ibex]

I wonder why that one wasn't closed automatically

the PR merged fine

Ah, that's why https://docs.github.com/en/issues/tracking-your-work-with-issues/linking-a-pull-request-to-an-issue#linking-a-pull-request-to-an-issue-using-a-keyword

[cold island]

Wasn't linked?

[vale ibex]

The keyword I as thinking of was resolves, not solves

[cold island]

ah

@timid sentinel you approved but it had a few commits since https://github.com/python-discord/bot/pull/2924

[clever wraith]

Omg Zig

Hello

[drowsy ravine]

Any feedback on my commits are also welcome. 🙏🏻
When I was learning about git I learned they should be small? Open to suggestions in any case! 🙂 thanks

[outer oasis]

Why are you adding your initials to them?
They already have your name on them.

[royal prawn]

maybe korty is a group account🤔. how many are you korty?

[rapid igloo]

by the way korty, regarding the case commands PR, I noticed you pushed a few commits that addressed my previous comments. If you're ready, please do re-request my review so I know I can review again.

[drowsy ravine]

Not sure! Saw it somewhere and decided I liked the looks of it and have been doing it since. Can stop it if thats preferred?

[drowsy ravine]

Its just me 😅

[royal prawn]

yeah, sure, whatever you say, swarm account

[drowsy ravine]

Copy! I think it’s done as for now, will request later today.

And regarding pypi:
I did some tests earlier on a few packages and all [-1] dists were the “source” files. Then if I understand correctly sometimes other distrubution wheels are added later or updated (im assuming hot fixes) and this can lead to a later date. The “front page” for each package seems to reflect the source file upload time.

[dim pelican]

Hello, your message has been removed due to violating rule 6

[spice blaze]

oh, sry

[tame pebble]

Also, comments are for discussing things related with a pull request / issue.

So please refrain from making comments such as 'hi bro' in a pull request

[drowsy ravine]

Hi tho 👋🏻

[brazen wren]

Also, comments are for discussing things related with a pull request/issue. So please refrain from making comments such as 'hi bro' in a pull request,

Not sure saw it somewhere decided i liked the looks of it and have been doing it since.can stop it if thats preferred?

Does this mean that if text is none the statement right of the or will be evaluated and we then assign that to text evaluated

[drowsy ravine]

?

[wise marten]

!dasm .../!da ... command (dis-assemble) for running !e import dis; dis.dis(...). Feel free to steal the idea.

along the same lines as having a timeit command

only reason it wouldn't be !dis ... is because !dis is the same as !hosting currently

[still comet]

I believe that is because the full tag name for !hosting is !discord-bot-hosting and !dis gets coerced into that

I'm in favor of using !dis anyway because it isn't clear that it leads to !hosting right now

[timid sentinel]

That idea is covered under this issue:

bot#1880

[dusky shore]

GitHub

[bot] #1880 More snekbox commands (black/mypy/flake8 etc)

[clever wraith]

what version of pinnwand are you using in production cause the new api https://pinnwand.readthedocs.io/en/latest/usage.html#api-v1-paste-a-z2-7
does not seem to work

[tame pebble]

V1.5.0-psql

What's not working?

[clever wraith]

ah cause the latest release is almost 1 year old

[clever wraith]

https://paste.pythondiscord.com/api/v1/paste/YOFA

this for example

this api route was introduced in supakeen/pinnwand#194

[dusky shore]

GitHub

[pinnwand] #194 add a 'show' route for the v1 API

[clever wraith]

I cant really work on bot#2679 if its not in production yet

[dusky shore]

GitHub

[bot] #2679 Code snippets for paste service links

[vale ibex]

Yea that's fine, just need to wait for a release of pinnwand to be made

might be worth asking in their freenode irc channel if there's anything else that they're waiting on before making a release

https://github.com/supakeen/pinnwand?tab=readme-ov-file#reporting-bugs

[rapid igloo]

hmm there's no one in #pinnwand on both freenode and libera

[tame pebble]

Ahaha yeah i tried myself

The versioning in pinnwand is not consistent

I'll open an issue for that

[vale ibex]

how so?

[tame pebble]

What i mean is, if you take a look at the docker images, you'll only find latest, and the 1.5 one.

There's no unique tag of a particular release that'll allow you to pinpoint a version.

[vale ibex]

pushing to ghcr was only added in v1.5.0

There are github releases for versions before that

[tame pebble]

For example, in our case, we use the sha hash to identify versions

[vale ibex]

do we?

I thought I pinned it to 1.5.0

[tame pebble]

I mean, for our packages

[tame pebble]

Yes you did

But pinnwand's ci has triggered a lot of times in the past period, and it's only overwriting 'latest'

[vale ibex]

ah right yea, which is why we use the release instead

[tame pebble]

So if i merge a PR now, the previous version's identifier/tag will disappear

[vale ibex]

this is quite common for apps that use releases, rather than building & pushing on each commit

usually it wouldn't be a problem if release frequency was higher

[tame pebble]

But pinnwand does build and push on each commit

[vale ibex]

as in it doesn't tag each commit because you're either on an unstable tag (latest) or a stable tag (release version)

I'm talking quite generally here for apps that follow this approach

adding a SHA tag would likely be easy to do with CI how it currently is, it has the possibility of burying the version tag though

[tame pebble]

Yeah

I'll try to see what Simon thinks about it

[vale ibex]

he may also reply in #pinnwand if you ping him

the diff would be adding type=sha to the list of tags here and type=sha,suffix=${{ matrix.container-stage }} here

[tame pebble]

Man why did i never see this

[rough shell]

Someone want to collaborate with me in a python project? Im making a encrypted password manager.

Can I send the link here?

[vale ibex]

If you're looking for help, #1035199133436354600 is the best place to go, this channel is for discussion python discord projects.

Keep in mind our rule about advertising and recruitment though.

[rough shell]

Ok thanks

[rough shell]

I havent found that, could you show me?

[dim pelican]

!rule 6

[stable mountain]

Rules

6. Do not post unapproved advertising.

[dim pelican]

Also, all of the #rules

[clever wraith]

No Access

[full fractal]

thoughts on site#1235?

[dusky shore]

GitHub

[site] #1235 Refactoring the timeline: YAML + Markdown entries and a slimmer CSS

[vale ibex]

yes, I like it

removing code (css) is great

and reducing the effort to add new items to the timeline is a big plus

[full fractal]

pls approve then :P

cc @rapid igloo

[vale ibex]

I have an idea, gimme like 10 mins

Is this something you wanted to implement youself hedy?

or just something you wanted to raise

[rapid igloo]

if it's approved I'd be interested in having a go at it

but I might not have enough time so if I realize that's the case I'll drop a comment there saying so...

thanks @full fractal by the way!

[rapid igloo]

hello, I have two fairly trivial PRs here from the site repo which needs one contrib/staff approval

1. site#1226
2. site#1231

and also site#1210, though it's not as trivial
reviews would be appreciated!

[dusky shore]

GitHub

[site] #1226 Document how to add new resources
[site] #1231 Add docs for development on Bulma, CSS, and dark mode colors
[site] #1210 Redesign of events page

[rapid igloo]

thank you Chris

[vale ibex]

No problem, there's some really nice stuff in that redesign that I'm looking at now

what might go a long way is a basic readme in the events directory on how to add a new event

which mentions the need to update when the event ends

[rapid igloo]

I have a draft of it in my git stash ready! will go about fleshing it out in the coming weeks

[vale ibex]

Ah very nice, good to know

I'm sure @thorny obsidian will appreciate being able to tell someone "go read this" and it just works for adding events to site :P

[clever wraith]

When did you become project leads?

[full fractal]

~about a day ago as she's been assigned to modernise our timeline impl.

[solemn vine]

!source

[stable mountain]

Bot's GitHub Repository

Repository

Go to GitHub

[rapid igloo]

should bot#1603 be closed since the functionality is now native in discord?

[dusky shore]

GitHub

[bot] #1603 Small Event Management System

[rapid igloo]

note, @thorny obsidian mentioned it could be (in a staff channel) but since that was three years ago, I'm bringing it up just in case.

[wise marten]

Looks like @stable mountain doesn't know what to do with a help post with no messages in it for auto-closing. Someone mentioned that a completely empty post has been open for about 9 hours. https://discord.com/channels/267624335836053506/1218220149283360808

[outer oasis]

Interesting
I don’t think I’ve ever seen one without the embed before

[vale ibex]

The bot can normally handle that just fine, there must have been a discord API blip, or the bot was down, when that channel was open so it didn't trigger the bot picking it up as something it needs to schedule for closing

[wise marten]

I c, freak accident