dev-contrib

[dusky shore]

GitHub

Update Github emoji to new icons

[vocal wolf]

there we go

[patent pivot]

✅

[velvet sequoia]

Thank you!

[short snow]

Can someone confirm this, https://github.com/python-discord/bot/pull/1634#pullrequestreview-682324315

[exotic ember]

whoops, I added that after you asked

then kinda forgot about it

[short snow]

@brazen charm I didn't understand what you meant by this, https://github.com/python-discord/bot/pull/1634#discussion_r650391876

[fallen patrol]

ah

this part i presume

Would need confirmation as I'm not sure if I got it right myself but I don't think the 16 hour limit was meant in this way, but the other way around (as a maximum). With a 16 hour minimum the moderator will always have less than 8 hours for personal non-ping time.

[trim cradle]

thanks

[cold island]

Yeah the 16 hour limit is a maximum

[clever wraith]

for the helpdm command can the dm be deleted once the channel is closed? the command is close to having no benefit

[short snow]

ah thanks for the clarification!

[short snow]

This is my code: https://github.com/python-discord/bot/blob/bb52bdf38e053935995ad4e9bc04f36831f93fbb/bot/exts/moderation/modpings.py#L188..L216 Whenever i do !modpings schedule 2h22m 2h23m i.e. it should schedule the task at 2:22 UTC timing but for some reason. It schedules the task at that exact minute when I invoked it

invoking time:

wait, does that mean past?

like it the time was in past, so it is getting scheduled immediately

[short snow]

Yeah, that was the issue, fixed it and pushed

[brazen charm]

The create_task util does the exception logging for you now. I noticed three errors when reading through it, those should now be logged as error level logs when the method gets ran as the bot starts up

[short snow]

Ah got it 👍

I have mentioned your changes btw

[vale ibex]

Ahh cool, we can free up some emoji slots on emoji server 2 now

We're actually at the limit atm

Can do it when I'm home

[green oriole]

lol, I guess it is time for server 3

[vale ibex]

Lol Yea

Realised deleting emoji was possible in mobile, so I've just done that

We've got 4 emoji slots left on there

[green oriole]

We probably have a few that aren't used anymore

[vale ibex]

Yea probably

Definitely on the first emoji server

There's a bunch of egg emoji

[green oriole]

I think we still have the egg chase cog in sir lance

[vale ibex]

Don't think I was around for that event

[green oriole]

It is very old

Maybe two years or so

I believe the last 8 or so emojis in the first server can go away

[torpid grotto]

Seems like @stable mountain is triggering typing for quite a while after the docs command was invoked. this might be a discord bug though

not sure if this was the right channel to send this to so lmk

[vale ibex]

Do you have example input that causes it?

Yea, this is the right channel for discussing it

[torpid grotto]

Just any documentation lookup it seems

I've only seen this in the discord.py channel

About every time it was used now

[vale ibex]

So the typing event stays even after the doc is sent?

[brazen charm]

!d str

[stable mountain]

str

class str(object='')``````py

class str(object=b'', encoding='utf-8', errors='strict')```
Return a [string](https://docs.python.org/3/library/stdtypes.html#textseq) version of *object*. If *object* is not provided, returns the empty string. Otherwise, the behavior of `str()` depends on whether *encoding* or *errors* is given, as follows.

If neither *encoding* nor *errors* is given, `str(object)` returns [`object.__str__()`](https://docs.python.org/3/reference/datamodel.html#object.__str__ "object.__str__"), which is the “informal” or nicely printable string representation of *object*. For string objects, this is the string itself. If *object* does not have a [`__str__()`](https://docs.python.org/3/reference/datamodel.html#object.__str__ "object.__str__") method, then [`str()`](https://docs.python.org/3/library/stdtypes.html#str "str") falls back to returning [`repr(object)`](https://docs.python.org/3/library/functions.html#repr "repr").

[torpid grotto]

Yes it hangs for a good few seconds I'd say around 5?

[tawdry vapor]

Isn't it a discord limitation

[brazen charm]

Seems alright to me, but there could be some client side issues

[torpid grotto]

It seems to be happening when more then one people do it at the same time though

[tawdry vapor]

That it has a minimum time it lingers for

[torpid grotto]

I didn't keep a close eye

[torpid grotto]

Feel like it is, remember reading about that

[vale ibex]

Yea, the typing events in discord don't have an end event, you send the typing event and then it ends when you send your message

Or it times out

[brazen charm]

I guess we could remove the typing as most fetches will be almost instant

[torpid grotto]

Ah yep this is a discord limitation seems like

[torpid grotto]

Agree with this, the documentation look up is quite fast

I don't think you need to trigger typing after every fetch or whatever it is in this bot

I haven't read the source code of this bot so I'm unsure

[vale ibex]

There's a context manager in d.py,which you just wrap the command in

I pretty certain that's what we're using

[torpid grotto]

Ah sorry, I haven't read through this bot's source code much

I see the context managers now

[green oriole]

The typing event either stops after you send a message or a 5s timeout

But the context manager isn't really a reliable method and may retrigger a typing just after the message is sent, meaning that will have to wait those 5s

[sleek steppe]

Doesn't it send the message outside the context manager?

[green oriole]

That's even weirder then

[fervent sage]

it shouldn't matter, what's meant to happen is the client receives the typing event, then wait for message from typing_author in channel then cancel_typing in the client, only if you send the typing and send events close enough together it's possible for delays to cause them to be desynced which makes the typing linger

[brazen charm]

The typing could also be moved to only trigger when a web request is made but the channel would need to go through a few method calls to get there

[fallen patrol]

Uh

#reddit message

[vale ibex]

?

[fallen patrol]

The linked message doesn't look bad to you?

[vale ibex]

fine to me

[vale ibex]

What's bad about it?

[fallen patrol]

Pain.

[fallen patrol]

First part

[vale ibex]

lol not sure what's happened there

I'm guessing something to do with the square brackets in the post title is breaking the android app

[fallen patrol]

Ah

The author used bold in the body

!raw #reddit message

[stable mountain]

You are not allowed to use that command here. Please use the #bot-commands channel instead.

[vale ibex]

it should be escaped, since it works on the desktop client.

I'm guessing it's a bug in the android app

[fallen patrol]

Sigh.

I can't get the raw of that message since it's over 2k characters

[green oriole]

Yep, you can blame that one on Discord

[fallen patrol]

Would need it uploaded to paste to see the exact markdown and why there's an issue

[green oriole]

!raw #reddit message

[stable mountain]

Sorry, an unexpected error occurred. Please let us know!

RuntimeError: Line exceeds maximum page size 1992

[fallen patrol]

I really hope that bots get the 4k limit

[green oriole]

Thank you bot

[fallen patrol]

Run an eval command to upload the output to hastebin 👀

[green oriole]

Well, not like I can do that

[placid ermine]

🤔

i think jason posted about this issue a while ago

[sleek steppe]

That's for .pridefact?

.pridefact

[dusky shore]

The year 2021 is not yet supported

[vale ibex]

yea, all through June the bot posts on a timer

but ofc, we don't have any 2021 facts in the repo

[short snow]

weeeee cj8 is coming

June 21st

wait June 14 for yesterday

Monday, June 14 - Form to submit theme suggestions opens

[patent pivot]

good spot

I guess it's June 14 in some of the world, but 15 in UTC where we put these calendars

[thorny obsidian]

<_< It's the 14th in the only important TZ

[short snow]

its 15 even in UTC i think, i am 5.30h ahead, and it is 7.43h now

[patent pivot]

yeah, it's 15th in UTC now, so we should amend that

[short snow]

spelling mistake, unimportant

vimeo should be capital right 😛

[patent pivot]

ya it could be

feel free to add review comment to caps youtube/vimeo

i'll disable auto-merge

[short snow]

welp,you disabled it just on time

[patent pivot]

lul

[short snow]

nice, its gonna be CLI/TUI this time

[thorny obsidian]

Well.... the precommit hook failed me ;_;

[patent pivot]

lol

[short snow]

lint before you push

[thorny obsidian]

I did!

It passed!

[short snow]

👀 interesting

[patent pivot]

it's because you merged in a suggestion that a trailing space

[short snow]

probably u merged by commit through UI, and it caused error

[short snow]

but i only did change in one word, not aything else

[patent pivot]

yeah, but git has no concept of a word

it's a line change, and the suggested line had a trailing space

[short snow]

oo

[patent pivot]

okay we're good to merge

[short snow]

whats github bootcamp? walkthrough it?

[patent pivot]

hahaha now you mention that the github is capitalised incorrectly

[vocal prairie]

It's a git crash course basically

[short snow]

welp, i forgot even github is GitHub

[patent pivot]

Github should be GitHub yeah

[short snow]

lol

[patent pivot]

oh we never changed the date to tuesday either lol

[short snow]

we should add docs link to to each of the frameworks

[tawdry vapor]

You forgot to change this

[patent pivot]

yeah, i noticed that too

[short snow]

and that lol

[patent pivot]

adding it to my patches branch

[vocal wolf]

we do a little bit of trolling

@patent pivot I've never seen such a complicated PR

[patent pivot]

lmfao

[short snow]

[vocal wolf]

this is a bug in english and needs to be solved IMMEDIATELY

[short snow]

lets see if there is something else remaining

[short snow]

what about this

[patent pivot]

could do, probably not in this PR though

[vocal wolf]

why don't we have a label for html in site

[patent pivot]

lol

[vocal wolf]

shouldn't we?

[tawdry vapor]

Doesn't the platform agnostic rule conflict with some frameworks which are not cross platform e.g.ö curses

[short snow]

is it straightforward or straight forward

[tawdry vapor]

The former or straight-forward

[short snow]

@patent pivot another one

[tawdry vapor]

The latter would be used like "look straight forward" which is a different meaning

[short snow]

yeah

[patent pivot]

yeah, good question. we mention WSL on the frameworks page so I guess that sort of works, but it is a conflict it seems

cc @thorny obsidian

[tawdry vapor]

Since it's only one framework maybe you can just list https://pypi.org/project/windows-curses/ as an alternative, though I am not sure how much of the same API it supports

[patent pivot]

hmmm

[tawdry vapor]

Never mind, urwid is also a problem

[patent pivot]

we could revert the PR adding events page, make all these necessary corrections there (both those in my PR & those mentioned here) and merge when it's ready for sure, having stuff picked up by scrapers & linked from various bits of the site when unclear probably isn't great

linking to docs for each of the frameworks seems smart

[short snow]

linking would be hyperlinking the heading or adding another line below?

[patent pivot]

probably adding a documentation anchor to each

[short snow]

better

[patent pivot]

it is still nice to be able to link to them

so

[thorny obsidian]

ugh, I've changed so many sidebars already ;_; they keep appearing

[thorny obsidian]

The platform agnostic is a holdover from Django. Considering the framework of the jam I don't think it's necessary since we have choices of frameworks

[short snow]

Did codejam8 page just go down?

[patent pivot]

yep, I rolled it back while we get stuff ready

[short snow]

ah alright, is there a PR open with the changes?

[keen valley]

can we get the template of python discord so that we can make one of clone like it
and then I can run the bot and make changes
Or I should create the new one with specific channels and roles

[short snow]

https://discord.new/zmHtscpYN9E3

[onyx acorn]

https://discord.new/tgi

[fallen patrol]

Doing more stuff

The escape is not escaping everything

[thorny obsidian]

auto-merge is spooky

[patent pivot]

lol

[thorny obsidian]

is site autodeploy? or is your k8s still blocking?

[patent pivot]

autodeploy

it rolls back until a new deployment is created

all deployed fine

[sleek steppe]

I think I found a bug

does it do this for anyone else?

https://pythondiscord.com/events/

[gritty wind]

oh boy

haha

I see it in the repo

[sleek steppe]

I know the problem

it's the fact that it has a number in its name

[patent pivot]

cache cleared

it works now

[sleek steppe]

oh

[patent pivot]

b https://pythondiscord.com/static/images/events/summer_code_jam_2021.png?v=1

[sleek steppe]

still doesn't work though

ah I needed to clear too

[thorny obsidian]

It worked locally ;_;

[gritty wind]

Clear your cache

[thorny obsidian]

I swear

[gritty wind]

ah

[thorny obsidian]

I tested it

[gritty wind]

It's working, just caching lol

[placid ermine]

@exotic ember should I close the PR?

nvm

[torpid grotto]

Does this happen a lot? Or is this just a rare issue

[patent pivot]

uhh

[torpid grotto]

Visually what happened was, I saw the typing indicator for the bot for a few seconds. Then it just vanished, then a minute later it sent

[patent pivot]

right now we're upgrading our kubernetes cluster

[torpid grotto]

Ah

[patent pivot]

so services are coming and going

[torpid grotto]

Ok that makes sense

[vale ibex]

!ext r help_channels

[stable mountain]

:ok_hand: Extension successfully reloaded: bot.exts.help_channels.

[vocal wolf]

@fervent sage tag has been reviewed.

[spring fern]

!ext r help_channels

[patent pivot]

huh?

[wide elm]

So I was wondering, why do you guys add init.sql as a volume in your site's repo? https://github.com/python-discord/site/blob/main/docker-compose.yml#L22

[stable mountain]

docker-compose.yml line 22

- ./postgres/init.sql:/docker-entrypoint-initdb.d/init.sql```

[vale ibex]

afaik /docker-entrypoint-initdb.d/init.sql gets picked up by the postgres container on start and runs it for you

so having the init file in that path means the init.sql gets ran on startup

[wide elm]

It runs on startup?

[vale ibex]

It's a volume, rather than in the docker file, as we don't want that to be part of the image that get's pushed to prod

[wide elm]

ah, alright

[tough imp]

you can look at what the script does

https://github.com/python-discord/site/blob/main/postgres/init.sql

it inits the metricity db and adds some testing data

since metricity isnt managed by django's ORM

you need to create the tables manually

@patent pivot can prob correct me if I'm wrong

[patent pivot]

yeah, that's it

we use two DBs for prod, but postgres docker image only inits 1 by default

[dry folio]

hey, i'm wondering how does one implement a tag? is the md file in resources/tags the only step?

[vale ibex]

yup 😄

[dry folio]

yay :D Time to PR ™️
thanks!

[dry folio]

bot#1644 is now on 😄
although this isn't my first time contributing to pydis... it says i need workflow approval

[dusky shore]

GitHub

Add dunder methods tag

[vale ibex]

approved the workflow run

[dry folio]

thanks!

[sleek steppe]

First time contributor to the bot repository

[dry folio]

ah... so it's different for each repo!

[vale ibex]

I'd really suggest reading https://pythondiscord.com/pages/guides/pydis-guides/contributing/contributing-guidelines/supplemental-information/#writing-good-commit-messages too

[dry folio]

yeah, that's my bad, it's been a long time since i've commited to a formal repo (been recently working on my own repos) so i forgot, but i listed the correct commit messages here:

[vale ibex]

Yea, those still aren't ideal either, since they don't really tell me what's happened in those commits

[brazen charm]

you can always ammend or rebase while someone is not working with the branch or if they won't affect reviews

[dry folio]

how would i do that?

[dry folio]

ah... alright

[vale ibex]

git rebase -i would be the easiest. You can change the commit names, save the file and close it. Then git push -f

at least that's how I usually do it

[dry folio]

@vale ibex are the messages better now?

[dry folio]

so i run that right in my command-line?

[brazen charm]

git commit --amend is used to add to the last commit, for going further than that you'll need the rebase, git rebase -i works or if you have a GUI utility for git it most likely provides an interface

[dry folio]

oof- i don't have a gui

wait wat- speedrun

yay only one approval needed

@vale ibex i added a new commit and it's asking for approval again- why? isn't an approval enough?

[vale ibex]

It'll need to be approved for every change you make

until you get a PR merged

[dry folio]

no way
bruhhhhhhhhhhhhhhhh

[dry folio]

alright, i guess that's how it works 😄

[dry folio]

guess i'm waiting for this first pr then lol

[gritty wind]

For a tag, we should probably squash merge anyways

There is no reason for 5 commits to get a tag

[dry folio]

i agree, just squash it all to "create tag" lol

[gritty wind]

It’s still good practice to give them meaningful messages during the process, to aid reviewers, or at least get the experience

[dry folio]

got it!

[trim cradle]

Is anyone available to review bot#1638?

[dusky shore]

GitHub

Proposed alternative "available help channel" message.

[green oriole]

Yus

Smh @sleek steppe, not pressing the update branch button

[sleek steppe]

No point when I can't merge lol

oh wait auto merge was on

I thought it updates it automatically

[green oriole]

lol

Also Stel was so ready for the changelog

That's one perm I really miss haha

[sleek steppe]

👀

[green oriole]

lol

I'm going to ping you every time I need something to be posted there haha

[trim cradle]

@green oriole @sleek steppe thanks!

[trim cradle]

I had time to write it so 🤷🏻‍♂️

[dry folio]

is any core dev available to review sir-lancebot#756? last approval then it can be merged, this pr had been waiting for almost a month. thanks so much in advance!

[dusky shore]

GitHub

Add "cs" and "python" categories to the .quiz command (and more in desc)

[fervent sage]

are mine good :P

[sharp zephyr]

thought it'd be cool to have token detection added to the !paste command . So if someone pastes a token , and clicks save .. before it is saved it reads the whole text looking for the word "token" ... and then checks with the user before confirming save . what do you guys think?

!paste

[stable mountain]

Pasting large amounts of code

If your code is too long to fit in a codeblock in discord, you can paste your code here:
https://paste.pydis.com/

After pasting your code, save it by clicking the floppy disk icon in the top right, or by typing ctrl + S. After doing that, the URL should change. Copy the URL and post it here so others can see it.

[short snow]

Issue: bot#1385 can be closed, since it is much clear right now

[dusky shore]

GitHub

Redis Startup Error

[sharp zephyr]

https://github.com/python-discord/bot/issues/1645

[fervent sage]

why would it search for 'token'?

wouldn't it be better to use our existing token detection & somewhat validation (also, the pastebin isnt part of the bot)

[sharp zephyr]

The time that i almost posted my token into a pastebin , it had "token" listed .

so i see it could also search for a series of numbers/letters that look like a token , and verify that its not .

also there was an update on that issue on github

basically heard that the pastebin service is one that is not easy to add onto . As its a hosted application from another company

was just thinking of any ideas that could support users in not accidently posting their token . I nearly did . (i turned on logging for something else .. and it inadvertently turned on logging of my token in the log)

i just scroll through the log prior to posting . I see others will not

[sharp zephyr]

I haven't seen this existing token detection

[fervent sage]

1 sec

https://github.com/python-discord/bot/blob/main/bot/exts/filters/token_remover.py

[sharp zephyr]

very cool

hmmm

well ... if theres a way to put a message on the "backburner" before being posted.... a bot could grab every pastebin and search the html text for anything that looks like a token .. in the same way.

and if it doesnt pass, deny the message post . And if it passes, paste the link

[fervent sage]

not possible

[sharp zephyr]

the backburner idea is not possible?

[fervent sage]

no, discord doesnt let you do that

[sharp zephyr]

i see ...

the link could be passed to the bot , instead of being directly posted into the channel , and then the bot posts the link when it passes

using a private message to the bot option .. using the / commands

[fervent sage]

thats just becoming harder and harder to actually use though

its not worth it for the minimal benefit it'll provide

[sharp zephyr]

idk , i see there could be a lot done with that . The bot could even create a new pastebin that doesnt include the tokens , and instantly post that instead.

and users could use it and know that their code is safe .. could be optional

and preferred

[fervent sage]

people can still scrape the link before it does that anyway

[sharp zephyr]

mhm , yes . i see tht

i see there could be another way to post bulk code with it being checked like the token_remover does

maybe passing it to a bot instead of using pastebin

[fervent sage]

again its just removing ease of use for little added value

we've considered various ways of adding token removal and in the end it boils down to just not being worth it for the few occasions it actually helps

[sharp zephyr]

i generally saw preserving tokens as high value .

understanding that it has already been looked into

[gritty wind]

We won’t be adding features to the pastebin if it can be helped

We’ve shot this idea down many times before. There is some gain, but not worth the hassle of maintaining our own fork of hastebin

The hastebin also doesn’t have as much a risk of scraping as on server messages, because accessing it involves more steps and resources than scraping the server

[gritty wind]

The issue and linked PR both state that the current setup is only a partial resolution of the issue. I’d still like to see the other suggestions

[green oriole]

I wonder if we could proxy the requests to the pastebin

Yeah, we can totally do that (CC @patent pivot)

Put a proxy that analyse POST requests, and if a token is found put a placeholder instead

[patent pivot]

Yeah that sounds doable

Also worth checking the error responses, we might be able to forge one of those to display it to clients

[green oriole]

Hmm

I guess I could try to reach the characters limit

Seems like it answers with 400 and a json message body

cool

Could I make that a traefik middleware?

[short snow]

Ah right! I will go through the related code and see if there is something I can do

[patent pivot]

nah, would have to be cloudflare workers

[green oriole]

right

[fallen patrol]

it would be nice if it did, but it would have to be a permission thing. Ig like manage messages.….
discord has the api to make this possible, but i wonder if it could handle that many requests.…
what if bots could set up regexes within discord to be deleted so discord blocks them because they match a regex and then it lets the bot know with the interactions api or smth?
hm.…
I'm rambling at this point lmao, respond in #ot2-never-nester’s-nightmare

[patent pivot]

@green oriole do you wanna open an issue on workers to add token filters

[green oriole]

sure

!remind 14h hope you did it, you lazy bastard

[stable mountain]

Yeah okay.

Your reminder will arrive in 13 hours, 59 minutes and 59 seconds!

ID: 2405, Due · Jun 18, 2021, 6:43 AM

[patent pivot]

lolol

[stable mountain]

@green oriole

It has arrived!

Here's your reminder: hope you did it, you lazy bastard.
[Jump back to when you created the reminder](#dev-contrib message)

[green oriole]

I am officially a lazy bastard

[dry folio]

lmaooooooo

[stable mountain]

Missing required argument

expiration

Reminders for Objectivitix#9891

No active reminders could be found.

[dry folio]

wat, i set an 8-year reminder...

[fallen patrol]

WOOO

https://github.com/discord/discord-api-docs/pull/3135/files

inline timestamps

<t:1618953630:d>

seems worth adding to @stable mountain for when stuff infractions end

it will show it in the user's local time

oh wait nvm it doesn't work on mobile (android) yet

[thorny obsidian]

Mhm, that's most likely the plan once it's on stable

[fallen patrol]

oh u already knew of this?

smh

[short snow]

oh that's wonderful, but I don't think it is gonna get stable anytime soon lol, see files

[fallen patrol]

wdym

<t:1624047476:t>
<t:1624047476:T>
<t:1624047476:d>
<t:1624047476:D>
<t:1624047476:f>
<t:1624047476:F>
<t:1624047476:R>

[patent pivot]

it'll hit stable soon

it's been on canary since this morning

[fallen patrol]

ye, the question is mobile

[patent pivot]

android has it

[fallen patrol]

uh no

are you in android alpha?

[fallen patrol]

what version and build

[patent pivot]

i don't know i don't use android lol

friends screenshot

[fallen patrol]

idk what to tell you

i don't have it lol

but i am up to date on beta

[patent pivot]

yeah, it's probably alpha

but we won't look at implementing it at all until docs are merged and it hits stable on all 4 platforms

[fallen patrol]

4?

[patent pivot]

desktop, web, ios, android

[fallen patrol]

o

web and desktop are identical lol

[patent pivot]

in parts

gateway for desktop and web is completely different

desktop uses erlang term format (ETF), web uses JSON

[fallen patrol]

why even lol

[patent pivot]

because browsers can't handle ETF very well

[fallen patrol]

#ot2-never-nester’s-nightmare

[patent pivot]

the TL;DR is that web & desktop operate differently

[vale ibex]

Anyone seen this on bot before? ```
postgres_1 | 2021-06-18 19:03:41.124 UTC [69] ERROR: integer out of range
postgres_1 | 2021-06-18 19:03:41.124 UTC [69] STATEMENT: INSERT INTO "api_role" ("id", "name", "colour", "permissions", "position") VALUES (476190141161930753, '@everyone', 0, 4397059136, 0)
web_1 | Internal Server Error: /bot/roles
web_1 | Traceback (most recent call last):
web_1 | File "/usr/local/lib/python3.9/site-packages/django/db/backends/utils.py", line 86, in _execute
web_1 | return self.cursor.execute(sql, params)
web_1 | psycopg2.errors.NumericValueOutOfRange: integer out of range

loading this on the site directly, it seems the permissions column in only an integer, not a bigint

hmmm, wonder why this is pushing a larger number than normal

ah ok I got it

d.py 1.7 added 3 new permissions

so we will need to bump site max val to allow for that, before bumping bot to 1.7+

Soooo, bumping d.py up to 1.7 for @stable mountain will require changing a column type on site from int to bigint

would that just work with a migration?

[fervent sage]

lol so the breaking change is breaking due to integer handling

[fervent sage]

should be something like ALTER TABLE <table> ALTER COLUMN <column> TYPE BIGINT

actually, exactly that

might take a sec though, it has to rewrite all of the rows in the table because the size changes from 4 bytes to 8

[vale ibex]

yea, django does it all for you

was mostly asking if there's other things to consider

site#534 bot#1648

[fervent sage]

shouldnt be, should be pretty smooth as long as the table isnt like millions of rows

[dusky shore]

GitHub

Don't voice verify ping users who join a stage channel
Add support for Discord.py 1.7+

[vale ibex]

only other thing I saw was the new linesep arg in the Paginator class

[fervent sage]

if it is massive no issues, just the table would be locked until the alter finished

but thats a pretty smol table iirc

[vale ibex]

yea, it's the roles table too, so it's not huge

[green oriole]

Ello, can I get a review on https://github.com/python-discord/bot/pull/1649 please?

[green oriole]

Thank you wookie and Chris

[vale ibex]

you now owe me 1 groote

[fallen patrol]

Hm....

Not sure that's the best solution

I feel like it should check a user's flags and have the 4.2k limit for nitro users and 3k for non nitro users

[remote wigeon]

you cant do that

a discord bot cant see nitro flags

[green oriole]

We are not going to implement any nitro gate either way

Every user should be equal no matter what

and for what is worth I'm really not happy with the 4k chars change

[fervent sage]

im confused why we increased the limit of a spam filter because it was possible to send more? surely no matter how much discord raises the limit, what qualifies as spam remains the same?

[green oriole]

Well, I we don't think we should be muting someone for sending a full length message

At least not automatically

[fallen patrol]

They can tho

[fallen patrol]

So ig the bot spam filter raise is good, however I feel like two back to back full messages should mute

[patent pivot]

how

they definitely cannot

[fallen patrol]

Huh?

OK brb

[patent pivot]

you can see if a user boosts or has an animated avatar

[green oriole]

Well, tell that to discord ¯\_(ツ)_/¯

[patent pivot]

you cannot see if they have a nitro subscription

[fallen patrol]

I thought you could lol

[patent pivot]

nope

[green oriole]

Two full length for a non nitro is the same as one full length for a nitro user

I don't see why we would punish the former but not the latter

Again, nitro is a Discord thing, we are not discord

We don't care if the user has nitro or not

[fallen patrol]

True, idk what I'm thinking

Ah yep

Can't get the nitro flags

Can get every other flag

[patent pivot]

yep

https://github.com/discord/discord-api-docs/issues/295

[remote wigeon]

i dont think they'll ever add that

just think the amount of data that would mean for bots like dyno or mee6

[patent pivot]

yeah, it's closed

[remote wigeon]

data discord definitely doesnt want in competitioners hands

[fallen patrol]

Yeah and there's already two ways for bots to figure out if one has nitro

Three, I guess

Which all require user input

[cold island]

Is there a way to check if a user accepts DM's without sending them one?

[brazen charm]

Don't think you can do that

[fervent sage]

nop

[sleek steppe]

I think so

Well it's still making an API call, but you can send them a blank message and see if you get a 403 vs a 400 :P

[cold island]

Hah, I'd rather not spam people with blank messages

[fervent sage]

thats the thing

the blank message will 400 if its not a 403

and never be delivered either way

[cold island]

o

🤔

[fervent sage]

im curious though which takes precedence

if I were designing the api (as i have done lol) then I'd do that sort of message validation pre-database query to see whether I could send it

so in my api a blank message would always 400, even if it couldnt be sent to a user

unsure if thats how discord handles it though

huh, discord validates that a message can be sent before it validates content length, very interesting

[sleek steppe]

Yeah I tested it and it 403's before it 400's

[vocal wolf]

@timid sentinel that's one small pr

[timid sentinel]

lol. small pr's are my speciality

[vocal wolf]

and this is more of a bug than an enhancement I think

[fervent sage]

I love small PRs

[sleek steppe]

Hmm on mobile it says I can enable auto merge but when checks pass it doesn't merge 😔

[fervent sage]

wait you can enable auto merge on bot repo prs...

[sleek steppe]

Nah I think it's a github mobile bug

[fervent sage]

https://i.modmail.io/17a25745a42.png

it looks like it went through

[sleek steppe]

I disabled because it didn't merge

[vocal wolf]

@sleek steppe @fervent sage here take https://github.com/python-discord/bot/pull/1650

[fervent sage]

already on it

[vocal wolf]

mmmm good

[fervent sage]

lol

[green oriole]

I am speed

Much faster that both of you combinated

[fervent sage]

I dont have bot issues piped into my brain tho

[clever wraith]

You're turning into aboo lmao

[fair spoke]

.

[stable mountain]

:incoming_envelope: :ok_hand: applied mute to @fair spoke until 2021-06-19 20:38 (9 minutes and 59 seconds) (reason: duplicates rule: sent 4 duplicated messages in 10s).

[fallen patrol]

add some of the tui stuff to @snekbox

[brazen charm]

how would you be able to use it

[patent pivot]

yeah snekbox isn't suited for that sort of stuff

[fallen patrol]

rich.tracebacks 😏

!d rich.traceback.install

[stable mountain]

rich.traceback.install

rich.traceback.install(*, console: Optional[rich.console.Console] = None, ...) → Callable[[Type[BaseException], BaseException, Optional[types.TracebackType]], Any]```
Install a rich traceback handler.

Once installed, any tracebacks will be printed with syntax highlighting and rich formatting.

[green oriole]

I don't think we should be adding rich knowing that only a couple of functions will work corectly

It may not even work because it is lacking a big part of the tty interface

[vale ibex]

Could we get bot#1653 reviewed pls

[dusky shore]

GitHub

Use the correct constant name for deleteing after a delay

[green oriole]

no

[short snow]

@brazen charm Could you give bot#1634 another review? :D

[dusky shore]

GitHub

Modpings schedule

[brazen charm]

Have you fixed the errors in the rescheduling method?

[short snow]

I don't get any errors now

[brazen charm]

I can still see them, do you have anything in the redis cache for it to reschedule?
From the ones I checked, there's an unawaited coro for the mod var, it's trying to unpack int keys from the dict instead of using items directly and is passing a string to fromtimestamp

[short snow]

That would have been the issue probably, i probably would have forgotten to disable fake redis when I shifted to docker for testing

i will have a look

[molten perch]

Hey, I'm kind of new to this open-source world, however I think I spotted an issue(in Sir-Lancebot) that I might be able to fix, and haven't been assigned to anyone.
If I'd like take it, my only job is to leave a comment? 😄
(Sorry, I haven't done such things before)

[exotic ember]

yep, leave a comment saying you're interested in solving this issue and someone'll assign it to you

[molten perch]

okay, thank you!

[vale ibex]

If you get stuck while reading through the setup guide, or want some help with git in general, feel free to post here 😀

[molten perch]

Okay! However that issue doesn't seem to be a hard one.

[vale ibex]

Yea, should be a good introduction to the code in general

[brazen charm]

Any other opinions on bot#1641 ? Also waiting for an approval/consensus on bot#1545

[dusky shore]

GitHub

Create tag groups
Add flake8-quotes to linting workflow

[vocal prairie]

I like the idea of 1641

[vale ibex]

1545 had a core dev ping 2 days ago, so you should get a consensus soon

[vocal prairie]

I think 1545 might break too many things, but I haven't gone through the codebase much

[vale ibex]

It'll be a decently sized change to the users, but it opens up a lot of command name space

[brazen charm]

Do stale issue ping or was that for some other reason/manual

[green oriole]

We created a poll to decide how we should interact with the namespaces

[vale ibex]

Yea, it was manual

[clever wraith]

Hey

I found a bug that might be a security issue

[brazen charm]

modmail or an email to admins would probably be the best if it's an actual vulnerability

[clever wraith]

Normally how ir looks when I launch my python script

[brazen charm]

ah

[clever wraith]

One typo

[brazen charm]

this channel is for python discord projects

[clever wraith]

And it seems to load parts of random images on my desktop bar.
Hovering with mouse makes it go grey then change permanently into another random image, until I hover over it again and it changes

[clever wraith]

Oh, ok

[brazen charm]

I'm not sure what that framework is but they should have their own tracker or other contacts

[clever wraith]

Oh I see, that is tkinter for python, to make graphical stuff

Well, I am off to sleep, found it when I did finishing touches before bed. I'll do more bug testing tomorrow and try to find more info. On top of contacting the right place

[brazen charm]

you can find the python bug tracker here https://bugs.python.org/, tkinter should have its own component category

[dry folio]

@sharp timber your changes have been all implemented :D

btw, absolutely genius lmao

[short snow]

oh is api open for contributions?

[sharp timber]

Alright, giving them a look over. FYI (literally fyi, just so you don't do more work) you don't need to merge master in very often when working on a part as isolated as this. Github can just merge it back up so long as there's no conflicts (which, given its a new file and nothing is changing around it, is cool)

[sharp timber]

Alright, just two last things

[gritty wind]

Github (discord probably) sets that on it's own, not sure we can pull it for ya

It's mostly just an inverted the github logo tho

[patent pivot]

hmmm we probably shouldn't use <a>'s here for this navbar item

using a <span> seems to mimic behaviour

[gritty wind]

If you wan the CDN link, that's easy enough

https://cdn.discordapp.com/avatars/409309396040941570/df91181b3f1cf0ef1592fbe18e0962d7.webp?size=128

[vocal wolf]

https://github.com/logos

[patent pivot]

https://github.com/python-discord/site/pull/538

[dry folio]

ah, i see. got it, thanks!

[dry folio]

will look over them now

@sharp timber changes implemented

[dry folio]

thanks a lot bast!! also, why is it not merged? everything's approved

[gritty wind]

It’s linting

[dry folio]

Ah... well i need the workflow to be approved

[gritty wind]

I triggered it now

[dry folio]

Woohoo, thanks!

[dry folio]

Everything works!

!dunder-methods

Aww, still not updated?

[short snow]

let it build and deploy

#dev-log message like that

[dry folio]

Hm, i guess it takes a while

[vocal wolf]

!dunder-methods

[stable mountain]

Dunder methods

Double-underscore methods, or "dunder" methods, are special methods defined in a class that are invoked implicitly. Like the name suggests, they are prefixed and suffixed with dunders. You've probably already seen some, such as the __init__ dunder method, also known as the "constructor" of a class, which is implicitly invoked when you instantiate an instance of a class.

When you create a new class, there will be default dunder methods inherited from the object class. However, we can override them by redefining these methods within the new class. For example, the default __init__ method from object doesn't take any arguments, so we almost always override that to fit our needs.

Other common dunder methods to override are __str__ and __repr__. __repr__ is the developer-friendly string representation of an object - usually the syntax to recreate it - and is implicitly called on arguments passed into the repr function. __str__ is the user-friendly string representation of an object, and is called by the str function. Note here that, if not overriden, the default __str__ invokes __repr__ as a fallback.

class Foo:
    def __init__(self, value):  # constructor
        self.value = value
    def __str__(self):
        return f"This is a Foo object, with a value of {self.value}!"  # string representation
    def __repr__(self):
        return f"Foo({self.value!r})"  # way to recreate this object


bar = Foo(5)

# print also implicitly calls __str__
print(bar)  # Output: This is a Foo object, with a value of 5!

# dev-friendly representation
print(repr(bar))  # Output: Foo(5)

Another example: did you know that when you use the <left operand> + <right operand> syntax, you're implicitly calling <left operand>.__add__(<right operand>)? The same applies to other operators, and you can look at the operator built-in module documentation for more information!

[dry folio]

Let's go!

[short snow]

that is a big one 👀

[placid ermine]

you're implicitly calling <left operand>.__add__(<right operand>)
type(<left operand>).__add__(<left operand>, <right operand>) 👀

[green oriole]

I mean it is almost the exact same thing

[sharp timber]

!e ```py
class A:
def add(self, other):
raise Exception("Fail")

a = A()
def add2(self, other):
return A()
a.add = add2

print(a + a, a)

[stable mountain]

@sharp timber :x: Your eval job has completed with return code 1.

001 | Traceback (most recent call last):
002 |   File "<string>", line 10, in <module>
003 |   File "<string>", line 3, in __add__
004 | Exception: Fail

[sharp timber]

huh

TIL

[green oriole]

I am not sure what could be causing this

[exotic ember]

@green oriole what operating system are you on?

[green oriole]

Linux Manjaro

[exotic ember]

run grep cgroup /proc/filesystems and see if there is a cgroup2 line

if so, it's because you're using cgroup v2, so you'll have to switch it back to v1 https://wiki.archlinux.org/title/Cgroups#Enable_cgroup_v1

at least, I think that's how I fixed it

[green oriole]

I have v1 and v2

[exotic ember]

yeah, you're probably using v2 since it supports it

check the link for switching it to v1 and try running it again

[green oriole]

hmm, will do later on

I probably won't need it before a long time

Hey @sharp timber, can you please have a look at https://github.com/python-discord/site/pull/536#pullrequestreview-688238871 ASAP? If not I can take care of it, there are just a couple of lines to remove

[sharp timber]

It's not running because coverage is misconfigured

It's testing the test itself's coverage

[green oriole]

Yep

and because a branch in the test hasn't ran it isn't happy

[short snow]

@exotic ember @sleek steppe could you give sir-lancebot#745 and sir-lancebot#618 another review

[dusky shore]

GitHub

Add logic for preserving unloaded Cogs.
Wikiguess Game

[sharp timber]

I can change this to just an assertEqual(bool(), True) but that's bad practice. We should be using --omit or something on the tests so coverage isn't testing recursively for coverage

I'm not all familiar with the tool so I'm not really sure I want to touch it

[green oriole]

Couldn't you just do without the try/except?

[sharp timber]

Yeah, although that turns any failure fo the test into an exception rather than a failure

[green oriole]

Well, is that a problem?

It is still a failed test

[sharp timber]

Eh. It feels wrong to not test it and return results and instead error out

I can change it to that though

We do have a coveragerc file it just doesn't have the test locations in it for some reason

[green oriole]

we can do that then

[sharp timber]

Should I make that part a separate PR?

I can do it now over the next, say, 30 minutes

[green oriole]

you can do that in one PR if you wish

[sharp timber]

Sorry, #539 already filed >.>

[green oriole]

no problem haha

now let's poke some people

Good morning @cold island, do you have a second?

[cold island]

hey

[brazen charm]

Having coverage on tests can be useful, I'd just ignore that one branch

[green oriole]

could you please review site#539 and site#536 ?

[dusky shore]

GitHub

Fix .coveragerc testing branch coverage of tests as well as project code
Adjust Message model to support new nitro messages with over 4000 chars

[cold island]

I approved the first one, not available to look into the second

[dry folio]

EYYYYYY CONGRATS ON HELPER POOP!

everyone in the cat dev is ascending

so happy

[placid ermine]

thank

[short snow]

For some reasons reaction adding is reallly slow

on @stable mountain , did !tags

[green oriole]

is it getting rate limited

!int socket

[stable mountain]

WebSocket statistics

Receiving 1.53 events per second.

TYPING_START

28,006

MESSAGE_CREATE

19,278

GUILD_MEMBER_UPDATE

11,112

MESSAGE_UPDATE

2,527

CHANNEL_CREATE

1,264

CHANNEL_UPDATE

1,022

MESSAGE_REACTION_ADD

965

CHANNEL_PINS_UPDATE

462

MESSAGE_DELETE

453

GUILD_MEMBER_ADD

400

VOICE_STATE_UPDATE

259

GUILD_MEMBERS_CHUNK

225

MESSAGE_REACTION_REMOVE

203

GUILD_MEMBER_REMOVE

201

MESSAGE_REACTION_REMOVE_EMOJI

171

GUILD_JOIN_REQUEST_DELETE

48

MESSAGE_REACTION_REMOVE_ALL

17

RESUMED

7

GUILD_BAN_ADD

7

GUILD_BAN_REMOVE

4

CHANNEL_DELETE

4

MESSAGE_DELETE_BULK

4

GUILD_CREATE

3

READY

1

GUILD_UPDATE

1

[green oriole]

Wait, that doesn't count outgoing

1.53/s is a lot

but that's unrelated

!help

[stable mountain]

Command Help

Big Brother
!bigbrother
Monitors users by relaying their messages to the Big Brother watch channel.
!bigbrother unwatch <user> <reason>
Stop relaying messages by the given user.
!bigbrother watch <user> <reason>
Relay messages sent by the given user to the #big-brother channel.

Bot
!echo [channel] <text>
Repeat the given message in either a specified channel or the current channel.
!embed [channel] <text>
Send the input within an embed to either a specified channel or the current channel.

BotSource
!source [source_item]
Display information and a GitHub link to the source code of a command, tag, or cog.

Branding
!branding
Control the branding cog.

Clean
!clean
Commands for cleaning messages in channels.

Page 1/10

[short snow]

hmm

this was fast

[green oriole]

Yeah, it does look rate limited

[short snow]

yeah, thanks

[green oriole]

there was a small break between the second and third emoji

[fallen patrol]

Switch to buttons

Buttons are the future

[green oriole]

maybe later

[sleek steppe]

when discord.py 2.0 is out

[vale ibex]

@stable mountain hasn't gone up in flames 🎉

we're now on d.py 1.7.3

[patent pivot]

poggr s

[fringe sphinx]

yet 👀

[fallen patrol]

Give it a few minutes

[vale ibex]

My code is 100% perfect and never breaks

until it does

[trim cradle]

There are two pinned questions in #help-peanut 😮 what could cause this?

[trim cradle]

like when you got the Python bot stuck in an infinite loop

[vale ibex]

Doesn't sound like something I'd do 🤔

[fallen patrol]

Idk but last time they were manually removed

Hm

Rather than actually fixing the bug, seems like it would make sense to just make the bot iterate through all pins and remove them when a channel is closed

[tawdry vapor]

It could happen if the cache is cleared.

Or if an exception occurs during the process of making the channel dormant, and before unpin gets called.

That would likely only happen if Discord is having issues

The first case has a log message but it's a debug-level message, so it doesn't show up in production logs.

[green oriole]

Seeing how laggy discord was yesterday, I'm blaming them

[hoary vessel]

I was looking at the Wolfram command as I am using it for myself...

I got some error at this point and I literally have to idea how to deal with it.

 async with j1.http_session.get(query) as response:

It was this. But it is a little modified.

There used to be a self.bot there but as I am not using a class so I removed it.

There is no http_session in Bot so I don't know how to deal with it

[green oriole]

You can use your own http session here

[hoary vessel]

I mean, I literally know nothing about it...

How do I even do that?

[green oriole]

You need to create an asyncio session

erm, aiohttp

[balmy sparrow]

http_session needs to be an aiohttp.ClientSession, make one yourself and use it for the requests https://docs.aiohttp.org/en/stable/client_quickstart.html

[green oriole]

anand too fast

[balmy sparrow]

[hoary vessel]

Erm

Anyways, I think I solved it?

[placid ermine]

!d decorators

[stable mountain]

decorators

= decorator+ decorator ::= "@" assignment_expression NEWLINE parameter_list ::= defparameter ("," defparameter)* "," "/" ["," [parameter_list_no_posonly]] | parameter_list_no_posonly parameter_list_no_posonly ::= defparameter ("," defparameter)* ["," [parameter_list_starargs]] | parameter_list_starargs parameter_list_starargs ::= "" [parameter] ("," defparameter) ["," ["" parameter [","]]] | "" parameter [","] parameter ::=...

[placid ermine]

what's up with this embed 👀

[green oriole]

Uhhh

Isn't that the EBNF syntax?

Okay, yeah

It is misdetecting a segment of the syntax as a valid symbol name

[brazen charm]

There is a doc symbol which points to the middle of the code block for every token so there's not much to be done there. The symbol names are from the inventory file so the bot can't misdetect them, just a bit of not so great parsing

[green oriole]

That's stupid

Why did they put a symbol there

I wonder if we can submit an issue for it

[brazen charm]

I mean, it makes sense to scroll you in there with the fragment, just not for our use without some heavier parsing which may be incorrect in other cases

[trail pilot]

Uh I read the how to contribute to the bot but it confuses me even for the sir lancebot

[vocal prairie]

What's confusing?

We can help you out

[trail pilot]

TBH, the entire thing. Apparently we need a docker container and what not?

[thorny obsidian]

Docker makes it very easy to start up, I'd highly recommend using it

[vocal prairie]

Do you know what Docker is?

The application is set up in docker so you can get it running with 1 or 2 commands.

[trail pilot]

Ohh ok

[trail pilot]

Yeah it's a container right?

idk what it exactly is htough

[vocal prairie]

Docker helps "containerize" applications. Basically, that means it run it in an isolated environment. It's not as heavy as a VM, since it doesn't use it's own part of an OS or anything like that. It also helps with compatibility across OSs

[short snow]

I mean if you are using a docker container you don't really need anything else, just your config.yml

if you are not using docker container, then you need

[dense bear]

In the cj8-qualifier repo's readme, it says that "If an item cannot be centered evenly, the extra space character should be to the right of the item (see example 3)".

However, after line 90 in local_test_suite.py, there seem to be 2 outputs for each of the test case. This indicates that the extra space character can be put either to the left or to the right of the item.

Shouldn't the readme be changed so as to reflect the condition used in the test cases?

[cold island]

Yeah, we'll update that, thanks

[dense bear]

np :)

[clever wraith]

wouldn't it be handy if we kinda have a command for meaning of mutating arguments since people seem to stumble upon it quite a bit in qualifiers?

[thorny obsidian]

!mutable

[stable mountain]

Mutable Default Arguments

Default arguments in python are evaluated once when the function is
defined, not each time the function is called. This means that if
you have a mutable default argument and mutate it, you will have
mutated that object for all future calls to the function as well.

For example, the following append_one function appends 1 to a list
and returns it. foo is set to an empty list by default.

>>> def append_one(foo=[]):
...     foo.append(1)
...     return foo
...

See what happens when we call it a few times:

>>> append_one()
[1]
>>> append_one()
[1, 1]
>>> append_one()
[1, 1, 1]

Each call appends an additional 1 to our list foo. It does not
receive a new empty list on each call, it is the same list everytime.

To avoid this problem, you have to create a new object every time the
function is called:

>>> def append_one(foo=None):
...     if foo is None:
...         foo = []
...     foo.append(1)
...     return foo
...
>>> append_one()
[1]
>>> append_one()
[1]

Note:

• This behavior can be used intentionally to maintain state between
calls of a function (eg. when writing a caching function).
• This behavior is not unique to mutable objects, all default
arguments are evaulated only once when the function is defined.

[thorny obsidian]

eeeeey

[clever wraith]

oh it is!

wow good to know, now we can just use it in there!

[vale ibex]

takes up like 1/3 of my portrait monitor 😬

[fervent sage]

we could shorten it by removing a bit of whitespace and using foo = foo or []

[sour wolf]

Theres a typo on the last line, "evaulated"

Also wouldnt it be easier on the eyes if the tags were like a brief description of the concept and linked to a full article on the PyDis site mayhap?

[thorny obsidian]

That's the overall goal now that dewikification is done. Just requires some people to ~do the thing~ which hopefully we'll have a bit more time after the jam

[sour wolf]

Oh, neat

[dry folio]

hey, is any core dev available to review sir-lancebot#756? it's been sitting there for a month now, and i'd greatly appreciate any review! thanks a lot in advance!

[dusky shore]

GitHub

Add "cs" and "python" categories to the .quiz command (and more in desc)

[short snow]

https://github.com/features/issues

[short snow]

@thorny obsidian https://github.com/python-discord/site/pull/543/files#r657571738 I don't think this had been mentioned, atleast I can't see the change

[remote wigeon]

repost from #community-meta:
you use sentry for error collection dont you? I'm setting it up rn and cant figure out how to get it to send issues to discord

@gritty wind but how do i do that

i saw a guide that used a legacy slack integration but i cant find that anymore

[gritty wind]

I was about to link that guide

Umm, that is what we use, but I don't have permission to mess with it really

It doesn't show up in search like the issue says, but it definitely shows up if you directly visit the link

following the guide on my own project, seems it definitely does work

[remote wigeon]

[gritty wind]

Which step is giving you trouble?

[remote wigeon]

i might have overread the part that i need to directly visit the link

[gritty wind]

You can technically get to it through the menus, but that's an overly complicated process

It's hidden under the "legacy integrations" menu, which only shows up on individual projects not orgs

[remote wigeon]

Tried it there as well

But this is kinda annoying, having to configure it for each project

Kinda tempted to make my own integration

[gritty wind]

Shouldn't be too hard, just copy the slack integration, and change it's name to discord :troll:

I'm totally 100% kidding
https://github.com/getsentry/sentry/tree/master/src/sentry_plugins/slack

[remote wigeon]

Internal plugins and integrations are different

Tho kinda tempted to make a pr like that

[short snow]

Perfect timing albedo, I just came here to ask that question lol

oh, so slack plugin works with discord, nice

[gritty wind]

Discord has native support for slack webhooks, for the most part

https://discord.com/developers/docs/resources/webhook#execute-slackcompatible-webhook

[short snow]

Ah alright thanks

[remote wigeon]

tho seriously it shouldnt be that hard, even using the native discord format

https://github.com/getsentry/sentry/tree/master/src/sentry/integrations/slack

su uh the one currently used is a bit more complex

[sleek steppe]

I think sir-lancebot#594 can be closed.

[dusky shore]

GitHub

Only allow 1 concurrent image processing command

[short snow]

pythondiscord got tasklists 👀 and yeah that can be closed

[sleek steppe]

actually I don't think it can be closed

I think we can just do an asyncio.Lock on the in_executor function

or change the max_workers here to 1 https://github.com/python-discord/sir-lancebot/blob/main/bot/exts/evergreen/avatar_modification/avatar_modify.py#L23

[stable mountain]

bot/exts/evergreen/avatar_modification/avatar_modify.py line 23

_EXECUTOR = ThreadPoolExecutor(10)```

[vale ibex]

I meant that to be 1

had it as 10 during dev

[sleek steppe]

yeah I thought I had saw it as one earlier, which is why I thought the issue could be closed

I will make a PR that

[vale ibex]

Ah right, yea that's what i had intended to do

I was going to look into if there was a way to detect when there were a bunch of commands pending, so that we could deny them until the queue lowered

[sleek steppe]

we could use an asyncio.Queue with a maxsize

[vale ibex]

Yea, that would be a good solution

[patent pivot]

@vale ibex power move

[vale ibex]

lmfao

it had to be done

[dry folio]

bump ^

[trail pilot]

Do not open a pull request if you aren't assigned to the issue.
So I had an idea for a project idea command in lancebot which gives one idea instead of the kindlings page on the python disc bot but --
am I not allowed to contribute because I need to open a pull request to suggest the idea because i presume this is already an issue that is occupied

I"m a bit confused by the wording

[vocal prairie]

You can go ahead and open an issue on the Lancebot repo.

You just shouldn't make a PR unless it's confirmed you're allowed to make that change.

[trail pilot]

Oh ok

[trail pilot]

So I just ask to make a PR, otherwise I just open an issue with what I suggest?

[vocal prairie]

The issue is your way of asking if you can make a PR

On the issue template, you can choose either "I would like to implement this" or "Anyone can implement this"

The fomer is saying "I'd like to implement this, can I do so?" and the latter is saying "This is an idea for anyone to implement

[brazen charm]

@green oriole missing a space after the -U in the dump

Probably a good time to bump postgres in bot's compose as well

[vale ibex]

psql doesn't mind missing spaces between single char args like that, so it won't error, but it does look weird

I agree we should take this opportunity to upgrade everything to pg13 really.

huh I seem to have missed that

[green oriole]

Yeah it is correct, both works. I'll add it for consistency though, thanks.

https://github.com/python-discord/bot/pull/1657

[gritty wind]

Hey if anyone's got a Mac, I'd like to get someone to test bot#1626. More testing in general is helpful, but mainly Mac is needed right now

[dusky shore]

GitHub

Pytest-xdist

[vocal prairie]

Hello! This channel is for discussion of Python Discord's projects. If you have a project you'd like to share, you might be better off doing that in a place like r/python.

[wild prism]

!rule 6
Sorry, but we don't allow advertising on the server, @terse vine

[stable mountain]

Rules

6. Do not post unapproved advertising.

[clever wraith]

I am Ayman Essalai, I'm a young freelancer and I really appreciate creating stuff. One of my best qualities is my problem-solving skill. I can debug errors fast and this has really helped me during my coding journey ! I love programming, I dream code and I know : HTML, CSS3, basic javascript and python. With my different services, you're guaranteed to be happy.

[rapid swallow]

Feel free to check out our open-source projects here https://github.com/python-discord

but please dont spam your message in multiple channels

[clever wraith]

I'm rly sry. Won't happen again

[rapid swallow]

no worries 👍

[trail pilot]

I forked the sir lancebot repo but the entire thing is just confusing me

for one I have to download all the requirements so that'll take a while

two where is the file that reads from the yaml files?

[vocal prairie]

Have you taken a look at this guide? https://pythondiscord.com/pages/guides/pydis-guides/contributing/sir-lancebot/

It explains everything you need to know to get started

[trail pilot]

raise DockerException(
docker.errors.DockerException: Error while fetching server API version: (2, 'CreateFile', 'The system cannot find the file specified.')

I got this error trying to do docker-compose up

[brazen charm]

is docker running?

[trail pilot]

I forgot to install docker 🤦‍♂️ oops

[trail pilot]

docker: error during connect: In the default daemon configuration on Windows, the docker client must be run with elevated privileges to connect

getting this error

[molten perch]

Hey! Can I get assigned to issues that hasn't been approved yet?

[sleek steppe]

Generally you wouldn't get assigned until the issue is approved, if you're unsure whether it's approved or not then you can ask here

[molten perch]

Well, there's no approved label there. (https://github.com/python-discord/sir-lancebot/issues/743)
However, that's a relatively small change.

[vale ibex]

approved

Could you comment on the issue so I can assign you? 😄

[molten perch]

Done, thank you! 🙂

[vale ibex]

You're welcome 😄 Feel free to message here if you need a hand with anything 😄

[molten perch]

Sure, thank you! 🙂

[gritty wind]

Have you managed to get it working. Running with elevation usually entails right clicking, and selecting “Run As Administrator”

[trail pilot]

Nope because I kinda gave up for now since I'm tired

so I run docker desktop as administrator?

problem is i'm accessing this from my ide's terminal

[gritty wind]

I’ve never run into this problem personally, but the top google search results imply it’s a problem starting the docker engine itself. If the engine is started, you should be able to start the bot itself from the terminal fine.

As for starting the engine, you can try the right click method as described above, or alternatively this from the top SO result:
https://stackoverflow.com/a/61062715

You don’t need to try this now, just ping me if you’d like more help in the future

[trail pilot]

Ah ok - it said the docker engine had trouvle starting itself

[gritty wind]

Ah

Interesting

[short snow]

What do you guys think of introducting pytests sugar to the testing setup on bot and site?

[short snow]

https://docs.docker.com/docker-for-windows/troubleshoot/ Does this help? I am not windows so can't help myself

[trail pilot]

I'll try

[crude edge]

I installed the sir-lancebot today but evergreen games commands are not working

It is showing HTTPException('400 Bad Request (error code: 10014): Unknown Emoji')> i didn't see any emoji requirements and which emoji Ids to add in documentations

[gritty wind]

Can you post the full traceback

!paste

[stable mountain]

Pasting large amounts of code

If your code is too long to fit in a codeblock in discord, you can paste your code here:
https://paste.pydis.com/

After pasting your code, save it by clicking the floppy disk icon in the top right, or by typing ctrl + S. After doing that, the URL should change. Copy the URL and post it here so others can see it.

[crude edge]

06/29/21 18:17:46 - asyncio ERROR: Task exception was never retrieved
future: <Task finished name='Task-65' coro=<Message.add_reaction() done, defined at C:\Users\Raghav\AppData\Local\pypoetry\Cache\virtualenvs\sir-lancebot-AZvk8AtN-py3.9\lib\site-packages\discord\message.py:1192> exception=HTTPException('400 Bad Request (error code: 10014): Unknown Emoji')>
Traceback (most recent call last):
  File "C:\Users\Raghav\AppData\Local\pypoetry\Cache\virtualenvs\sir-lancebot-AZvk8AtN-py3.9\lib\site-packages\discord\message.py", line 1221, in add_reaction
    await self._state.http.add_reaction(self.channel.id, self.id, emoji)
  File "C:\Users\Raghav\AppData\Local\pypoetry\Cache\virtualenvs\sir-lancebot-AZvk8AtN-py3.9\lib\site-packages\discord\http.py", line 254, in request
    raise HTTPException(r, data)
discord.errors.HTTPException: 400 Bad Request (error code: 10014): Unknown Emoji

[gritty wind]

That's probably the trashcan emoji

[crude edge]

or should i paste whole logs after startup

[gritty wind]

It isn't necessary to add that emoji, and it's safe to ignore

If you want to add it, you can add it under the TRASHCAN_EMOJI environment variable

it's the output of \:emoji_name:

[crude edge]

ok

[crude edge]

but .ships and other commands are throwing these is error when used

[gritty wind]

Without a fuller traceback, I can't help since that command doesn't even use any custom emojis

[crude edge]

full traceback meaning? exception origin from sit-lancebot package ?

[crude edge]

.

[gritty wind]

Honestly anything you can post, since the error above is missing any sort of traceback, it's just an http exception

All the code in the battleship command uses default emojis, so I'm not quite sure how you're getting that error

[crude edge]

yeah but it is the only thing showing up in my terminal

[gritty wind]

a reproduction method would also help

[crude edge]

I can share name of environment variables which i am using

other than that all is standard procedure

[gritty wind]

Can you share the commands/steps you took

[crude edge]

poetry install
poetry run task start

[gritty wind]

I'm confused, do you get this error when starting the bots, or when using the ships command?

[crude edge]

https://paste.pythondiscord.com/uketafusug.apache

i am getting these error after running the ships command

Bot instance is running and logging the whatever exception is generated

[vale ibex]

Could you add BOT_DEBUG=true to your .env file, and see if more info it output?

[crude edge]

ohk

[green oriole]

is it not just DEBUG?

[gritty wind]

definitely bot_debug haha

[vale ibex]

for lance it's bot_debug yea

[green oriole]

oh okay

but it is debug for Python, right?

[vale ibex]

I think debug mode on bot is inferred on whether site is local

[gritty wind]

doesn't exist

what chris said

[green oriole]

[vale ibex]

DEBUG_MODE = 'local' in os.environ.get("SITE_URL", "local")

and then trace is enabled via the cool per cog thing we have

[vale ibex]

any luck? 😄

You'll need to restart your bot after enabling debug mode fyi

[crude edge]

yeah i did that

and .ships is command right?

[vale ibex]

That's a command yea

[gritty wind]

I'm pretty sure it's not the command

It seems to be right after the reddit cog loads, so I'm thinking it might be that emoji

But that error is invoked before the bot is done loading, and is ready to accept commands

[crude edge]

oh i looked it in source code so thats why i am using thats

[gritty wind]

If it was the ships command, you'd see this before it:

06/29/21 16:05:07 - bot.utils.checks TRACE: Scaleios#8200 may use the battleship command as they are in a whitelisted channel.
06/29/21 16:05:07 - bot.utils.decorators DEBUG: Scaleios#8200 tried to call the 'battleship' command and the command was used in an overridden context.

[vale ibex]

Can you put TRASHCAN_EMOJI=:x: in your .env and see if that fixes it?

[crude edge]

i am not using env

[vale ibex]

you must have a .env file

since that's where you put your bot token

[crude edge]

i am using powershell script for initialzing environment

[vale ibex]

not poetry task run start?

[crude edge]

only environment variables

[vale ibex]

ok, then can you put that above in whatever you're doing then?

[crude edge]

yeah i did that

[vale ibex]

and...?

[crude edge]

it is the ouput i am getting :emoji_name:

is it correct?

[vale ibex]

output from what?

[crude edge]

or i need to put its id

[crude edge]

.

[vale ibex]

the env var TRASHCAN_EMOJI needs to be set to an emoji.

[green oriole]

if it is a custom emoji you need <:name:id> I believe

[vale ibex]

I recommended you set it to :x:

if you want to use a custom emoji, rather than a default one, then you need to do what Ak said

[green oriole]

!pithink

[gritty wind]

!python -> \:python:

[crude edge]

ohk i will do that

[vale ibex]

But be aware, your bot needs to be in the server with those emojis, otherwise you will get that 400 error

[crude edge]

is there any emoji which is present in all the server?

[vale ibex]

yes.

the default ones

[crude edge]

so i can use there code right?

[vale ibex]

if you open up your emoji bar

yea

[crude edge]

👁️

i am going with these

[vale ibex]

ok

[crude edge]

what about command .ships was wrong right?

[vale ibex]

There's nothing wrong with the command no

[crude edge]

it is showing command not found

after i turned bot_debug on

[vale ibex]

Could you show the logs?

[crude edge]

https://paste.pythondiscord.com/ijuvijiduv.apache

[vale ibex]

ah ok,. .ships is a subcommand

it's .battleship ships

[gritty wind]

It should work, but try it in the correct bot command channel

It doesn't make sense, but I hit that same problem till I tried that so

meh

[crude edge]

Well i didn't research properly

[crude edge]

yeah it worked

I am not defining .env files but logs are showing Found .env file, loading environment variables from it. Why?

does load_env loads empty file when .env is not present?

[gritty wind]

it's a little misleading

[vale ibex]

That is always printed out by us

[gritty wind]

That should say it found the module

[vale ibex]

if the import works

[gritty wind]

joe is just bad at copy pasting

[crude edge]

Yeah or not found the module

[crude edge]

?

[vale ibex]

if the module isn't found atm, it doesn't output anything

since it's only used in dev

[crude edge]

I think module is automatically installed by poetry

[gritty wind]

It's a dev dependency

[vale ibex]

It's a dev dependancy

[gritty wind]

so it isn't installed in prod

(you install prod by doing poetry install --prod)

[crude edge]

it doesn't make sense to depend on import error

[gritty wind]

Why not?

[crude edge]

How does env variables initialized in production environment

[vale ibex]

We use kubernetes in production

All of those env vars are already loaded as secrets in the environment

[crude edge]

My bad it makes sense if we aren't installing python-dotenv in some any case

[vale ibex]

Yea, if you look in the Dockerfile, you can see that we build deps with poetry install --no-dev

[vale ibex]

--no-dev

[gritty wind]

blame yarn

[vale ibex]

lol

[vale ibex]

More info on our infrastructure 😄

[crude edge]

But it also leaves edge case when we use dev environment

[vale ibex]

We have documented steps on how to setup a dev environment, using both poetry and Docker

if people want to do things themselves, then that's on them, including any issues

we can always help with them here though

[gritty wind]

The print is misleading, but that's less of an intentional decision, and more of an oversight

[vale ibex]

Yea, since we assume if python-dotenv is installed, a .env file is going to be there

[gritty wind]

Like chris says, dotenv is an optional helper, even if we do recommend it for most

[crude edge]

Thanks for the help

[short snow]

#ot2-never-nester’s-nightmare message some more 👀

It is a complicated one lol

[molten perch]

Hey! I have a question. When using Sir Lancebot's "issue" command I experienced a strange behaviour when the first argument (numbers) was not shipped, or was put in the wrong place, the numbers parameter will be an empty array, and because of that it won't fetch any issue and just simply will return a embed with a Github link that points to either sir lancebot's repository or the repository you typed in or in case of wrong order it points to a 404 Github page, but no issues from Github or error message whatsoever, is this an expected behaviour?
Because, it would be cool if it sent an embed with an error message, and invoked the help command.

[cold island]

Can you reproduce?

[molten perch]

While testing, I managed to reproduce, yes. Would you like me to send one into the bot commands?

[vale ibex]

PS C:\Users\Chris\src\kubernetes> (Get-ChildItem -Recurse | Measure-Object).Count
179
```lol yea

on the issues command, i remember us saying we can just remove it entirely, in favour of the regex listener

[cold island]

yeah, it's a bit difficult to understand without an example

Here is fine too

[molten perch]

.issue site 1

[vale ibex]

.issue site 1

[dusky shore]

GitHub

[sleek steppe]

Shouldn't it be

.issue 1 site

[dusky shore]

GitHub

adds gunicorn server_ready hook to set up database

[molten perch]

Yes, it should. But the bot doesn't throw any kind of error, or something.

[sleek steppe]

Yeah it takes all the first items that are ints

so you fetched 0 issues

[molten perch]

And since that's not an int it will return an empty list, yeah.

[cold island]

I wouldn't mind it throwing the help embed in that case

[molten perch]

Yes, that was my idea. Actually, I might have done it wrong, but I've already implemented a solution for that.

[vale ibex]

if not numbers:
    raise commands.UserInputError("You must enter at least one issue/pr number to fetch.")