Lemme get on my PC because then it'll be half the amount of lines and more processing power
#dev-contrib
static canyon
sour viper
Just write it to another file, dont display it ? would that work ?
static canyon
Stop being big brain smh
brisk brook
Wait what does that mean?
static canyon
I guess we want the dict format right
Or just a 1d list of all the anagrams put together?
short snow
dict format probably
sour viper
was looking at some other resources, how they have worked with jsons
short snow
from what i understand of maxy's appraoch
brisk brook
I am so confused, for all I know you may be cooking drugs
static canyon
It just means we then won't have a list of stuff to see if it's a valid anagram
I guess we just sort letters and get the values, nvm
brisk brook
What's the purpose of printing 5106 lines 😂
static canyon
That'll be even better because only checking possible matching anagrams (same letters and same length) rather than every anagram in existance
static canyon
I put the entire thing into a json easily
Wait shit that's the wrong one
Sec
this is the correct one @sour viper
There's just some words that are questionable as to whether they count
Like "aahed"
short snow
"aaby": [
"abay",
"baya",
"yaba"
],
"aa
wow
did u verify those words?
or just generated them
static canyon
Neither
sour viper
"aanst": [
"antas",
"nasat",
"santa",
"satan"
]
static canyon
It's every word from https://raw.githubusercontent.com/dwyl/english-words/master/words_alpha.txt that's at least 3 chars @short snow
sour viper
its working
short snow
ah okay i saw your script
sour viper
i checked some random words, which i didnt think were words, they were words
static canyon
So this will make the actual command part really easy now
short snow
but those words lol
static canyon
but those words lol
Remember the keys aren't words, that's just the sorted letters
sour viper
end product is gonna be like poor mans scrabble
short snow
yeah i know that
static canyon
But yeah there is some questionable words
short snow
i was looking into teh values
static canyon
I'll probably filter through it tbh
short snow
https://github.com/first20hours/google-10000-english maybe try this
sour viper
just a small concern, other peoples resource files are around 36kb max then there's our file, around 2MB
static canyon
https://github.com/first20hours/google-10000-english maybe try this
That gives 1302 from 9868 words
28kb
Although that's including 2 letter words which I don't think we wanted
Here we go, 23kb
sour viper
huh santa and satan not present
static canyon
Because they're not top 1k words I guess
gleaming jay
must've went on vacation
static canyon
The file has 9474 words that are 3 letters or more
And santa and/or satan aren't a part of them
sour viper
i see a potential problem not sure though, what if the word is anagram but not in the popular list ?
static canyon
Then it wouldn't match
sour viper
But it would fulfill the rules of anagram wouldnt it ?
static canyon
Like sayjson "aemn": [ "name", "mean" ],another valid anagram would be "mane"
But it's not included
short snow
welp
stable mountainBOT
It has arrived!
static canyon
I'm gonna filter through the massive file for a bit and see what comes of it
@sour viper if you fork the sir-lancebot repo, create a branch and add me to it I'll push the file when I'm done
Then I'll leave the rest of the issue to you
sour viper
Sure i have it forked, im creating the branch right now
static canyon
My GH is TizzySaurus / tizzysaurus@gmail.com fyi
molten perch
Oh, no worries. The solution ended up to be StringIO
sour viper
My GH is TizzySaurus / tizzysaurus@gmail.com fyi
you must have the invitation to collaborate now
clever wraith
i won’t have time to work on anagrams anymore so i’m just gonna same day the current code i have
y’all feel free to adjust do whatever
static canyon
you must have the invitation to collaborate now
yep, ty
static canyon
My wifi just died but fyi I've filtered a fair amount
Thing is there's just so much
I've done >5k words but then there's a total of like >70k (so only ~7% in 2 hours)
So it's at this point I think we have to make a decision:
-
Use the smaller word list and accept that it's going to have missing anagrams
-
Use the bigger dataset without refining any further, and accept there may be some false-positives in terms of anagrams (means we may have "unwinnable" games)
-
Use the bigger dataset and keep refining, but instead of removing bad words, we keep things we know are right and remove everything else (will have missing anagrams, but less than option 1)
-
Use the bigger dataset and fully refine removing all bad (will probs take days to fully refine and don't think is worth it)
I personally would probably go for the third option but interested in your input @sour viper (and I guess @short snow & anyone else if you'd like)
stable mountainBOT
It has arrived!
fervent sage
I am in need of a list of safe english words to use for other things, what we could do is create a quick & easy website that shows a word and a yes/no button give it to a bunch of people and it should be done fairly quick
static canyon
Yeah I mean that's basically what I've got rn just locally so only me doing it
This word list would also have other uses tbf (like the hangman PR)
But then this is a list of anagrams, not a complete list of all safe words, so I suppose that's something to keep in mind
The full list of words is 369,646 words, whereas list of anagrams from those words is currently 72k ish (can't get exact number rn)
So to make a full list is 5x more work
It makes more sense to remove from the 370k list though and then just regenerate the anagrams since anagrams can be generated in <1sec
sour viper
I was thinking a mix of option 1 and 2
We can have a list of missing anagrams, whenever someone enters the missing anagram we run through 2 checks same letters as scrambled and check if that word is present in original list of all words to see if it's valid?
static canyon
Eh I don't think that's a good solution personally
We want to be minimising what we store, so not storing both
Especially when as you pointed out earlier it's like 4Mb
static canyon
I am in need of a list of safe english words to use for other things, what we co...
I think this is the best option in an ideal world but don't know how well it would really work
fervent sage
do as we do in minecraft@home
create 2 (or more) tasks for each word
give the tasks to different people
if the results match eachother accept it
else put the tasks back into the pool
stable mountainBOT
It has arrived!
static canyon
Right, yeah. What do you mean by "tasks"? Just having at least two people verify it?
I guess, two different people
So it'll require two "votes" for accepting, and one for denying? @fervent sage
fervent sage
So basically you have a database (it can be sqlite, not important) of "tasks" which in this case will just be words, when a user visits the website an uncompleted task is selected and given to them, they send a result and the task is marked as validating, then that task is sent to another user to complete, and that also gets marked as validating. When both (3 may be more reliable actually) are marked as validating you check those results to see if all of them have the same completed result, if not you release the tasks back into the pool of tasks and they get sent out again. Repeat until that check returns true
That's what we do for validating mc@h results to make sure people dont just send in bullshit results
the db would look something like
CREATE TABLE Tasks (
word VARCHAR(255),
status INT, -- An enum for the status of the task, `uncomplete`, `validating`, `complete`
result BOOLEAN -- whether the word is good or not
);
Right, that makes sense
austere hornet
Hey @vale ibex , could you please review sir-lancebot#843 when you have a moment? I would greatly appreciate it. Thanks!
trim cradle
@cobalt sun tells me that there are python to js transpilers and I wonder if this is something we could use for our projects
In particular, it would give me a lot more ability to work on smarter resources.
cobalt sun
I am new to hear about it as well, and I'll try to find out whether it's worth looking into seriously by tomorrow, unless anyone already has experience
trim cradle
Sounds like Joe and Alex aren't into it
patent pivot
Transcrypt is something we should avoid at all costs
it's just writing JS in python flavor, it's not Python stdlib at all
trim cradle
Well fuck
patent pivot
no stdlib packages, still got to use functions like addEventListener and so on, still got to use document
fervent sage
So it doesn't even save you any work
it actually makes it harder
not only do you need to write it in js format, but you have to forget certain things you'd normally do in python
clever wraith
i know this has been long and really annoying, but sir-lancebot#849 could use reviews
remote wigeon
Does anyone know why the converters are classes and not simply functions
you can iterate over functions as well
trail pilot
What if…
geeksforgeeks command
That grabs an article from geeksforgeeks
I might make an issue for it on sir-lancebot once I’m done with one of my two current PRs
eternal owl
Sounds like a great idea, maybe we can combine other sites similar to geeksforgeeks for the command ? I guess there are a few more notable ones @trail pilot
A thing to consider is, does geeksforgeeks provide a rest API to interact with which is robust and convinient ?
placid ermine
That grabs an article from geeksforgeeks
there are concerns about the quality of some geeksforgeeks articles though
trim cradle
What
trim cradle
there are concerns about the quality of some geeksforgeeks articles though
The python examples on geeks for geeks are embarrassing
trim cradle
What changes? We discussed what the MVP is in the staff meeting yesterday and I finished implementing it except for an issue with some redirects
That's not what mine looks like
short snow
🤷♂️ My branch is upto date, and i ran task start
gritty wind
I had that same problem when I checked the branch out a while ago (but I can't replicate right now because I'm getting warnings about the lockfile being out of date, and navigating to the page throws an error)
File "pydis_site\apps\resources\views\resources.py", line 11, in _parse_checkbox_options
return set(options.split(",")[:-1])
Exception Type: AttributeError at /resources/
Exception Value: 'NoneType' object has no attribute 'split'
Working around that bug, this doesn't seem to be a problem anymore
short snow
I had that same problem when I checked the branch out a while ago (but I can't r...
even i get lock file out of date
clever wraith
can anyone tell me how to setup snekbox inside my discord.py bot
green oriole
Sure, you will need to run a local instance of snekbox. You can find some instructions there https://github.com/python-discord/snekbox#running-snekbox.
After that, you can add a cog to your bot calling the API. You can take inspiration or even copy our own implementation.
!source eval
stable mountainBOT
clever wraith
the thing is, i want to set it up inside heroku
green oriole
Ah, I don't think that's possible. You should look at public evaluation API for that, I believe Piston is often suggested? Or something like that, I don't remember honestly
clever wraith
After that, you can add a cog to your bot calling the API. You can take inspirat...
but will this work?
and will it use too much memory??
green oriole
Snekbox is a totally different service, most probably incompatible with heroku
austere hornet
That grabs an article from geeksforgeeks
I love GeeksForGeeks but it looks like some are against it, so probably not worth making an issue for this one.
trail pilot
Just read RealPython
I agree that RP is much more specific but there are some very specific topics geeksforgeeks cover that doesn’t exist on RP
brisk brook
Any other article you'll find is usually better covered than geeks for geeks, sometimes it may be the only result
But at that point, a command isn't enough. You need to Google search with a bit more effort
trail pilot
Hm, what if realpython could be refactored to article, as a generic one, and G4G could be a fall back if RP doesnt exist
cold island
Yeah I wouldn't like a command for a learning resource we wouldn't show on our website
green oriole
I really don't want to like to G4G, their "article" suck sooooo bad
They are badly formatted, outdated and sometimes even outright not working
trail pilot
Hm
I’ll look at some other resources on the website and see if there’s any alternative
clever wraith
Yeah g4g is sketchy
surreal venture
Is it possible to add a "practice" section to the resource page of the pydis website?
vale ibex
The resources section is being overhauled as part of site#582
dusky shoreBOT
cold island
w3schools?
Seems even worse to me
surreal venture
really? isn't w3schools like regularly updated and is maintained by a set group of people?
vocal wolf
w3schools is pretty bad, lots of incorrect information and what not
tbh I'd prefer if we'd stay away from both g4g and w3, realpython will do.
vale ibex
Yea, I don't see a need to expand further than real python
late wolf
Why the use of ] prefix?
#bot-commands message
thorny obsidian
It's a temporary bot and other common prefixes would collide
late wolf
ah didn't notice the differnece
cold island
Don't get me wrong, I still sometimes use g4g and w3 for quick reference sometimes if they're in the first google results, but I wouldn't use it as a teaching aide
Lately though I just use the python docs more often than not
dim pelican
At what point do we have too many article grabbers? It would be interesting to have a generalized one, but at that point google is probably better and faster
(I did the realpython one, so I definitely see the potential to adding more, just raising the question)
clever wraith
Realpython is probably all we need imo
molten perch
Hey! Could someone take a look at sir-lancebot#857 ? 🙂
dusky shoreBOT
clever wraith
sir-lancebot#849 could use your reviews
surreal venture
I'll check out both ¯_(ツ)_/¯
clever wraith
Why?
vale ibex
Discord py has the message intent already
fallen patrol
no
vale ibex
the change discord's end is jsut making it privileged
fallen patrol
that's different
no?
its really not lol
message content != messages
bot still can know that messages are being sent, but not see the contents
fallen patrol
the change discord's end is jsut making it privileged
their support doc literally says When can I apply for the new intent?
thorny obsidian
Isn't that only for verified bots?
vale ibex
The way I read the announement is that you will be sent messages as normal via the gateway with the message intent, but if your bot doesn't have access to the privelidged intent, you just won't receive content, embeds, attachments, components in the message object
vale ibex
Isn't that only for verified bots?
You still need to toggle on the privileged intents, it just bots in 100+ server need to apply for it to be enabled
vale ibex
possibly
fallen patrol
just look in the ddevs server smh
its a new intent which has to be declared like any other
dim pelican
So the bots in the repo are only used here right? Which is less than 100 servers and therefore unverified?
fallen patrol
just like presence and guild member data.
eg. turn on in developer settings and declare intent when connecting
patent pivot
yes, but then the bot still needs to request said intent on logging on
where have they clarified that content is becoming a new intent
enabling in settings does not necessarily mean that there is a new intent, it just means that message data comes through with messages
you are jumping to conclusions with pretty limited evidence, if there are links please send them, if there are messages in ddevs also feel free to link them, but as far as I can see this is just a toggle that attaches data already, you have it or you don't at the app level, not the session level
I could well be wrong, but as far as I've seen there is no evidence pointing towards your suggestions either
fallen patrol
@vale ibex someone with your exact pfp, username, and ID has entered the chat where I asked 👀 is that u
vale ibex
yea
currently waiting for someone from discord to weigh in
rather than someone saying it's implied
silent walrus
Where is the code that implements the bot responding, when a user incorrectly enters python code? on the pybot github.....
vale ibex
!e Are you referring to this error? ```py
print(n)
@vale ibex :x: Your eval job has completed with return code 1.
001 | Traceback (most recent call last):
002 | File "<string>", line 1, in <module>
003 | NameError: name 'n' is not defined
or some other error?
silent walrus
when the above is incorrectly inputted
vale ibex
ahhh yea, that makes sense
silent walrus
<https://github.com/python-discord/bot/tree/main/bot/exts/info/codeblock>
thanks!
@brazen charm can the bot edit a user's previous post?
silent walrus
I'm curious if it's possible to "auto-correct" incorrectly pasted Python code
vale ibex
You could have the bot reply with what it thinks is correct
but then you have 2 messages with the code taking up a lot of screen space
silent walrus
No, you can't edit another user's messages
discord itself prevents this?
vale ibex
if you delete the user's "incorrect" code, they then lost the ability to edit it
vale ibex
discord itself prevents this?
yes
vale ibex
it's quite a lot of effort to allow that, since if a user edit's another user's message, then you need to store the fact that the original author didn't type the new message, and have that visible in the client somewhere
otherwise a mod could change a user's message to say whatever they want, and possibly lead to the user being banned
silent walrus
I see...
vale ibex
and it's effort that Discord just hasn't deemed worth it so far
silent walrus
is the "python" tag used to paste python code, markdown?
vale ibex
both py and python can be used
silent walrus
both `py` and `python` can be used
I know but what supports them? is it markdown?
or part of the discord client itself..?
vale ibex
Yea, it's part of the client
Discord uses an internal fork of https://highlightjs.org/
silent walrus
interesting
cold island
Idea for lancebot because I'm on the bus and bored: a command to convert a sentence to pirate speech invoked with .arrgh
clever wraith
Hey I don't know why but why list methods are not in !d
letting people know about list methods is needed IMO.
brazen charm
the python docs do not expose them
clever wraith
you mean its coming from some api call of theirs?
brazen charm
The bot parses the html of the looked up symbol by going through the HTML tag's id, the base list methods don't have symbols associated with them so they don't have ids to look them up with
clever wraith
The bot parses the html of the looked up symbol by going through the HTML tag's ...
i can see list on html of their site.
clever wraith
The bot parses the html of the looked up symbol by going through the HTML tag's ...
oh
uhm
brazen charm
The list methods page people usually look at is a tutorial page, then the normal documentation would be https://docs.python.org/3/library/stdtypes.html#sequence-types-list-tuple-range
The source doesn't provide anchors for them, so the file we parse for symbols doesn't contain the methods and it's impossible to find them directly on the page without knowing their contents
clever wraith
oh so str works perfectly because they gave it out there!!
so I mean, if we can do a little bit of conditioning and parse that html for list, wouldn't it be worth?
reasoning: lists are the data structures which are anyways going to get used a lot. and !d helps in helping a lot. and it would not affect the speed since it just involves 1 static conditioning.
dim pelican
#dev-contrib message
Any tips or recommendations for this one?
Preview not rendering a created .md or .txt file
brazen charm
so I mean, if we can do a little bit of conditioning and parse that html for lis...
No, I don't think specialcasing this would be worth it. If the methods are necessary they can be looked up easily (or even explained with their simple behaviour)
clever wraith
wouldn't disagree with can be explained easily part, but seems worth to me.
should i make an issue or just ping Xith here? so that atleast we can see opinion on this. i kinda feel it may worth.
brisk brook
https://discord.com/channels/267624335836053506/635950537262759947/8865885336267...
I am not sure if this is the best way, for many these do not render in the client (mobile users).
dim pelican
I am not sure if this is the best way, for many these do not render in the clien...
Right, I was just trying to implement it from Joe's suggestion
brisk brook
Yeah, neither options (Markdown file or embed) are perfect
dim pelican
I think the embed with a link to the section is ok for both mobile and desktop
But the original issue author reached out mentioning they wanted to add this additional feature
gritty wind
should i make an issue or just ping Xith here? so that atleast we can see opinio...
You can make an issue at any time (no need to ping Xith unless you explicitly want his opinion). Opening an issue is by no means an indication something is approved or will happen.
In terms of the idea itself... eh, I don't care much for it. There isn't a great place in that docs page to link to begin with. Like the main mention I'm seeing of it is under Mutable Sequence Types, and that's literally a single line among a list of other functions that will probably be more confusing than helpful to someone (espically if it's someone just starting out and needing to learn remove)
clever wraith
You can make an issue at any time (no need to ping Xith unless you explicitly wa...
you mean this link?
https://docs.python.org/3/tutorial/datastructures.html
gritty wind
The list methods page people usually look at is a tutorial page, then the normal...
I was actually looking at this page lol
Still, not particularly convinced it's all that worthwhile
clever wraith
yeah that's where its currently referring to i think.
oh alright, thanks!
I'll leave this idea then.
gritty wind
(Are the tutorial pages actually included in the symbols?)
Thanks for suggesting the idea
Different opinions and creativity are what keep things alive
cold island
I just built it
cold island
I didn't pull the image, I built it locally
Since I needed to configure metricity to the test server
vale ibex
The metricity image?
cold island
yeah
vale ibex
are you able to exec into the metricity image and cat entry_point.sh
cold island
Because I'm on windows?
vale ibex
uhhh, not sure, I'm on windows and didn't have a similar issue
cold island
yyyeep
gritty wind
Ah
cold island
Isn't this running *inside* docker anyways?
yes but it pulled the files from my local machine
clever wraith
(Are the tutorial pages actually included in the symbols?)
oh I'm not sure where to look for for that, I'm new to /bot repo.
did look into ext/info/doc but couldn't find symbols
gritty wind
You can configure git to not add CRLF on it's own, but for whatever crazy reason that isn't the default, so it may not be the most viable
vale ibex
Ah, I'm guessing this isn't an issue for the image I built locally as I have autoclrf set in my global git conf
git config --global core.autocrlf input
gritty wind
oh I'm not sure where to look for for that, I'm new to /bot repo.
did look into ...
The symbols are pulled dynamically from the website
!docs getdoc python
vale ibex
it could, if we had one
cold island
yeah ok, it doesn't go insane anymore
gritty wind
But from the docs command, it doesn't seem to pull them
clever wraith
wait that made me download this inv file
cold island
However it seems it failed to set up the database
gritty wind
Now you probably wouldn't actually browse that by hand
^
clever wraith
i tried opening with vscode, didnt work lol
gritty wind
It's a compressed file
Sphinx inventory version 2
Project: Python
Version: 3.10
The remainder of this file is compressed using zlib.
cold island
I'll kill all the containers and try again
cold island
I'll kill all the containers and try again
Resulted in
vale ibex
If you're still having issues in ~45 min I can debug over screenshare if you want
clever wraith
It's a compressed file
oh i see so python docs are created using this Sphinx thingy and that creates this compressed file where symbols(not sure what they are 100% tho) are, and you guys basically do crawling depending upon those symbols?
brazen charm
the inv file tells the bot what symbols (e.g. str etc.) are available and where they are, then it uses that information to scrape the doc pages and convert that to Markdown to show here
There are no symbols for the list methods in the tutorial page so the bot can't parse them as it would be impossible to find them. Then linking to it would be weird without an anchor too if it had something to accommodate them
gritty wind
Well considering we're trying to solve the problem so people don't need to run other things themselves 🤡
gritty wind
Basically, the thing Zig and Chris are working on is the fix for the issue, so manually running the sql is not an option
The plan is to have docker running it
clever wraith
There are no symbols for the list methods in the tutorial page so the bot can't ...
oh i see, i really thought the process was straight forward when you mentioned html parsing.
my earlier assumption was that you guys were just fetching html from certain pages by some mapping and then with help of soup you'd filter content and then show.
I do partially understand this now.
green oriole
Resulted in
did you pulled again though?
gritty wind
There probably is no way to know from context, I've just been following this for days lol
cold island
did you pulled again though?
I'm not pulling anything, this is a local image
green oriole
because if the child container hasn't been updated it will cause issues
oh
well
you are running the site, right?
cold island
The only thing I can think of is that metricity isn't speaking with the DB, which happened last time I set it up
green oriole
but did you made a change to the compose to use a local image?
cold island
yes
green oriole
Interesting
cold island
The only thing I can think of is that metricity isn't speaking with the DB, whic...
actually no, that shouldn't be the case, since I'm not starting it up separately
gritty wind
Yup, sorry that wasn't my intention 😅
vale ibex
free now <@!512354988157173763>
.
work project
lol
it's a bot that integrates with slack
forgot I had it open tbh
just using vsc as a terminal rn lol
I never mute
gritty wind
I'm muting because I've got a terrible mic 🤡
Discord is having a day, and I can't watch the stream
Good luck you two
cold island
I'll be available in like 10 minutes
Ok I'm around
Don't have a mic 
Sec
Need to make sure I'm not streaming something I shouldn't
By the way
When I set up metricity
ok
could just build lol
green oriole
time for me to go, cya 👋
merry bone
bubye
green oriole
Debugging nights with Chris and Zig, sounds like an event title to me
cold island
It is true
@vale ibex
# The guild ID that Metricity should be reporting about
guild_id = 476190141161930753
# This is the role ID of the role that should identify staff members, for Python Discord this is the helper ID
staff_role_id = 476190429960732672
# Channels that should be flagged as staff only by category ID
staff_categories = [
609529926366003201,
686646323985907808
]
# Don't report messages for the following categories
ignore_categories = [
609530625233387540
]
What I set ^
gritty wind
clever wraith
What is this code about?
clever wraith
I didn't follow from the start
brisk brook
Tl;Dr Docker being a whale needs to blow out water from it's nose-thingy
vale ibex
volumes:
- ./config.toml:/metricity/config.toml
gritty wind
join us
gritty wind
👀
gritty wind
for key, value in config.items():
if isinstance(value, dict):
if env_var := value.get("env"):
if env_value := environ.get(env_var):
config[key] = env_value
elif default := value.get("default") is not None:
config[key] = default
That's- what
gritty wind
I should just move to pydantic
yesyesyes
And you wrote most of it - I wrote the last two lines (the broken ones)
patent pivot
And you wrote most of it - I wrote the last two lines (the broken ones)
that's what we like to hear
patent pivot
That's- what
tbf, works well
clever wraith
Lol
gritty wind
if env_value := environ.get(env_var):
if (key_type := value.get("type")) is not None:
if key_type == "int":
env_value = int(env_value)
config[key] = env_value
🤡
for key, value_type in dictionary["__annotations__"].items():
config[key] = value_type(config[key])
dictionary.update(config)
@patent pivot we need your opinion on a fix
_type = dictionary["__annotations__"][key]
if type(_type) is type:
config[key] = _type(config[key])
🤡
cold island
Wtf is this
cold island
Every day we stray further from God
vale ibex
lol
vale ibex
1 word
gritty wind
Go on 👀
vale ibex
# magic
_type = dictionary["__annotations__"][key]
if type(_type) is type:
config[key] = _type(config[key])
Thank you for your infinite wisdom good sir
vale ibex
does python refer to str int float etc as primitive types?
or does it use a different term
brazen charm
I don't think it uses that term anywhere around builtins
patent pivot
I revised tihs config system in Olli to have type support with pydantic, so long term I could look at something like that
patent pivot
lul
vale ibex
the issue all came from wanting to run metricity via the bot compose
so we need to give it guild id etc
so env vars sounded good
but env vars are strings
🙃
patent pivot
imagine env vars
vale ibex
imagine a place
cold island
so we need to give it guild id etc
Tbh at that point giving the bot a toml file to override with in dev sounds ok
Putting it in an .env doesn't seem any better
cold island
Did you just push the dev config file? Lol
vale ibex
yup
No point in pushing the actual config-default, since this will never be used in prod
so might as well use our dev values lol
patent pivot
this is so webscale
vale ibex
cold island
Hmm. I'm not sure how I like it in the repo
patent pivot
i am also somewhat hmming it
vale ibex
yea, it doesn't feel right
cold island
I imagined we'd add it like config.yaml
vale ibex
the file needs to be present in order to mount a volume
cold island
But it doesn't need the file in prod
vale ibex
We use docker-compose to shove this file into the metricity container, since that's what needs it
docker-compose won't let you use a volume if the file doesn't exist
cold island
Right, but we don't want a volume in prod
cold island
Hmmmmmm
vale ibex
but yea, how do you see init scripts working joe?
we'd need to get the local config into the metricity container on bot somehow
cold island
And it only mounts a volume when you set use metricity to true?
vale ibex
it always mounts the volume
cold island
Hmmmm
vale ibex
there isn't a way to conditionally mount a volume as far as I could find
patent pivot
but yea, how do you see init scripts working joe?
init script to pull the config from the repo, but thinking about it the values do need to get changed
vale ibex
yea, we need a way to overwrite with local values
patent pivot
maybe the pydantic approach with env vars is the best
vale ibex
yea, it's just a decent peice of work due to the number of env vars
so we'd need to write a bunch of validators
patent pivot
hmmm
vale ibex
actually, it's not so bad if we force the use of env vars
even in prod
rather than having config default and env vars
patent pivot
what other service did i write that does this
oh king arthur
hold on
this also works
if we force env vars, which I'm not opposed to
cold island
You lose the yaml structure though
vale ibex
lists in env vars are yikes too
patent pivot
grafana solve this by using _ as a .
patent pivot
lists in env vars are yikes too
csv isn't too bad tbh
patent pivot
would assume so, haven't tested though
METRICITY_BOT_TOKEN=X.Y.Z
METRICITY_PYTHON_LOG_LEVEL=INFO
METRICITY_DATABASE_URI=postgres://foo:bar@baz/qux
I will have to double check whether pydantic allows that
it certainly has sub-models in the config examples at https://pydantic-docs.helpmanual.io/usage/settings/
but doesn't show the env vars
ah
yeah it does
hm
export my_prefix_more_settings='{"foo": "x", "apple": 1}'
thta's not friendly
vale ibex
yeaa
there is parse_obj
which we could pass in the output of the current get_config()
which merges the toml files
so we could use a combination of toml and env vars
hmm, still not the best if you need to overwrite a value of a sub model via env var
patent pivot
hmmm
vale ibex
we could flatten the config into 1 class
and prefix certain things that need it
I'd prefer that to json encoded env vars
fallen patrol
`export my_prefix_more_settings='{"foo": "x", "apple": 1}'`
you don't need that lol
ah
nvm
torn idol
austere hornet
Do you get the contrib role after making at least 1 contribution to one of the open-source projects or is there a set number of contributions that you have to reach to get it?
brazen charm
austere hornet
Thx
clever wraith
Hey all, sir-lancebot#867 is ready for review!
dusky shoreBOT
fallen patrol
@clever wraith https://pythondiscord.notion.site/Squashing-a-Pull-Request-191e8efa2fbb4d72840fd17ca37cf7fa 
hmmmmm
I have some other things on 867 to review
clever wraith
i rebased
patent pivot
we can squash at PR level, which for a PR implementing a single feature isn't a bad bet
clever wraith
oh i already squashed :P
patent pivot
yea all good - just saying that's an option for those that don't want to get into deep git for really small PRs
clever wraith
kay
fallen patrol
we can squash at PR level, which for a PR implementing a single feature isn't a ...
ah, i thought you had that turned off
which the guide above literally says 😛
Since squashing a Pull Request is an exceptional operation, we don't allow squash merge, you will have to do it from the command line, let's jump right into it!
patent pivot
which the guide above literally says 😛
> Since squashing a Pull Request is an e...
probably written at a time where that was policy, for small PRs like this that modify a file or two I'm fine with it, I'll make a note to update our docs at some point
don't have notion on this machine
yet
patent pivot
uh
not to my knowledge
let me actually see where this page is hosted in terms of on notion
patent pivot
i'm not sure whether we set aside an area on notion for this
clever wraith
For what
patent pivot
hmmmm so
patent pivot
<@456226577798135808> https://pythondiscord.notion.site/Squashing-a-Pull-Request...
this document is part of a 3 page collection we have under our dev domain
code review guide, squashing and then I think an internal poetry vs pipenv page
code review guide got migrated to site, so I guess at some point the plan may be to convert this to a guide, it's not a part of anything else I think we just drafted it on notion to get things together quickly
green oriole
I wrote that as an internal reference, but it will probably need some polish before becoming an article
Hey @cold island @vale ibex, so what was the issue?
cold island
Hey <@512354988157173763> <@126811506632294400>, so what was the issue?
Some file conflict that made the git pull not work well
But we're now trying to understand how to allow development with metricity on a different server. Since that's the whole point of the changes
The issue being that you need to change metricity's config from the bot
vale ibex
toml files
with a default & override
with some values being pulled from env vars
I think the consensus atm is to migrate metricity to pydantic
We still need to discuss it a little more, since we need to figure out a nice way to deal with it
see my conversation with joe last night
yea, scroll up like 2 pages
green oriole
toml files
Why don't we bind mount a config file into the container?
green oriole
Hmmm
What about mounting the current folder, since it will always exist? Then we will be able to look whenever there is a config file on that folder
cold island
Mount the bot folder in the metricity container?
green oriole
Yeah
We could have the entry point try to move the config to the right place
So it is transparent for the app
cold island
Not sure how to do that. Maybe Chris knows
green oriole
We could have compose mount . to /tmp/bot and add cp -f /tmp/bot/metricity-config.toml ./config.toml to the entry point
Well, at least that's the rough idea
cold island
cc @vale ibex
gritty wind
Eh the whole idea feels like a lot just to set one variable
I like the idea of pydantic (totally unbiased), so if we’re considering doing that anyways
Might as well save the extra effort
We’re in no rush either, things work as is, the only reason we need this is for people that want to run metricity
So basically only people who are working on very limited features
green oriole
I'm not a pydantic pro, how will that help with configuration?
gritty wind
Well basically we already have it working cleanly with env vars
What Chris and I were working on was just getting that env var to the right data type since meta classes do not type casting at all
Pydantic on the other hand
Does
Unrelated, but it would actually be great if we could just drop the toml configuration. I dislike it in all projects we use it in.
For most of our projects it’s literally just forcing us to define the same content in two different spots. The only project it sort of makes sense for is bot, because it makes it easier for people to clone and change the variables they need
But even bot can just be completely circumvented if we have smart config
Which is, ya know, one day™️
green oriole
I mean, we could roll smartconfig to more projects than the two bots
gritty wind
What other projects need it?
Isn’t smartconfig creating a server and setting the IDs?
green oriole
I mean, any project that uses a large config file can benefit from it
The core of smartconfig is about having a modular config system
Then there will be an extra to add a discord integration
gritty wind
I don’t follow, what would a modular config look like
green oriole
Don't mind me, I'll copy paste my message from the staff channel so it is publicly available
So, smartconfig. You might have heard that name here and there, the goal of this library is to make our whole configuration system easier to use.
Right now, we rely on this huge default-config.yml file, which has more than 500 lines or so. Additionally, all of those constants are repeated in a constants.py file for typehinting sakes. This is far from ideal, and we can do better.
Smartconfig will allow us, instead of having centralized constants, to move them to their parent cog if you will. Additionally, smartconfig will automatically parse your configuration file and allow you to override constants using the module path.
I know this is quite confusing at first, so here is an example: everyone's favorite cog, DuckPond ! It has a related constant, ducky_thresold
# file: bot/exts/fun/duck_pond
class Config(smartconfig.ConfigEntry):
ducky_thresold = 7
```and `Config.ducky_thresold` can be used like normal. But here is the cool part! Imagine I have a config file:
```yaml
# file: config.yaml
bot.exts:
fun:
duck_pond:
ducky_thresold: 1
```then the constant will be overridden.
The advantage of this are:
- It is clear what constants belongs to where
- You don't have a massive and daunting config file
- You only override what you need to override
- Pulling changes isn't a mess
And more importantly, we now have easy to parse configuration classes, which will allow us to create a test server and fill up the configuration file. All of that automagically with a single command! At least that's the plan There, so everyone can follow
So like here you have a small configuration entry for the duck pond cog
You could imagine having different entries (you could also call them units or modules) for the different parts of your app
Or even just have a single constants file, but you can override it out of the box
Anything is possible, really
gritty wind
I don’t think that’ll be too helpful for the other projects that use toml configs, but it seems cool for bot
cold island
Hmm how do the config classes help with creating a test server
green oriole
It is a lot easier to attach metadata saying "this should be that emoji" for instance
It isn't required per se, but it will be much easier to do both at the same time
vale ibex
Yea we could mount the whole bot folder, it just felt wrong lol
cold island
Honestly for dev purposes it seems the simplest and most flexible solution
vale ibex
we also lose the ability to rename the file, IE we can't do ```yml
volumes:
- ./metricity-config.toml:/metricity/config.toml
so it would need to be named config.toml or config-default.toml
vale ibex
wdym?
cold island
If you have cp then I guess you can check if the file exists?
vale ibex
I guess we could wrap it in ```bash
if [ -e /tmp/bot/metricity-config.toml]
then cp -f /tmp/bot/metricity-config.toml ./config.toml
fi
lol ye
it seems less effort than migrating metricity to pydantic lol
cold island
haha
green oriole
I wonder what cp behaviour is if the file doesn't exist
If the force flag is set it won't error out if the file doesn't exist
Don't ask me why
vale ibex
lol how odd
I used the if statement anyway so that I could echo something conditionally
ok that should be it @cold island
I've force pushed some changes, so I'm not sure if a git pull will be fine, or if you need to git reset
someone with more experience of git can answer that 😄
remote wigeon
git pull --force should do the trick
vale ibex
nice
remote wigeon
👍
molten perch
Hey! Can I get a review on sir-lancebot#857 and maybe on api#16 ? 🙂
dusky shoreBOT
clever wraith
Hey @patent pivot https://jb3.dev seems to be offline
patent pivot
hmmm working for me?
what do you see
also this is for pydis projects lol, better pinging in off topic
clever wraith
Oop
clever wraith
sir-lancebot#867 needs reviews! Should be quick, it's a 7 line file 😄
dusky shoreBOT
clever wraith
Yeah
vale ibex
@cold island
cold island
Hmm our site uses django-simple-bulma, which says bulma-collapsible is available, but I can't seem to get it to work
For reference I tried copying this snippet into one of the pages https://bulma-collapsible.netlify.app/usage/#collapsible-card
Ahh, I think it's just not enabled
And with the extension it doesn't un-collpase
@crude gyro if you're available maybe it's something you're familiar with? I did see an open issue in the repo related to collapsible
crude gyro
@cold island make sure the extension is actually turned on in settings.py
there's a config dict where extensions must be added for them to be available.
cold island
Yeah, without the extension the card doesn't collapse, with the extension it doesn't un-collapse
green oriole
It would be a dict move to not add it
dim pelican
Hello github issues page
Include me in the screenshot
molten perch
I don't think that's a good idea, if it's not a valid Cog or tag, the bot should give some feedback about it.
I see, but it's not a valid Cog or tag, regardless. 🙂
austere hornet
Hey @vocal wolf, could you please review sir-lancebot#843 when you have a moment? I will greatly appreciate it. Thanks so much!
austere hornet
Ok, what do you suggest instead?
dim pelican
json?
austere hornet
Can you link pls?
dusky shoreBOT
clever wraith
@vocal wolf could you please re-review sir-lancebot#849
vocal wolf
@austere hornet @clever wraith I shall.
austere hornet
<@!853309207922212965> <@456226577798135808> I shall.
Thank you!
dim pelican
any recommendations for sir-lancebot#859? I tried the file preview thing with the .md and .txt file creation but that didn't work. Anything left to do for it?
dusky shoreBOT
austere hornet
Love the reasoning in your PR description @dim pelican lol
dim pelican
Love the reasoning in your PR description <@698273827448291379> lol
Lol, I took over the previous PR with this one
And copied their reasoning too
austere hornet
Lmao
patent pivot
btw this is a thing now
green oriole
You know I cannot PR at that time of the day haha
patent pivot
true, true
green oriole
But with some in-line html you can do some cool stuff
patent pivot
yea
brazen charm
isn't the logo enough?
green oriole
!remind 10h get those swag points
stable mountainBOT
green oriole
I feel like having a banner would look cool
patent pivot
we could remove a bit of the blurb and add some action items
visit webpage, join discord, read contribution guides, etc
green oriole
Huh, it isn't python-discord/python-discord
Ya ya
Buttons would be nice too
Lmfao @patent pivot github just published the changeling for the org readme
How fast did you added one
green oriole
Oh really, was it not yet released back then
green oriole
Aaaaah
brisk brook
Hey <@196664644113268736>, could you please review sir-lancebot#843 when you hav...
Aaahhh, I have some "unpushed" comments for this one I think.
One sec
brisk brook
Tl;Dr it's all about capitalization today
austere hornet
Tl;Dr it's all about capitalization today
Ah
brisk brook
Okay that should be submitted
patent pivot
lol
fallen patrol
lol github revealed some things in that screenshot they probably weren't supposed to
stable mountainBOT
!remind 10h get those swag points
It has arrived!
brisk brook
4.9k repos
Yeah? It's a big company
remote wigeon
damn 1.8k teams
vale ibex
Hey @gritty wind was the idea of https://github.com/python-discord/sir-lancebot/pull/578/files#diff-2272f4f1bd8af0ac66cdd3a1c7e0428d97be9bc4d41fa5065a36d7251452be48R281 to bypass all checks entirely?
IE including the global check
gritty wind
Yeah?
vale ibex
I see this passes it through to the whitelist_check, which removes all the kwargs for when it hit in_whitelist_check
but since channels, categories and roles are all empty, that means that it'll always be denied right?
since the global check still applies
gritty wind
I don’t remember much of the code rn, I could look at it in a bit
It’s entirely possible I missed testing it
vale ibex
alr, just mentioning it since I was reviewing sir-lancebot#856
dusky shoreBOT
vale ibex
and saw that it's already pretty much implemented
gritty wind
Lol I was wondering why that issue was taking so long
green oriole
!remind 6h 
stable mountainBOT
gritty wind
Right, so now I remember. The idea was to allow you to block it even in the globally allowed channels
Not the other way around
gritty wind
Wording should be fixed
gritty wind
Well, maybe, but I think it would allow you to add a channel that isn’t in the whitelist, and only that channel
Same for roles
vale ibex
ah yea, it resets first, then iters
gritty wind
So I can enable this channel, but disable OT
gritty wind
Could also update it to have both behaviors
vale ibex
since I'm going to suggest we call the kwarg in sir-lancebot#856 bypass_all_checks rather than every
dusky shoreBOT
vale ibex
same 😄
vale ibex
Could also update it to have both behaviors
how would that work?
fwiw bypass_defaults isn't used anywhere rn 😦
gritty wind
Actually yeah never mind I’m misunderstanding the intention here
Yeah, this was created as mainly a util I aim to port to bot-core at some point
So it has some redundancies
gritty wind
Is this being implemented on the wrapper or the actual check function
It should probably be implemented on the wrapper to avoid unnecessary overhead
vale ibex
ah yea seems like it's in the check
gritty wind
It doesn’t seem to be?
Actually I know what’s happening
Confusing wording again lol
Wrapper I'm referring to is whitelist_check
gritty wind
check is in_whitelist_check
vale ibex
I was referring to the wrapper whitelist_override
gritty wind
We could probably add it there too using a lambda or something
vale ibex
I think we implement it the same as bypass_defaults is
gritty wind
Yeah that's probably the most reasonable
vale ibex
have it a a kwarg on whitelist_override and then getattr in whitelist_check
gritty wind
This could actually be easier solved by using the everyone role, no?
No actual changes to the wrapper or anything lol
gritty wind
It also means we don't get DM invocs
vale ibex
@green oriole I've noticed that my bot is getting gateway events for messages even if the bot isn't in a thread
So doing that thread join for every new thread isn't needed
but could be useful to just have the bot in the member list
not sure how many api requests it'll end up being joining every single thread automatically
green oriole
Huh, that's interesting
I use that on Python to keep track of what new thread we logged or not
In term of rate limits it is impossible we reach it unless we get raided
vale ibex
Yea, I don't think it's an issue, just pointing it out
green oriole
That's good to know, yeah
cold island
Reviews on bot#1793? 
dusky shoreBOT
cold island
Also, does someone know of an easy way to create collapsible elements on the site without bulma?
austere hornet
sir-lancebot#843 needs reviews as well. At this point it would be best for contrib/core dev reviews since there weren't any yet. Thanks again to @clever wraith, @dry folio, @brisk brook, @dim pelican, and @vale ibex for your reviews!! They were really helpful to me and Fronto :)
trail pilot
Also sir-lancebot#860
dusky shoreBOT
static canyon
@trail pilot would you like nitpicky comments as well?
Like # there are --> # There are (capital)
Just keeps your code consistent with the style of the bot
clever wraith
sir-lancebot#843 needs reviews as well. At this point it would be best for contr...
no problemo
trail pilot
<@!584817706646110333> would you like nitpicky comments as well?
Sure
clever wraith
Hey there everyone, does anyone have a gitpod account? I'd like someone to test the .gitpod.yml
If so, please test it at https://gitpod.io/#/github.com/kronifer/sir-lancebot/tree/gitpod
trail pilot
Just submitted 👍
Are you the tooboredtocode person
static canyon
Are you the `tooboredtocode` person
No, TizzySaurus
trail pilot
Ah, it’s on Hangman?
static canyon
Challenges
static canyon
I may have fked up
Do you see it in #dev-log? @trail pilot
Try clicking on the link to it
static canyon
Where? Because when I looked around lancebot there's basically none
thorny obsidian
Our style guidelines and PEP8
Trailing commas are a good practice to make sure the git blame is accurate
static canyon
Trailing commas are a good practice to make sure the git blame is accurate
But basically nowhere uses them currently
Other than when it's making a datatype into a tuple e.g. aliases=("x",)
thorny obsidian
But basically nowhere uses them currently
So? I don't think that's justification for not following good practice
static canyon
I thought we had a thing of consistency was more important
green oriole
I am honestly thinking about suggesting to remove that rule because of consistency
brazen charm
consistency is important, but commas aren't going to make things confusing on which to choose and there is an actual non style benefit to it
static canyon
I am honestly thinking about suggesting to remove that rule because of consisten...
Wdym?
thorny obsidian
If it's an area where it might make sense in the future to expand on it, then I think trailing commas are a good practice. Sir-Lancebot is also a learning repo so I'm hesitant about opting for consistency over good practice to make it more "consistent" because that very well could mean making things consistent with bad practice
static canyon
I'm not strongly against trailing commas myself, just thought we had a thing of not doing them
thorny obsidian
Maybe we do? But you'd have to point me to that discussion
static canyon
just thought we had a thing of not doing them
By this I just mean "I've seen people say to remove them"
dim pelican
I think its not using trailing commas in a function definition right? Since they are usually not going to add additional arguments
static canyon
e.g. here (this is just the first one I found)
thorny obsidian
Well, that makes sense because it's not correct usage of a trailing comma, if you're referring to the decorator
static canyon
Well then what defines as "correct usage"?
thorny obsidian
Having a trailing comma on a single line does nothing to help with git blame. The trailing comma is used when you split it up on multiple lines, because adding an element will only assign you the blame for the single line you changed, not the line previous as well (when you have to add a comma)
static canyon
So do you always add a trailing comma when it's over multiple lines, or would you leave it out p.e. in a function defintion since you're unlikely to get extra added arguments (what brad asked above)
thorny obsidian
When trailing commas are redundant, they are often helpful when a version control system is used, when a list of values, arguments or imported items is expected to be extended over time.
dim pelican
So do you *always* add a trailing comma when it's over multiple lines, or would ...
Now I'm questioning myself, what if there is a lot of args and split across newlines?
static canyon
is expected to be extended over time.
So if it's not expected to be extended (like a function definition), it shouldn't have a trailing comma
thorny obsidian
For instance, let me pull an example from our AoC cog
dim pelican
> is expected to be extended over time.
So if it's not expected to be extended ...
Oh, confidence returned lol
brazen charm
There's nothing to lose if the comma is included, and even if you don't expect it at the time of writing, functions are commonly extended
static canyon
I think in the case of this PR it makes sense for the list of valid languages, since that will change as codewars add more languages
But it doesn't make sense in the function definition because that won't change
thorny obsidian
https://github.com/python-discord/sir-lancebot/blob/main/bot/exts/events/advent_of_code/_cog.py#L167
This has a trailing comma because it's not unlikely that we'll add more information
The function definition is a single line, no?
static canyon
stable mountainBOT
bot/exts/events/advent_of_code/_cog.py lines 167 to 172
info_str = [
"To join our leaderboard, follow these steps:",
"• Log in on https://adventofcode.com",
"• Head over to https://adventofcode.com/leaderboard/private",
f"• Use this code `{join_code}` to join the Python Discord leaderboard!",
]```One of the __init__ is multiple lines
thorny obsidian
aaaah, okay
static canyon
https://github.com/python-discord/sir-lancebot/blob/main/bot/exts/events/advent_...
And yeah, that makes sense. Just like the example I gave above of the languages that codewars supports
thorny obsidian
But I'm also with Numerlor in that it doesn't hurt much. So if it's there I'd leave it. If it's not *shrug*
static canyon
I think if we're going to have a "rule" we should stick to it
If it's likely to be changed (or is being used to make a tuple), including a trailing comma, otherwise don't
thorny obsidian
Now, where the bike shedding enters: how do we determine likely?
dim pelican
Add black to the ci and try to survive?
static canyon
Now, where the bike shedding enters: how do we determine likely?
Just a scenario where you can see it changing, like the supported languages etc.
thorny obsidian
But I don't have a horse in this race, I just like trailing commas and will generally advocate for them where they make sense~
static canyon
will generally advocate for them where they make sense
Do you think it makes sense here?py def __init__( self, language_embed: Embed, tags_embed: Embed, other_info_embed: Embed, main_embed: Embed, ): ...
I'm assuming that wouldn't classify as a "logical" place to have it
static canyon
Add black to the ci and try to survive?
bot#1773 is kinda this but went stale 🤷
dusky shoreBOT
thorny obsidian
you keep mypy out of lancebot
static canyon
lol
dim pelican
<:spray:703023678497095700> you keep mypy out of lancebot
But I don't have a horse in this race
lol
I haven't used mypy very much, is it a general formatter?
thorny obsidian
ooooh... hurmmm...
Okay, this is for the Information Dropdown, which I can imagine being changed if we want to add other embeds / more info in the future
thorny obsidian
I haven't used mypy very much, is it a general formatter?
static type checker
static canyon
Kat said taht we should use trailing commas tho?
So @trail pilot I think my conclusion on this is that keep the trailing commas where you think it's likely to change in the future (e.g. the lists of supported languages) but remove elsewhere (such as the SelectOption calls)
thorny obsidian
give me true dynamic typing or give me death (not really, I just think mypy would be worse for 1st time contribs and lance is aimed for 1st time contribs it's not appropraite)
brazen charm
> will generally advocate for them where they make sense
Do you think it makes s...
It may not be logical for the blame/diff reason, but like I said previously there's no downside to including it in there. I wouldn't introduce a commit to add the comma but I also definitely wouldn't introduce a commit to reverse it
static canyon
I mean I don't think of it as adding a commit to reverse because I'd just squash it but yeah, I do agree
thorny obsidian
In that specific case of __init__, I can imagine that changing in the future since it's adding new/more embeds to the dropdown
static canyon
I mean yeah, I guess I can see that potentially happening
SelectOption(
label="Main Information",
description="See the kata's difficulty, description, etc.",
emoji='🌎',
),```you wouldn't add a field to these though(?)Yeah, doesn't look like it
clever wraith
Hey @brisk brook, I'll add the log.error after school. GitHub mobile is having a bad time at the moment so I'm responding here
brisk brook
Are you the `tooboredtocode` person
@remote wigeon
brisk brook
I haven't used mypy very much, is it a general formatter?
Bad static type checker
remote wigeon
<@!244459328847872000>
yeah thats me
brisk brook
Pyright is better imo
dim pelican
Hmmm. What are the choices for type checkers? I guess I should do my own research but curious what the people think in this channel
brisk brook
Pyre, Pyright and Mypy
You can also use Cython, but that requires cythonizing your code
patent pivot
@thorny obsidian do we have plans to revert code jam on frontpage?
thorny obsidian
No because it totally slipped my mind. Will make a PR for it today
patent pivot
👍
clever wraith
Pyre, Pyright and Mypy
What about pydantic
brisk brook
That's not a static type checker
It enforces types at runtime
Not statically, that's the difference 😉
clever wraith
Ah
surreal venture
Wait now that discord has released replies and threads, would it be a good idea to change spooknamerate game to use threads?
sir-lancebot#509
dusky shoreBOT
surreal venture
Now I realise that it is in serious need of updating
stable mountainBOT
!remind 6h <a:partygopher:472490455096098827>
It has arrived!
thorny obsidian
Wait now that discord has released replies and threads, would it be a good idea ...
I'd be down to see what that can look like with threads! (but also open an issue and get coredev approval >_>)
surreal venture
alr
dim pelican
Here to learn forms
thorny obsidian
So! Forms! I'd like to use forms for our upcoming esoteric challenges (cc: @placid ermine) and also get the full functionality for future code jams.
placid ermine
i will watch your progress with great interest
thorny obsidian
@gritty wind is our fearless leader for forms, but joe and Akarys also help out a bunch. I think we may have conned Blue into helping as well.
dim pelican
Interesting. I must say that I would be starting at 0% for TypeScript as well as Web Apps in general
thorny obsidian
I will also be starting at 0% as well, but I do plan on contributing to forms :D
But I believe we have a few issues to conquer regarding forms which should be ... fun? definitely intersting
dim pelican
Sounds a lot like my manager saying "We have a great opportunity for you"
gritty wind
Hey hey brad - you may be currently unable to contribute because the issue board and setup guides aren’t public yet (it’s on my agenda I swear), but in the mean time, it’s probably best to get familiarized with TS and react from a basics perspective
dim pelican
I do agree with that statement lol
gritty wind
The contributing guide may be public, my memory is not 100%
thorny obsidian
But looking to future contribution stuff, forms is going to be a top priority for me in the coming months
dim pelican
Sounds pretty neat, looks like I have a good bit of projects to learn with! Besides sir-lancebot I'm trying to help out the catcord peeps, and now forms
thorny obsidian
Hey hey brad - you may be currently unable to contribute because the issue board...
let me know if there's anything I can help with regarding that~ Code Jam is wrapped up so I've got some time before I refocus
gritty wind
!remind 23H Forms stuff ^
stable mountainBOT
molten perch
Now that you're talking about projects, may I ask what's the situation of the API migration? 🙂
gritty wind
I think that’s been a little slow because Ves was busy with real life (that makes two of us!)
But I’m not too sure
thorny obsidian
I honestly don't know. Uhhhh @last patio and @hardy gorge would know iirc
remote wigeon
if you need help with frontend dev i can help as well
gritty wind
I’ll be making the entire repo public eventually, probably will put out a call to action then
Err, the repo is actually public, issues are not
thorny obsidian
(I think I'll be helping Scale do some of the converting)
hardy gorge
I've unassigned myself from the one ticket on API I was assigned to, but if no one else has taken it in the mean time, I'm going to work on it again this weekend
molten perch
I've unassigned myself from the one ticket on API I was assigned to, but if no o...
The Alembic setup needs the models to be set up I believe. I've opened an issue regarding that matter. However, it hasn't been approved.
hardy gorge
We can probably configure it with only the declarative base of SQLAlchemy, although that is not very useful in itself
I'll see if I can run the codegen for the models this weekend
Although I guess anyone with the Django site setup could do it
surreal venture
Also, would it be a good idea to migrate the pagination to use the new buttons feature instead of reactions?
molten perch
I'll see if I can run the codegen for the models this weekend
You mean with SQLAcodegen, as suggested? I believe it has to be reviewed throughly though.
(Also I think I might be able to do it if it gets approved, just in case)
hardy gorge
It will probably generate some bare Table definitions instead of Models as well
I'll approve the issue; I think it's an approach worth trying
sour viper
? Isn't tomorrow Thursday ?
dim pelican
Yes, unless you are 12 hours ahead and it is already Thursday
Or their weekend is Thursday lol
sour viper
I would joke that this isn't true and earth is flat and time is same everywhere, but people may actually take it seriously
molten perch
I'll approve the issue; I think it's an approach worth trying
Thank you, will you also assign me? 🙂
gritty wind
? Isn't tomorrow Thursday ?
Weekends here start on Friday 😃
fallen patrol
<@!407110650217627658> I've noticed that my bot is getting gateway events for me...
Yeah, this is in the discord docs lmao
thorny obsidian
The repos have always been public, we're working to migrate the guides & issues to the repos
austere hornet
My PR like ^
Lmfao
thorny obsidian
we're working on making it open for contributions from non-staff
Because it was an internal project and we didn't see a need to at the start when it was first getting off the ground.
celest charm
give me true dynamic typing or give me death (not really, I just think mypy woul...
Why do you like dynamic typing?
sorry, I reread your message and understand the point
If we don't want beginners to care about the type hinting, why do the linting rules require type hints on all public functions? Beginners will (and do) use the wrong type hints, which will hurt readability, not assist. So maybe we drop this requirement from lancebot?
thorny obsidian
True, I'd be fine with not type hinting all public functions. But that's going to be a core dev and admin decision.