-
-
Notifications
You must be signed in to change notification settings - Fork 666
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Update Dependencies #196
Update Dependencies #196
Conversation
I'm still not sure entirely why were are using requests instead of aiohttp which we use for all web requests in the rest of the bot. Feels like we should only need one of them. Any chance we can refactor some code to get rid of it altogether, or would that be a large enough job to warrant a separate PR? |
I'm pretty sure we are only using it for intersphinx and specifically an error which intersphinx might raise (iirc it's only imported once). |
Yeah, that's why. I should probably remove that somehow. |
What exactly went wrong there? |
Like I said, this was happening before updating any dependencies so it's unrelated. I just wanted it to be clear I cannot properly test this command in case there is something else that would break once this is resolved. |
@gdude2002 Were @scragly's comments regarding the git dependency addressed? |
@MarkKoz looks like you have some example entry in your local database with an invalid entry. can you remove that and retry? |
@jchristgit Ah yeah, it was the illustrious |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good.
There's still some discussion points that need to be resolved:
|
Why not?
I'll take care of requests in a separate PR; I have a few issues with setting up the bot locally currently. |
@gdude2002 had something against it though he hasn't gotten back to me on that. |
Regarding the git dependency, @scragly made a good point yesterday in the dev-contrib channel about how any breaking change to the discord.py API (I believe cogs was used as an example) will cause us to have problems when we start auto-deploying the new version. AFAIK there is no way to pin to a specific ref with a zip. I think that the risk of deploying a breaking change is not one we should be taking our chances with. |
I'm in favor of switching out for a git dependency. I think the issue about it being "harder for beginners" is not a really valid thing here. Most of our contributors will already have git installed (to clone the repository) and there was also a mention of it making web editing harder (?) which wouldn't really be valid because you don't download dependencies in the GitHub web editor. @python-discord/staff if any of y'all have an opinion which is against adding a git dependency please state it now. |
Build is failing because it's now linting discord.py's source. It's in For what it's worth, running the linter locally yields no errors. |
|
It isn't linting |
We can either specify the directory to checkout editable dependencies into with |
We could use |
What is the cache directory? |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
👍
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
looks good to this guy
Closes #99.
discord
todiscord-py
.discord-py
was changed to a git dependency.multidict
andsympy
from Pipfile because they weren't explicitly being used anywhere.websockets
andyarl
from Pipfile because they were only there to circumvent previous dependency issues with discord.py.requests
from dev packages to "normal" packages because the docs cog imports it.flake8
was updated to 3.6 and all new lint errors were fixed.Things seem to still work fine, though I did not rigorously test every single command in the bot. One thing I couldn't test is the eval command. Also, I had trouble adding new docs (
!docs set
). However, both of these issues were present for me before these changes.