Poetry?

[ryansurf]

I haven't used poetry before, but it seems like it would make setup/maintaining the project more simple compared to pip... seems easy enough to set up. Maybe convert to using it?

@K-dash thoughts?

[K-dash]

@ryansurf
I agree:)

By introducing Poetry, the dependencies will become clearer, and we can separate the libraries used in production and development environments, which I think is a good idea.

[K-dash]

@GUKWAT
Oh. Cause identified.

This is because the definition of API_KEY of GPT is missing in .env.

Normally, you need to register OpenAI's API_KEY here, but if you don't use GPT, you can set API_KEY in .env as follows, and enter appropriate value here, and the problem should be solved.

API_KEY="tmp key"

[K-dash]

Or you can follow the instructions in README.md, clone the latest repository, and copy it with cp .env.example .env .

This commit requires API_KEY in .env.

[ryansurf]

Hey @GUKWAT, sorry about that. I worked on adding an openai GPT and accidentally didn't set a default value for the api key in settings.py, which would have avoided this error (settings.py should have gave the API_KEY value a default value if missing from the .env file).

I was a little hasty in my work. Going forward I'll start testing more thoroughly before I push any code. Hopefully you've gotten it working, and thank you @K-dash for figuring out what was going on

Let me know if you've gotten the server running! I'll be here if you have any questions/more errors 😳

Fix implemented in c75b75f

Thinking about this, maybe I should also be submitting PRs to be fully transparent about what I'm changing? Obviously, I've introduced some bugs that should have been prevented before pushed (like this). What do you guys think?

[K-dash]

@ryansurf
Regarding this discussion, since Poetry has already been introduced, it might be better to close it:).
If it's alright with you, I would be very grateful if you could mark my reply as the answer and close this discussion!

[ryansurf]

@K-dash will do 😃

[K-dash]

@ryansurf

Also, have you heard of a tool called Rye that has recently come out? It's a tool that combines both Python version management and pip library management.
However, since Rye is described as an experimental attempt, it might be better to introduce the more stable Poetry.

https://dev.classmethod.jp/articles/get-start-rye-python/

[ryansurf]

Haven't heard of this but It seems like a good option, ships with ruff too! But yeah, poetry may be the move for stability reasons. Gonna sleep on this one

[GUKWAT]

@K-dash @ryansurf
Finally!!! I was able to run the server.py without having any errors now I can finally start working on issues and making code contributions.
At first I added API_KEY="tmp key" to the .env file but I was only able to run the app in my commandline.
I kept on having file path errors when I tried to run server.py
In order to make server.py run I ended up pip installing linux and installed docker.
I think you are working on this project using Mac OS. I am working on a windows machine which is why I was having issues when it comes to the file path

[K-dash]

@GUKWAT
Welcome back!
I'm glad you were able to get it up and running successfully😆

There have been a lot of policy changes over the past week, so I think you should check with @ryansurf to see if you would like us to start on any of the issues!

[ryansurf]

@K-dash @ryansurf Finally!!! I was able to run the server.py without having any errors now I can finally start working on issues and making code contributions. At first I added API_KEY="tmp key" to the .env file but I was only able to run the app in my commandline. I kept on having file path errors when I tried to run server.py In order to make server.py run I ended up pip installing linux and installed docker. I think you are working on this project using Mac OS. I am working on a windows machine which is why I was having issues when it comes to the file path

Great! Actually for the API_KEY you can leave it blank like API_KEY=. See the .env.example file, you can leave it as is!

And yeah, you're correct. I work mostly in Linux/Mac, sorry about the issues using windows. My main PC is running windows too, I'll see if I can replicate a path error.

Sorry for not being as active in the past few days, life got a little busy. This weekend I'll have lots of time to work and help you debug if needed!

[K-dash]

@ryansurf
I've noticed that our current server.py, send_email.py, and cli.py files require careful consideration of relative paths and execution directories when importing modules.
This is because the PYTHONPATH is not properly set. As a result, we're experiencing some issues.
For example, we currently need to run python3 server.py from the /src directory, but when running test_server.pytests, we encounter ModuleNotFoundError.
To address this problem, I believe we should prioritize the introduction of Poetry. Poetry allows us to define a pyproject.toml file where we can specify the project's root directory as part of the PYTHONPATH.
This means we won't have to manually set the PYTHONPATH environment variable every time we run the application!
I think it's a good idea to tackle this before diving into writing more test code. I'll start working on integrating Poetry into our project.

[ryansurf]

This is necessary, I kept getting annoying path errors when messing around with code. Reviewing the PR now

[K-dash]

@ryansurf
(This is just a casual chat, by the way! 😎)
So, I found out about this Python project management tool called hatch yesterday.
It's an all-in-one tool that handles Python version management (which Poetry can't do), dependency management, and even provides built-in commands for linting, formatting, and testing!

I thought I'd give it a try myself, so I wanted to share it with you~

[ryansurf]

Seems interesting, I like the simplicity of an all in one tool. I'm still getting used to Poetry at the moment but perhaps I'll give this a go after. I've read a few of your writeups on your website, you know your stuff haha

Thanks for the recommendation 😳

[K-dash]

I think it's good, Also, thanks for taking the time to read my articles! hehe 😎