Sort requirements.txt based on package name only

[deepyaman]

Description

The current sorting algorithm uses the full line, instead of the package name. For example:

Current:

# Pin problematic traitlets release -  https://github.com/jupyter/notebook/issues/7048
ipython~=8.10; python_version >= '3.8'
jupyterlab_server>=2.11.1, <2.16.0
jupyterlab~=3.0, <3.6.0
jupyter~=1.0
kedro-telemetry>=0.3.1
kedro~=0.19.1
traitlets<5.10.0

Expected:

ipython~=8.10; python_version >= '3.8'
jupyter~=1.0
jupyterlab~=3.0, <3.6.0
jupyterlab_server>=2.11.1, <2.16.0
kedro~=0.19.1
kedro-telemetry>=0.3.1
# Pin problematic traitlets release -  https://github.com/jupyter/notebook/issues/7048
traitlets<5.10.0

I just noticed it also throws the comments at the top, which takes them out of context.

Development notes

Initially, I removed the sort_requirements function, but it seems the other starters also rely on it, so will leave it as a wrapper.

Developer Certificate of Origin

We need all contributions to comply with the Developer Certificate of Origin (DCO). All commits must be signed off by including a Signed-off-by line in the commit message. See our wiki for guidance.

If your PR is blocked due to unsigned commits, then you must follow the instructions under "Rebase the branch" on the GitHub Checks page for your PR. This will retroactively add the sign-off to all unsigned commits and allow the DCO check to pass.

Checklist

• Read the contributing guidelines
• Signed off each commit with a Developer Certificate of Origin (DCO)
• Opened this PR as a 'Draft Pull Request' if it is work-in-progress
• Updated the documentation to reflect the code changes
• Added a description of this change in the RELEASE.md file
• Added tests to cover my changes
• Checked if this change will affect Kedro-Viz, and if so, communicated that with the Viz team

[SajidAlamQB]

Initially, I removed the sort_requirements function, but it seems the other starters also rely on it, so will leave it as a wrapper.

You would need to update all the post_gen_project.py in the relevant starters.

[SajidAlamQB]

LGTM, thank you @deepyaman. 👍

[deepyaman]

Initially, I removed the sort_requirements function, but it seems the other starters also rely on it, so will leave it as a wrapper.

You would need to update all the post_gen_project.py in the relevant starters.

Yeah, and would need to do that before updating the code in main. But this works with more minimal changes. :)

[noklam]

Great catch!

[astrojuanlu]

Not extremely happy about adding a runtime dependency on pre-commit-hooks though. At least it's unconstrained, but (1) it's unclear to me if the project is meant to be used from its Python API (not that we're using a private function or anything, but I don't see references to it on the READMEs) and (2) this dependency only exists for kedro new and users are already complaining that we have too many of them, this seems to go in the opposite direction.

[deepyaman]

@astrojuanlu If you feel strongly against it, I can just copy the implementation of that particular function. I added this because pre-commit-hooks is an extremely lightweight library (only dependency is ruamel.yaml, which itself is very light).

[astrojuanlu]

@merelcht what do you think?

[merelcht]

I agree with @astrojuanlu, this sort of dependency isn't core to Kedro as a Framework and so it doesn't belong as a run-time dependency. If anything it should be something like a "dev" dependency.

[deepyaman]

@merelcht I don't think that's @astrojuanlu's point. I am using pre-commit-hooks as a runtime dependency here, because the starter requirements get sorted.

The question is whether to vendor the requirements sorting logic or leave this lightweight dependency.