During this week, I started working intensively on GitHub on my Repository-Context-Packager project. This is a command-line tool that analyzes a local Git repository and generates a single, formatted text file containing repository context. This makes it easier to share your project with Large Language Models instead of copy-pasting files one by one. I believe most of us have struggled with this while working on LLM as you ask help on a file and it tells you that you need to upload the rest of the project files so it can better assist you.
Why I built this
As mentioned earlier, we can have a hard time working with AI to quickly provide enough information about a repository as it will require us to upload files one by one which is highly demanding and can be frustrating for large projects. This is where Repository-Context-Packager comes in as it enables us have a quick summary of a repository, filter which files we include and save the results in a single text file which we can share with a Large Language Model.
I had to collaborate with classmates working on the same project and answer the below questions:
How did you go about doing your code reviews? Do you prefer an async or sync approach? Why?
The code reviews were done asynchronously. I prefer it because i currently struggle between work and school which makes me have a tight schedule and i donot have availability after classes or on some early hours as other students do.
What was it like testing and reviewing someone else's code? Did you run into any problems? Did anything surprise you?
It was interesting to look at another persons code. I enjoyed the fact that the language was different and I could see how they approached certain things in a different way to lead to the same results. Yes, I ran into problems as the readme.md file wasn't clear enough to guide me throughout the installation till testing process.
What was it like having someone test and review your code? Were you surprised by anything?
It was good as sometimes we can't see what we miss until it is brought up by another person.
What kind of issues came up in your testing and review? Discuss a few of them in detail.
The issues that came were the structure not correctly outputed when we run some files, Git info which is not captured correctly when passing a filename and the optional include feature which works with --include and not -i.
Provide links to issues you filed, and summarize what you found
Issue 1
Issue 2
Issue 3
Issue 4
For this project I forked, i noticed it missed instructions on how the project was to be run; optional features were missed as well; some header files were included in main.cpp which didn't exist in the project and error handling was missed as well.
Provide links to issues that were filed on your repo, and what they were about
Issue 1
Issue 2
Issue 3
They were about the structure not correctly outputed when we run some files, Git info which is not captured correctly when passing a filename and the optional include feature which works with --include and not -i.
Were you able to fix all your issues? What was that like?
I haven't started fixing them. I will focus on them next week as I have other assignments due this weekend.
What did you learn through the process of doing the testing and reviewing?
First, I had never really done it this way from GitHub. So I like the fact that it pushes me to learn how to use GitHub deeply, make researches. Also, I feel a bit more confident now than when i started this course as I was scared due to my little to no deep knowledge of this. I think it is a good point for Open source projects as I now know how to do testing, leave clear reviews on issues.
Top comments (0)