mpoon / gpt-repository-loader Goto Github PK

Most repos don't have .gptignore, ignore .gitignore if .gptignore is not present.

We want to add a new end to end test for gpt-repository-loader. The test should use gpt-repository-loader to process an example repository and ensure that the output is what we expect.

I have an internal library for counting tokens on files and directories, should probably opensource it.

Given an ignore file of

/this_directory

Git will ignore the directory, this tool will include it. It'd be rad if they were consistent.

Add a .gitignore file to our project.

to ease for hosted repos.

Add a permissive open source license to this project.

  File "gpt_repository_loader.py", line 30
    output_file.write(f"{relative_file_path}\n")
                                              ^
SyntaxError: invalid syntax

I followed the instruction but getting the above error when trying to run the command on my repository.

Rename git_to_text.py to gpt_repository_loader.py

Currently .tox is included into the output.txt. This is a lot of files.

We want gpt-repository-loader to accept a new command line argument that allows the user to override the default preamble text generated in output.txt before any filenames or file content is output.

I see you said that you manually created PR #17 from the output of ChatGPT.

What if we could automate PR creation? The only manual step would be to create a GitHub account for ChatGPT. After creating that, I guess we would need a script that:

• Gathers GitHub credentials from an account (e.g. ssh public key or key token).
• Uses a tool that can feed the output of GptRepoLoader to ChatGPT (example: AskGPT): https://github.com/praeclarum/AskGPT
• Tell ChatGPT to write his solution as a .patch file.
• Apply it via git am locally.
• Push branch to ChatGPT's fork in ChatGPT's Github account.
• Hit GitHubActions API to create PR.

I compared, zipping a medium size repo to representing it with one big file, as per this tool. I got repeated response generation errors. When I ask GPT-4 the meta questions about which was better, each method had some pros and cons, but it was definitely leaning towards zip.

Manually copypasting from output.txt is still work, auto copy the content to clipboard for easy pasting into ChatGPT.

• We want the user to be able to specify the output file path and file name when running the gpt-repository-loader command.
• There should still be a default output file if the user choses not to specify
• Make sure to update any usage text, helper text, and readme files so users know how to specify the output file path and file name

Update the README file with:

• A short overview of the purpose of gpt-repository-loader
• Usage information on how to use gpt-repository-loader aimed toward getting new users started quickly

Add a test harness to gpt-repository-loader

• Add a placeholder test that always passes
• Update the README to include instructions on how to run tests in gpt-repository-loader

GPT-4, can you help me with these items:

• what are some ways to compress the file contents without losing data so that the output file is smaller than the default output file
• can you choose the most promising of your suggestions and add a new option to allow users to compress the file contents
• update the README so new users know how to use this option

https://github.com/FocusedFidgeter/gpt-repository-loader/blob/31e6fa9f54e4f0302055363e9604724ca6e8d2a4/analyzer.py#L255

Add the following files to .gptignore

• LICENSE
• everything in the .github directory

Thank you for the wonderful tool!

Problem

When inputting to GPT-4 at once, I get the error "The message you submitted was too long, please reload the conversation and submit something shorter."

Details

In my fairly simple React app, when I output output.text, the number of lines reached 6 million, and of course, most of it was node_modules. When I excluded descriptions unrelated to the app, such as node_modules, yarn.lock, and .git/, it reduced to about 1,500 lines, but I still got the same error.

Now, I've declared to send one file at a time, and by repeatedly copying and pasting, it seems that the files can be read.

I will now type in the code for an application, file by file. The following text is a Git repository with code. The structure of the text are sections that begin with ----, followed by a single line containing the file path and file name, followed by a variable amount of lines containing the file contents. The text representing the Git repository ends when the symbols --END-- are encounted. Any further text beyond --END-- are meant to be interpreted as instructions using the aforementioned Git repository as context.

Suggestion

It might be useful to have features like the following:

• A function to exclude large code, binary code, and code unrelated to the app
• A function to divide text files according to the maximum character count of each GPT API, and add a prompt to that effect

By the way

With GPT-4, I could send about 400 lines at a time (4-5 files).
Also, every time I sent it, the contents of the file were properly explained, and when I sent it up to --END--, it even created a summary for me, which was impressive.

From now on, I would like to verify whether GPT can handle code maintenance and adding new features.

Will you sell your GPT-4-32k API? If you would like to, please contact me.