https://github.com/keerthanpg/SwePT
AI sends pull requests for features you request in natural language
https://github.com/keerthanpg/SwePT
Last synced: about 1 month ago
JSON representation
AI sends pull requests for features you request in natural language
- Host: GitHub
- URL: https://github.com/keerthanpg/SwePT
- Owner: keerthanpg
- Created: 2023-02-25T20:45:56.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2023-03-25T21:03:43.000Z (about 2 years ago)
- Last Synced: 2024-02-13T02:51:38.579Z (about 1 year ago)
- Language: Python
- Size: 14.6 KB
- Stars: 111
- Watchers: 6
- Forks: 17
- Open Issues: 3
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- awesome-ai-devtools - SwePT
- awesome-ai-dev-tools - SwePT - Open source PR generator. (Code Generation & Analysis / Pull Request Automation)
- awesome-ai - SwePT
README
# Japeeti
AI sends pull requests for features you request in natural language
- Create frontend website for marketing
- Semantic search for what files to update and recursive update
- Slackbot that takes requests from users and responds with PR links
- Github api to retrieve PR comments and what file, line num it correspond to
## Usage
```
python swept.py -h
usage: swept.py [-h] -f FILE -i INSTRUCTION [-r REPO] [-d] [-pr]
Edit a section of code, PR with changes.
optional arguments:
-h, --help show this help message and exit
-f FILE, --file FILE location of a file
-i INSTRUCTION, --instruction INSTRUCTION
instruction on how to edit the file
-r REPO, --repo REPO location to git repo
-d, --diff show diff
-pr, --pull-request add change, commit, push and raise a PR
```
## Example
```
OPENAI_KEY="" GH_TOKEN="" python swept.py -f examples/import_indent_bug.py -i "Rewrite the given code and fix any bugs in the program." -d --pr
```
Here is a real PR opened by this above command: https://github.com/keerthanpg/SwePT/pull/8
```
OPENAI_KEY="" GH_TOKEN="" python swept.py
-f /Users/keerthanapg/robotics_transformer/tokenizers/image_tokenizer.py -i "Rewrite the given code by making the __call__ function always use token learner and remove use_token_learner and self._use_token_learner variable" -r "/Users/keerthanapg/robotics_transformer" -d -pr
```
Here is a more complex PR opened on an open source repo: https://github.com/keerthanpg/robotics_transformer/pull/5/commits/20dda2730774a414cfd6a59e59f8870c83ce6307