pipe gpt下載 - pipe gpt原始碼下載

pipe gpt

其他源碼

v0.4.0

下載

管道-gpt

直接從命令列將內容傳輸到 gpt。一個允許很多可能性的概念。

透過 Cargo cargo install pipe-gpt安裝，透過cargo build --release在本地構建，或從發布頁面下載。

請注意，您將需要 OpenAI API 金鑰。

設定開放 api 密鑰環境變量

在 Linux 中： export OPENAI_API_KEY=sk-12345abc
在 Windows powershell 中$env:OPENAI_API_KEY = 'sk-12345abc'

使用案例

本地命令列使用

tail -30 /var/httpd.log | pipe-gpt --p "Is there anything in the http log file I should fix?"
cat main.rs | pipe-gpt -p "How would you improve this code? Include line numbers in your comments so I can tell where you mean."
cat main.rs | pipe-gpt -p "Is this code production ready? If yes reply 'Yes'. If no, then explain why not. Be concise."
cat file.json | pipe-gpt -p "Convert this JSON to YAML" > file.yaml
cat french.txt | pipe-gpt -p "Translate this to English please."
git diff --staged | pipe-gpt -p "Code review this code change"
cat src/main.rs | pipe-gpt -p "improve the code and only output the replacement code as I will pipe the output directly back into a file, no explanations, just pure code please" > src/main.new.rs

用於本地開發的管道-gpt

 user@computer:/GitHub/pipe-gpt$ cat src/main.rs | pipe-gpt -p "how would you improve this code? include line numbers in your comments so I can tell where you mean"

Here are some suggestions to improve your code:

1. Line 2, 3: You can combine these two lines into one as follows:
`use openai_api_rust::{*, chat::*};`

2. Line 6-16: You can simplify the creation of `conversation_messages` by using a `Vec::new()` and then pushing messages as needed. This way, you avoid creating an unnecessary `Message` when `prepend` is empty.

3. Line 18-19: Instead of using `unwrap()`, it's better to handle the error properly. You can use `?` to propagate the error up.

4. Line 20-33: You can create a function to initialize `ChatBody` to make the code cleaner.

5. Line 35-38: Instead of using `unwrap()`, it's better to handle the error properly. You can use `match` to handle the possible `None` case.

CI 的管道 gpt

 jobs :
  build :

    runs-on : ubuntu-latest

    steps :
    - uses : actions/checkout@v3
    - name : Build
      run : cargo build --verbose
    - name : Run tests
      run : cargo test --verbose
    - name : GPT Code Review
      env :
        OPENAI_API_KEY : ${{ secrets.OPENAI_API_KEY }}
      run : pwd && find . -path './target' -prune -o -name '*.rs' -exec echo {} ; -exec cat {} ; | ./target/debug/pipe-gpt -p "how would you improve this code? include line numbers in your comments so I can tell where you mean"