Flask Server

A Flask-based REST API server that provides GPT model interaction and file management capabilities.

Git Tips

Enhance your workflow with these detailed Git tips:

# 1. Before pushing, update your local repository:
git fetch -p  # Fetch remote changes and prune deleted branches

# 2. If the main branch has new changes:
git checkout main
git pull      # Update your local main branch with the latest commits
git checkout your_branch
git rebase -i main  # Interactively rebase your branch to integrate changes and resolve conflicts

# 3. When starting a new feature or fix:
git checkout -b your_branch  # Create a new branch based on the updated main
# Develop your feature, then repeat step 2 to stay in sync with main
git push --set-upstream origin your_branch  # Push your branch to the remote repository

# 4. Once your changes are reviewed:
# Open a pull request on GitHub to merge your_branch into main

Additional tips:

• Use descriptive commit messages.
• Frequently rebase to catch conflicts early.
• Leverage branch protection rules on GitHub for quality control.
• Consider squashing commits for a cleaner history.

Setup

1. Install dependencies:

pip install -r requirements.txt

2. Configure environment variables:

• Create .env file with:

  OPENAI_API_URL=your_openai_api_url
  OPENAI_API_KEY=your_openai_api_key
  DEEPSEEK_API_URL=your_deepseek_api_url
  DEEPSEEK_API_KEY=your_deepseek_api_key
  

3. Run the server with additional arguments:

python src/app.py --debug --host 0.0.0.0 --port 5000

see more details

Features

1. GPT Integration

• Endpoint: /ask
• Method: POST
• Request Format:

  {
    "prompt": "Your question here"
  }

• Response:

  {
    "response": "GPT response content"
  }

2. File Management

Upload File

• Endpoint: /upload
• Method: POST
• Content-Type: multipart/form-data
• Parameters:
  • file: File to upload
  • path: Target directory path
• Response:

  {
    "message": "File uploaded successfully"
  }

List Directory

• Endpoint: /list
• Method: GET
• Query Parameters:
  • path: Directory path to list
• Response:

  ["file1.txt", "file2.txt", "directory1"]

Create Directory

• Endpoint: /mkdir
• Method: POST
• Request Format:

  {
    "path": "desired/directory/path"
  }

• Response:

  {
    "message": "Directory created successfully"
  }

Error Codes

• 410: No prompt provided
• 411: Unknown model specified
• 420: No file part in request
• 421: No path specified
• 422: Invalid path
• 423: Path does not exist
• 424: No selected file
• 425: Path already exists

Todo

• Add Database
• [] Add an item to distinguish IOS and Android Version.

login on ECS

Ip: 120.25.178.24

login

1. download private key in group chat.

the key file name is dataset_app.pem please contact Wenjie (wduaj@cse.ust.hk) and move it in ~/.ssh/

2. Add the config in ~/.ssh/config

chmod 400 ~/.ssh/dataset_app.pem

Host ecs 
  HostName 120.25.178.24 
  Port 22 
  User root 
  IdentityFile ~/.ssh/dataset_app.pem 

3. Use ssh login Terminal or VSCode

References

• Flask File Uploading
• Flask Database
• Run application
• MongoDB