Skip to content

Feature: A class to interact with the github API

Okay, here's a prompt designed to be given to Gemini Code Assist, building on our previous discussion and leveraging its ability to generate multi-file projects and detailed code.

Gemini Code Assist Prompt:

"My project needs an object-oriented Python utility to interact with GitHub repositories, specifically for fetching issues and pull requests. I already have a shared low-level API fetching utility located at scripts/github_api.py which contains a fetch_from_github(owner, repo, endpoint, params) function that handles pagination and authentication via GITHUB_TOKEN environment variable.

I need you to create a new Python package and populate it with two classes: Repo and Item.

Here are the requirements:

  1. Project Structure:

    • Create a new directory local_data_platform.
    • Inside local_data_platform, create an empty __init__.py file to make it a package.
    • Inside local_data_platform, create a new file named github.py.

  2. local_data_platform/github.py content:

    • Import: It should import fetch_from_github from scripts.github_api.

    • Repo Class:

      • Represents a GitHub repository.
      • __init__(self, owner: str, name: str): Initializes with the repository owner (username or organization) and name.
      • __repr__(self): Provides a helpful string representation.

    • Item Class:

      • Represents a single GitHub issue or pull request.
      • __init__(self, data: dict, repo: 'Repo'):
        • Takes a data dictionary (the raw JSON response for an issue/PR from GitHub API) and a repo object (an instance of the Repo class).
        • Determines self.type as either 'issue' or 'pull_request' based on the presence of the 'pull_request' key in data.
        • Initializes common attributes directly from data: number, title, html_url, state, created_at, updated_at, closed_at, user_login (from user.login), and labels.
      • is_pr(self) -> bool: Returns True if the item is a pull request, False otherwise.
      • is_issue(self) -> bool: Returns True if the item is a regular issue, False otherwise.
      • __repr__(self): Provides a helpful string representation including type, number, title (truncated), state, and repo name.

      • __getattr__(self, name): Implement this to allow accessing any key from the underlying _data dictionary directly as an attribute (e.g., item.body or item.assignee). If the attribute doesn't exist in _data, raise an AttributeError.

      • @classmethod fetch_all_items(cls, repo: Repo, status: str = "open") -> list['Item']:

        • This is the core fetching method.
        • It should take a Repo object and an optional status string (defaulting to "open").
        • It must use the imported fetch_from_github function, passing repo.owner, repo.name, the endpoint "issues", and params={"state": status}. (Note: The GitHub /issues endpoint returns both issues and PRs, which is suitable here).
        • It should then iterate through the raw data returned by fetch_from_github and wrap each dictionary in an Item object, returning a list of Item instances.
        • Include basic type validation for repo parameter.

  3. Example Usage Script (main.py):

    • Create a main.py file in the project root (sibling to local_data_platform and scripts).
    • It should demonstrate how to:
      • Import Repo and Item.
      • Define REPO_OWNER and REPO_NAME constants (e.g., "tusharchou", "local-data-platform").
      • Instantiate a Repo object.
      • Call Item.fetch_all_items to get open items.
      • Print the total count of open items.
      • Separate the fetched items into open_issues and open_prs lists using the is_issue() and is_pr() methods.
      • Print the counts for open issues and open PRs.
      • Loop through and print the number, title, type, state, user_login, created_at (formatted), and html_url for a few (e.g., top 5) issues and PRs.
      • Demonstrate fetching closed items as well.
      • Show an example of using the __getattr__ functionality (e.g., printing item.body if available).
    • Include a note about setting the GITHUB_TOKEN environment variable for authenticated requests.

Please provide the complete code for local_data_platform/__init__.py, local_data_platform/github.py, and main.py."