a python script to call the github search api looking for code

github-search.py

#!/usr/bin/python

import requests
import json
import os
import time

def search_github_code(query, language=None, sort=None, order=None, per_page=30, page=1):
    """
    Search for code on GitHub using the GitHub Search API.
    
    Parameters:
    - query (str): The search query
    - language (str, optional): Filter by programming language
    - sort (str, optional): Sort results by 'indexed' (default) or 'best-match'
    - order (str, optional): Order results 'desc' (default) or 'asc'
    - per_page (int, optional): Number of results per page (max 100)
    - page (int, optional): Page number for pagination
    
    Returns:
    - dict: JSON response from GitHub API
    """
    # Base URL for GitHub Search API
    base_url = "https://api.github.com/search/code"
    
    # Construct the query
    search_query = query
    if language:
        search_query += f" language:{language}"
    
    # Parameters for the request
    params = {
        "q": search_query,
        "per_page": per_page,
        "page": page
    }
    
    if sort:
        params["sort"] = sort
    
    if order:
        params["order"] = order
    
    # Headers for the request
    headers = {
        "Accept": "application/vnd.github.v3+json"
    }
    
    # Add GitHub token if available
    github_token = os.environ.get("GITHUB_API_TOKEN")
    if github_token:
        headers["Authorization"] = f"token {github_token}"
    else:
        print("Warning: No GITHUB_TOKEN found in environment variables.")
        print("Unauthenticated requests are rate limited to 10 requests per minute.")
    
    # Make the request
    response = requests.get(base_url, params=params, headers=headers)
    
    # Check for rate limiting
    if response.status_code == 403 and "rate limit exceeded" in response.text.lower():
        reset_time = int(response.headers.get("X-RateLimit-Reset", 0))
        wait_time = max(0, reset_time - int(time.time()))
        print(f"Rate limit exceeded. Reset in {wait_time} seconds.")
        
        if wait_time > 0 and wait_time < 300:  # Only wait if less than 5 minutes
            print(f"Waiting {wait_time} seconds...")
            time.sleep(wait_time)
            return search_github_code(query, language, sort, order, per_page, page)
    
    # Check if the request was successful
    response.raise_for_status()
    
    return response.json()

def display_results(results):
    """Display search results in a readable format."""
    total_count = results.get("total_count", 0)
    items = results.get("items", [])
    
    print(f"Found {total_count} code results")
    print("-" * 50)
    
    for i, item in enumerate(items, 1):
        repo = item.get("repository", {})
        print(f"{i}. {item.get('name')}")
        print(f"   Repository: {repo.get('full_name')}")
        print(f"   URL: {item.get('html_url')}")
        print(f"   Path: {item.get('path')}")
        print("-" * 50)

if __name__ == "__main__":
    import argparse
    
    parser = argparse.ArgumentParser(description="Search for code on GitHub")
    parser.add_argument("query", help="Search query")
    parser.add_argument("--language", "-l", help="Filter by programming language")
    parser.add_argument("--sort", "-s", choices=["indexed", "best-match"], 
                        help="Sort results by 'indexed' (default) or 'best-match'")
    parser.add_argument("--order", "-o", choices=["desc", "asc"], 
                        help="Order results 'desc' (default) or 'asc'")
    parser.add_argument("--per-page", "-p", type=int, default=30, 
                        help="Number of results per page (max 100)")
    parser.add_argument("--page", type=int, default=1, 
                        help="Page number for pagination")
    
    args = parser.parse_args()
    
    try:
        results = search_github_code(
            args.query, 
            args.language, 
            args.sort, 
            args.order, 
            args.per_page, 
            args.page
        )
        display_results(results)
    except requests.exceptions.HTTPError as e:
        print(f"Error: {e}")
    except Exception as e:
        print(f"An unexpected error occurred: {e}")