GitHub API Repository Search (Python 3.6.2)

keyword_search_github_repositories.py

#!/usr/bin/env python3.6.2
# -*- coding: utf-8 -*-

# run script from command line via python3 keyword_search_github_repositories.py

import click
import datetime
import time

from github import Github
from github.GithubException import RateLimitExceededException
from tqdm import tqdm


def search_github(auth: Github, keyword: list) -> list:
    """Search the GitHub API for repositories using an input keyword.

    Args:
        auth: A Github authenticate object.
        keyword: A keyword string.

    Returns:
        A nested list of GitHub repositories returned for a keyword. Each result list contains the repository name,
        url, and description.
    """

    print('Searching GitHub using keyword: {}'.format(keyword))

    # set-up query
    query = keyword + '+in:readme+in:description'
    results = auth.search_repositories(query, 'stars', 'desc')

    # print results
    print(f'Found {results.totalCount} repo(s)')

    results_list = []
    for repo in tqdm(range(0, results.totalCount)):
        try:
            results_list.append([results[repo].name, results[repo].url, results[repo].description])
            time.sleep(2)
        except RateLimitExceededException:
            time.sleep(60)
            results_list.append([results[repo].name, results[repo].url, results[repo].description])

    return results_list


@click.command()
@click.option('--token', prompt='Please enter your GitHub Access Token')
@click.option('--keywords', prompt='Please enter the keywords separated by a comma')
@click.option('--filename', prompt='Please provide the file path')
def main(token: str, keywords: str, filename: str) -> None:

    # initialize and authenticate GitHub API
    auth = Github(token)

    # search a list of keywords
    search_list = [keyword.strip() for keyword in keywords.split(',')]

    # search repositories on GitHub
    github_results = dict()
    for key in search_list:
        github_results[key] = []
        github_results[key] += search_github(auth, key)
        if len(search_list) > 1: time.sleep(120)

    # write out results
    timestamp = datetime.datetime.now()
    formatted_date = timestamp.strftime('%d') + timestamp.strftime('%b') + timestamp.strftime('%Y')
    full_filename = filename.strip() + 'GitHub_Search_Results_' + formatted_date + '.txt'

    print('Writing search results to: {}'.format(full_filename))
    with open(full_filename, 'w') as f_out:
        for key in tqdm(github_results.keys()):
            for res in github_results[key]:
                f_out.write(key + '\t' + str(res[0]) + '\t' + str(res[1]) + '\t' + str(res[2]))

    f_out.close()


if __name__ == '__main__':
    main()

Note. sleep timing could be improved with some API sensitivity experimentation.

python -m pip install tqdm click pygithub

python -m pip install tqdm click pygithub

Thanks so much @Abdur-rahmaanJ!

@callahantiff This could be a nice project of it's own btw