Aims

• To count the respective number of commits that any GitHub user has made to the respective default branch of his/her public repositories, excluding forked repositories. The necessary data is scraped from the API of the GitHub webpage.
• To sort these repositories according to the number of commits made by said GitHub user.

Methods

The main program in main.py takes 2 arguments from the command line:

• your_username (mandatory): the GitHub username of the user whose commits you want to count and sort by repository.
• my_token (optional but recommended): your own GitHub personal access token.

The latter argument is there only as a precaution; in case too many requests have been made for information from the GitHub API in the past hour, GitHub will return the following error message: API rate limit exceeded. To circumvent this, we can use a simple digest authentication of our requests by providing our own GitHub personal access token. This is because authenticated requests have a higher hourly rate limit.

Python Dependencies

• requests
• json
• re
• pandas
• numpy
• pycurl

In addition, make sure that Python 3 (not Python 2) is used. (This code has been tested on Python 3.7.4.)

Running

To create a local copy of this repository, clone it by first navigating to the path you want to store the local copy and then executing the following on the command line:

git clone https://github.com/liweiyap/SortingGitHubReposByCommits.git

run_main.sh is a Bash script that wraps around the running of main.py on the command line. Optionally, follow the instructions in this link to generate your own personal access token (if you haven't already done so). Then, open up run_main.sh and set the MY_TOKEN variable to your generated token.

Next, to run, simply execute the following in the root of the repository:

./run_main.sh <your_username>

whilst filling in the desired/appropriate value for the your_username argument. The output will be printed on the command line as a Pandas dataframe.

Tests

I have tested the code with 40 of the GitHub users whom I am following.

For example, when running ./run_main.sh liweiyap, the following output is obtained:

                               Name  Commits
                 LeetCode_Solutions       92
                 liweiyap.github.io       60
                  Conway_GameOfLife       54
 HackerRank-InterviewPreparationKit       43
                     ProteinFolding       43
                              MyCPU       40
        SortingGitHubReposByCommits       28
          DataStructures_Algorithms       26
                      metawear-impl       26
                                PvZ       22
   VectorMatrixElementaryOperations        6

Total repositories: 11
Total commits: 440