A script to get the count of open/closed issues/PRs on GitHub on a per day basis.
Python
Switch branches/tags
Nothing to show
Clone or download

README.md

GitHub issue count stats

This is a script that fetches all issues and PRs from GitHub for a specific repo. Then it calculates the count of open issues/PRs and their closed counterpart.

You can combine the output of this script with gnuplot and get following.

open issues

How to start

Install the requirements

    pip install github3.py

or

    pip install -r requirements.txt

Get an API key for GitHub and set it as GITHUB_TOKEN in import_issues.py.

Update the ORG and REPO constant in import_issues.py and process_issues.py.

You can also specify a list of bug labels (BUG_LABEL) that rate a ticket with 10, and critical bug labels (CRITICAL_BUG_LABEL) that rate a ticket with 100. All other tickets are rated with 1. These numbers just apply for open tickets.

Then there is a new list for all sub components of your software that contains tickets in the same bug tracker and have separate labels. Each of them will get a sub rating to compare sub components against each other. They are specified by the APP_LABEL config option.

Run fetcher part:

    ./import_issues.py

This will fetch all issues and PRs and save the needed information to a JSON file to work with in the next step, which saves GitHub API requests and speeds the overall process up.

No process this file and calculate the issue/PR count for each day:

    ./process_issues.py

This then saves the result to a file called result.tsv, which has following format(columns are separated by tabs):

    date	open_issues	closed_issues	open_prs	closed_prs
    2012-08-26	0	0	0	1
    2012-08-27	2	1	1	1
    2012-08-28	2	1	0	2

You can plot this file to a line graph via gnuplot and following commands:

    set xdata time
    set timefmt '%Y-%m-%d'
    plot 'result.tsv' using 1:2 title column with lines, 'result.tsv' using 1:3 title column with lines, 'result.tsv' using 1:4 title column with lines, 'result.tsv' using 1:5 title column with lines

For plotting the ranking graph:

    plot 'result.tsv' using 1:6 title column with lines

For plotting the ranking graph per app use this:

    plot 'result.tsv' using 1:7 title column with lines, 'result.tsv' using 1:8 title column with lines, 'result.tsv' using 1:9 title column with lines, 'result.tsv' using 1:10 title column with lines, 'result.tsv' using 1:11 title column with lines, 'result.tsv' using 1:12 title column with lines, 'result.tsv' using 1:13 title column with lines,  'result.tsv' using 1:14 title column with lines