How to merge pull requests using the carbon_pr.py

This code is heavily inspired by the similar tool in Apache Falcon.

This section documents the process for reviewing and merging code changes contributed via Github pull requests. It assumes you have a clone of carbondata's Git repository. carbon_pr.py is a script that automates the process of accepting a code change into the project. It creates a temporary branch from apache/master, squashes the commits in the pull request, rewrites the commit message in the squashed commit to follow a standard format including information about each original commit, merges the squashed commit into the temporary branch, pushes the code to apache/trunk and closes the JIRA ticket. The push will then be mirrored to apache-github/master, which will cause the PR to be closed due to the pattern in the commit message. Note that the script will ask the user before executing remote updates (i.e. git push and closing JIRA ticket), so it can still be used even if the user wants to skip those steps.

Installing Dependencies (Optional)

If you want to close JIRA issues using the script then you will need to install jira library. You can install it using easy_install with

sudo easy_install jira

or using pip with

pip install jira

Setting up

The script requires two remotes to be configured - one for the github repo of the project(to download pull requests), and one for the git repo of the project(to push the changes).

  1. Add aliases for the remotes expected by the merge script:

git remote add apache https://git-wip-us.apache.org/repos/asf/carbondata.git

git remote add apache-github https://github.com/apache/carbondata.git

If you execute the script without doing any setup, the script will output the commands that you need to execute to setup these remotes.

Executing the tool

Once the pull request is ready to be merged (it has been reviewed, feedback has been addressed, CI build has been successful and the branch merges cleanly into trunk):

  1. Run the merge script

python carbon_pr.py

  1. Answer the questions prompted by the script.

TIP: If you don't want to enter your jira credentials again and again then you can set the JIRA_USERNAME and JIRA_PASSWORD environment variables and the script will use them instead of prompting for credentials.