hubot-pr-fu

hubot-pr-fu

hubot-pr-fu is a chat bot built on the Hubot framework. It was initially generated by generator-hubot. The npm module name is hubot-pr-fu.

Installing the script

Integrating with exisiting bot

Install the script via npm:

npm install hubot-pr-fu --save

Add the script to your external-scripts.json file:

[ "hubot-pr-fu" ]

And start your bot

Running hubot-pr-fu Locally

You can test your hubot by running the following, however some plugins will not behave as expected unless the environment variables they rely upon have been set.

You can start hubot-pr-fu locally by running:

% bin/hubot -a slack

Notes

• There might be some inconsistency with determining the conflict of PRs. GitHub's API returns a mergeable key in the response that has three possible states: true, false and nil. nil state means that the mergeability of that PR has not yet been determined, and a background job has been scheduled. For reasons of making the logic simpler, we don't display information about these nil state PRs. So there might be a rare case where the total number of PRs in status all command doesn't match with the sum of mergeable PRs and non-mergeable PRs in the same command's response.

Commands/Features

User specific commands

status all

This command returns the PR stats for the repo. This information would include:

1. Total open PRs
2. Total PRs which are mergeable and those which are not
3. List of users who have open PRs and links to each of those

status conflicts

This command returns all the PRs which have merge conflicts. This has more detailed information for those compared to status all command. The title, PR number, link to that PR, assignee and the username who opened this PR is included in the information.

status <username>

This command returns all the PRs opened by this user. This includes all open PRs which are mergeable and non-mergeable. The username is assumed to be a valid Github username. For now, there is no authorization or authentication built-in. Another assumption is that a username in lowercase is same as capitalized or camelcased. That is, for a username bootstrap, the users Bootstrap, bootStrap are the same.

One caveat is that the robot will treat every string after status other than all and conflicts as a username. So, don't abuse this. Adding a way to make this better is in the works.

Configuration

Some env vars are assumed to exist:

• HUBOT_SLACK_TOKEN (Get this from Slack account page) Read this for steps to obtain a slack token and set the bot name steps](#getting-a-slack-token) to know the name of the bot.
• GH_AUTH_TOKEN (Get this from Github)
• PR_STATUS_GITHUB_ORG
• PR_STATUS_GITHUB_REPO

The org and repo are the username or organization name and the repository's name respectively. These have to be added to the environment based on which project you want to configure the robot to work with.

Getting a Slack Token

1. Signin into Slack via https://slack.com/signin
2. Head to https://.slack.com/services/new/bot
3. Register a new bot with whatever name you want
4. Save the service, and you'll be shown the API key. Grab this key and k set it to HUBOT_SLACK_TOKEN

Contribution

An example script is included at scripts/example.coffee, so check it out to get started, along with the Scripting Guide.