Skip to content

Latest commit

 

History

History
93 lines (68 loc) · 4.25 KB

README.md

File metadata and controls

93 lines (68 loc) · 4.25 KB
Raw

tapioca

tapioca is a slack app, that will make conversation groups with random people inside your organization and suggest something to talk about!

How to use it? (for users)

Just enter on https://www.yuca.live/tapioca/ and press on Add Tapioca to Slack and follow instructions. This will create a new channel named tapioca-time. Then you can invite people to join the channel. tapioca will take care of the rest, making groups with folks to talk and suggesting topics to them talk about.

How to deploy or make a clone of it? (for devs)

If you want, you can keep using the version hosted by Yuca, but if you want to deploy your own version of tapioca app, it will require you to make a new Slack App and
Amazon Web Services account to run for now.

The only required service so far is the S3 that is where it stores the slack's oauth tokens.

We suggest to deploy using Lambda with API Gateway as we did.

It will require two lambda functions:

1. oauthLambda.js

This function will receive the Oauth hook from slack and save workspace token to be used later. As Trigger for this lambda add an API Gateway endpoint, and save this URL on your slack app webhook.

2. tapiocaLambda.js

This function will require some kind of scheduled run, we use a CloudWatch Events/EventBridge, that is a CRON to schedule when it should run. We use: cron(30 20 ? * WED *)

How to build and run it?

First things first, if you will use Lambda to deploy it, we suggest you to install and use SAM to test it in your local machine:

For Gnu/Linux with python/pip:
pip install -U pip setuptools
For Mac OS:
brew install aws/tap/aws-sam-cli

Then with SAM installed you can trigger events to test like a real AWS Lambda call.

You will need to get an Oauth code from slack, that you can get when you give give access to the application to your workspace, replace this code on events/event.json, on queryStringParameters on code, unfortunately that token is for a single use only, so you will need to get a new one every time you run locally.

Then to test, the group creation and token saving, run inside tapioca folder:

sam local invoke "Oauth" --event events/event.json

And then to trigger conversation making run:

sam local invoke "Tapioca"

Then you run conversation making lambda, if everything works as they should, you have the tapioca-time channel on your slack you must see a new conversation with you and other friends who also are in the channel.

Required ENV vars
  • S3_ACCESS_KEY_ID and S3_SECRET_ACCESS_KEY: if you need help to get those values: AWS: Understanding and Getting Your Security Credentials.
  • S3_BUCKET_NAME: the name of bucket tapioca will use to save your files.
  • S3_TOKEN_FILE_NAME: the name of the file inside the S3_BUCKET_NAME that will hold the slack oauth token.
  • APP_CLIENT_ID and APP_CLIENT_SECRET: are your slack app client_id and secret. You get these values when you create your slack app, check Create a Slack app and authenticate with Postman
  • SUCCESS_TAPIOCA_PAGE: the redirect url of success page.
  • FAIL_TAPIOCA_PAGE: the redirect url in case some error happens.

Building

There is a script called build.sh that will assembly two ZIP files so you can upload on AWS Lambda, one for Oauth and other for Tapioca.

Contributing

If you want to make this tool even more awesome, please check CONTRIBUTING before start.

License

We use MIT License and any contribution to this project will automatically accept this license.

Please, check our LICENSE file for more information.