Welcome dear world to yet another survey slackbot. This started out as an attempt to learn some Golang and the Slack API. I wanted to build a bot that would be used to pose single simple questions to groups or individuals. The goal to drip health and engagement data ( okay and some fun team trivia) from Slack. Yes, this has be done before but there is nothing like writing something for yourself to learn. This is still a work in progress but the conversation flow goes like this. Open a private message with Carlos:
dana [10:39 PM]
create response poll
carlos_the_curiousBOT [10:39 PM]
Creating response poll. You can cancel the poll any time with `cancel poll 437c67df-febb-4350-b561-92296ea97f6a`
[10:39]
What was the question you wanted to ask?
dana [10:40 PM]
Did you remember to write tests first?
carlos_the_curiousBOT [10:40 PM]
What are the possible responses (comma separated)?
dana [10:40 PM]
yes,no
carlos_the_curiousBOT [10:40 PM]
Who should we send this to?
dana [10:41 PM]
@dana
carlos_the_curiousBOT [10:41 PM]
Here's a preview of what we are going to send:
Look good to you (yes/no)?
wunderbar
Did you remember to write tests first?
Recipients:
----------------
@dana
Possible Answers:
----------------
yes, no
dana [10:42 PM]
yes
carlos_the_curiousBOT [10:42 PM]
We have a question for you. You can answer via `answer poll 437c67df-febb-4350-b561-92296ea97f6a {insert response}`
Question
Did you remember to write tests first?
Possible Answers:
----------------
yes, no
[10:42]
Poll is live you can check in by asking me to `show poll 437c67df-febb-4350-b561-92296ea97f6a
Lots left to do:
- - test coverage and quality
- - refactor and code cleanup
- - better eventing/statemachineness
- - recipients can be channels or users
- - display the results of the poll
- - integration testing
- - support open ended questions
- - platform for analysis
- - schedule polls e.x ask poll every monday morning
First before you doing anything setup Docker for Mac. Once up and running You can build the project via make build. This will build the project inside the container. To use Carlos you will need to have a postgres database running somewhere. I would suggest using the postgres docker container which should setup everything at least for development. You can set this up via make database-build which will create the container. You can run the container via make database-up and bring down with make database-down.
Run the container with:
DATABASE_URL={insert database URL} SLACK_TOKEN={insert your slack token} make run
But, If you want to trigger the container yourself:
docker run --net=host --rm -it -e "DATABASE_URL=postgres://postgres:@127.0.0.1/carlos?sslmode=disable" -e "SLACKTOKEN={{insert your slack token here}}" carlos-the-curious