Meme Say custom Slack Slash custom

Intro

I just wanted to fool around and see if i could get slack slash commands working
requires graphicsmagick
uses dropbox for a free awesomely usable solution.

How to use

Provide in github secrets HEROKU_API_KEY, HEROKU_APP_NAME, HEROKU_APP_URL, HEROKU_EMAIL
ALso set up your dropbox app with full access and token key - provide this key in github actions secrets as DROPBOX_API_TOKEN. This will be passed in as an ENV variable.
it will use meme-say/bots and meme-say/generated-memes
make a folder named after what you'll call your custom command (for example alex-say) in the bots folder and place your memeable pictures in that folder - ensure these images aren't too big. about 1000px is goopd
add the endpoint as a custom slash command in your slack group - you can reuse this endpoint for as many custom slash commands as you want - for example you could create an ana-say integration, and elton-say and despite being seperate integrations - you can use the same url.
It's nice to give the hint [first-line];[second-line]

Example bot creation

Scenario: create a bot called lachlan

log into the same dropbox account the dropbox app you created has.
go into meme-say/bots and create a folder called lachlan and put images in there
in slack create a custom slash command called lachlan-say(naming here is important as it must match the folder name you created)
use the endpoint you created when you deployed this to heroku
success

How it works

Given a slack command (for example alex-say) - it will use this as the prefix for the configured bucket.

It expects there to be images in this prefix - and will choose one at random.

It will then print your meme text and return it to slack.

Getting started

install graphicsmagick
run yarn or npm install
run the app with yarn start or npm run start
Can test with: bot mode

curl --header "Content-Type: application/x-www-form-urlencoded" \
--request POST \
--data 'text=22222dasdgsadgaserpnot have;TP??&command=/ana-say' \
http://localhost:3000

meme-say mode

curl --header "Content-Type: application/x-www-form-urlencoded" \
--request POST \
--data 'text=will 22222dasdgsadgaserpnot have;TP??&command=/meme-say' \
http://localhost:3000

(if using within lambda - need to run npm install --arch=x64 --platform=linux sharp` for distribution)

Notes that aren't useful anymore but useful if we decide to use lambda again

zip distribution as a .zip and use aws s3 cp meme.zip s3://terraform-wlee-meme/v1.0.20/meme.zip
use terraform apply -var="app_version=1.0.20"