dpdug4n / slack_bolt_boilerplate

Boilerplate for containerized slack app

Slack Boilerplate

Docker

Execute docker-compose --env-file ./app_name/.env up to build and start the docker container.
docker-compose up --force-recreate to rebuild the container after making changes to app.py

Tips

When your app uses blockkit views for home pages, shortcut modals, etc -
store each view's JSON in ./assets/views.json with the property as the view name, and the value as the view's json.

• The boilerplate loads this file and assigns it to a views variable, reference each view via views['view_name']

To allow for dynamic parsing of views and their inputs, use a standard naming convention in the block ids and action ids.

• An input field named user_input would have a block id of user_input_block and the element's action id would be user_input_action

  params = {
      "input1" : view['state']['values']['input1_block']['input1_action']['value'],
      "input2" : view['state']['values']['input2_block']['input2_action']['value'],
  }
  #dynamic input validation
  errors = {}
  for field in params.keys(): 
      if len(params[field]) > 3000:
          errors[f'{field}_block']='Max character limit is 3k.' 
  if len(errors) > 0:
      ack(response_action="errors",errors=errors)
      return
  

Store utility functions in ./utils/helpers.py or in their own .py modules and import them into app.py

• This helps keep the code clean & organized and helps prevent the application code from being monolithic.
• Add an empty __init__.py file in a subfolder to enable python to recognize it as a module and import the submodules.

Rename .env.example to .env if you're using it for local development.

Useful Links

• https://api.slack.com/apps
• https://slack.dev/bolt-python/tutorial/getting-started
• https://app.slack.com/block-kit-builder/
• https://api.slack.com/reference/surfaces/formatting
• https://api.slack.com/methods
• https://slack.dev/bolt-python/api-docs/slack_bolt/