Skip to content

premisedata/slack-approval

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

93 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

slack_approval

PyPI PyPI - Python Version

This python library serves as the basis for managing and deploying a lightweight approval workflow based on Slack and GCP. This library contains two key classes, SlackRequest and SlackProvision as well as the logic to deploy them to GCP, which is based on goblet.

Install library with pip install slack_approval

Slack Approval Flow


approval_flow

The user iniates a request to the reqest function. This function uses the SlackRequest class to then post a message into a private Slack channel with the option to approve or reject the request.

Sample message in the approvers channel.

sample_message

If approved by a member of the private channel, the message is then sent to the provision function, which executes logic based on the request. There is a second Slack channel for requesters to view the status of their request.

Onboarding New Requests for Approval


Provision

  • First add a provision file for your new request provision_{request name}.py
  • In the file, extend the from slack_approval.slack_provision import SlackProvision class and define your own approved() and optionally a rejected() method. By default, rejections log the event in the function and updates the messages in Slack. You only need to define it if other actions are necessary.
  • Import your new class in provision.py. This should be the only thing you need to change anything else in the base files

Request

  • The only requirement for the events sent to the request function is the provision_class field. This is what maps the event to the right class in the provision function. You can add any other fields here and they will show up in the request in Slack.
    • The value for provision_class should be the name of your provision class with spaces

      ie. class ProvisionLBEndpoint -> "provision_class": "Provision LB Endpoint"

  • Optionally, you can add approvers_channel and requesters_channel fields in your request to specify the approvers and requesters channels respectively.
    • If added, add environment variables with the same values.
    • If not provided, the function will look for APPROVERS_CHANNEL and REQUESTERS_CHANNEL environment variables by default.

Example:

curl -X POST -H "Content-Type:application/json" https://REGION-PROJECT.cloudfunctions.net/slack-request -d '{
	"provision_class": "Provision Service",
    "approvers_channel": "APPROVERS_CHANNEL",
    "requesters_channel": "REQUESTERS_CHANNEL",
	"field1": "value1",
	"field2": "value2",
}'
  • The entire data json is available to the provision classes as self.inputs

Deploying Functions


The functions that need to be deployed are found in the /functions folder or can be created by the slack_approval init command.

See the blog post Tutorial: Setting Up Approval Processes with Slack Apps for more detailed slack and GCP setup steps.

Before deploying you will need to modify the config.json file found in the .goblet folder with the required environment variables.

Request

  • APPROVERS_CHANNEL
  • REQUESTERS_CHANNEL
  • SLACK_BOT_TOKEN

Provision

  • SLACK_BOT_TOKEN

To deploy the functions all you need to do is run the following two commands.

  • goblet deploy --stage request

  • goblet deploy --stage provision

Blog post


See the blog post Tutorial: Setting Up Approval Processes with Slack Apps for more detailed slack and GCP setup steps.