/wildflower-platform

Primary LanguageRuby

README

This README would normally document whatever steps are necessary to get the application up and running.

Things you may want to cover:

  • Ruby version

  • System dependencies

  • Configuration

  • Database creation

  • Database initialization

  • How to run the test suite

  • Services (job queues, cache servers, search engines, etc.)

Deployment instructions

Any merge to the main branch on the Github repository will automatically deploy to produciton and run migrations. You can follow the process by using the heroku logs --tail command.

Development Environment

The development host is https://api-dev.wildflowerschools.org Merging to the branch main will automatically deploy to this environment.

Staging Environment

The staging host is https://api-staging.wildflowerschools.org. Merging to the branch staging will automatically deploy to this environment.

Production Environment

The production host is https://api.wildflowerschools.org. Merging to the branch production will automatically deploy to this environment.

Platform API

  • Search APIs: people/schools

  • show/index

  • edit/update

  • create

  • main entities are people and schools

  • school relationships connects the two

  • tags are another dimension

Workflow

If you have a large task that needs to be completed over and over, chances are you need a system that can:

  1. break the large task into smaller piece of work
  2. manage dependencies between pieces of work; e.g. when can we start on what?

This system aims to do that.

You can have workers (users who do the small pieces of work) be told:

  1. what to work on next (based on availability)
  2. where are they in the work (in the larger process)
  3. see what dependencies are needed for a piece of work
  4. see what work is dependent on someone else

When there is changes in the state of the system, you can recompute the new work needed.

If there are work weights, you can project a completion date, or given a completion date, know if you are going to miss.

  1. what to work on next (based on urgency)

Workers can add manual tasks to track missing work.

Analytics

  • you can see average completion times
  • you can see what manual steps are missing
  • you can track which workers are working well

ActiveStorage

Using active storage to manage uploaded profile image's storage and the attachment to the person model.

The workflow is as such:

  • client sends a POST request to /rails/active_storage/direct_uploads with some blob information, including checksum
  • API creates a blob record in the database and sends back a token, a URL for the client to directly upload the image to, and a signed_id to identify the blob
  • client then uploads image to the direct URL with the token. API verifies the token is in fact for a blob, and then proceeds to upload to whichever service (production: amazon, local: local).
  • client then updates the person record with the signed_id as the profile_image

Configuration

  • storage.yml lists the different storage services we use.
  • Set which type of service for which environments via the line config.active_storage.service
  • View aws credentails by running rails credentials:show in the terminal.

AWS S3

  • If you want to test aws locally, change config.active_storage.service = :local to config.active_storage.service = :amazon in development.rb. It should upload to the ssj-local bucket setup in S3.
  • The client's direct upload url's host will be https://ssj-local.s3.amazonaws.com
  • Note: If you receive a 403 forbidden error, check the CORS policy in S3.

Summary

  • this system includes
    • data models
    • commands/services for core actions (should really be exposed as an api or callable code...)
    • background jobs for all the domain agnostic hooks
    • it was gonna include the workflow author UI (admin UI) but that's just a spreadsheet for now
      • eventually UI will need to be more smart.
    • the views to do the actual work need to be hosted elsewhere
      • read (link to content, iframe), watch/listen: youtube videos
      • form input
        • eventually the conditionally is coupled to the data model
        • build a language for the primitives that we are conditional on
        • then we need a data mapping from host application which meets API (when doing any workflow action, we need to give the instance's conditional context...)