Add CbvFlows model and controller for flow logic (FFS-679) #7
Conversation
* Store the Argyle User ID for the current user in a model called `CbvFlow`. * Create a CbvFlow controller to handle routing between those pages. This controller may get big; we can break it up if it does. * Add a `bin/update` script to run after you `git pull` to make sure you have all the new dependencies and everything. * Add some light testing of each page separately
The filename needs to end with _spec.rb for Rspec to discover and run it.
| def fetch_and_store_argyle_token | ||
| return session[:argyle_user_token] if session[:argyle_user_token].present? | ||
|
|
||
| raise "ARGYLE_API_TOKEN environment variable is blank. Make sure you have the .env.local from 1Password." if ENV['ARGYLE_API_TOKEN'].blank? |
There was a problem hiding this comment.
Eventually we could have a config system that validates this stuff elsewhere
|
|
||
| raise "ARGYLE_API_TOKEN environment variable is blank. Make sure you have the .env.local from 1Password." if ENV['ARGYLE_API_TOKEN'].blank? | ||
|
|
||
| res = Net::HTTP.post(URI.parse(USER_TOKEN_ENDPOINT), "", {"Authorization" => "Basic #{ENV['ARGYLE_API_TOKEN']}"}) |
There was a problem hiding this comment.
This is where we'd call @GeorgeCodes19' service class (or a private wrapper method for it).
| @@ -1,8 +1,10 @@ | |||
| <h1>Hello, Matt</h1> | |||
| <h1>Hello</h1> | |||
|
|
||
| scope '/cbv', as: :cbv_flow do | ||
| get '/entry' => 'cbv_flows#entry' |
There was a problem hiding this comment.
Nit: As far as naming goes, is the cbv_ prefix redundant? I don't feel strongly about it, it's a convention in Login, too (idp_{object}).
| def change | ||
| create_table :cbv_flows do |t| | ||
| t.string :case_number | ||
| t.string :argyle_user_id |
There was a problem hiding this comment.
In a later PR, we may want to store the Argyle-supplied JWT? Although that may be a case of "inappropriate intimacy" between our domain modeling and the 3rd parties' architecture...
Edit: actually I think that is managed by Argyle in its own cookie. Will have to investigate...
Basically the presence of an Argule cookie would affect how we determine flow state.
Additionally, At the moment the user is Linked, we are provided with an account_id that we'd use to query for paystubs in the next step. So maybe that context info is captured in the CbvFlow.
| end | ||
| helper_method :next_path | ||
|
|
||
| def fetch_and_store_argyle_token |
There was a problem hiding this comment.
Future service class stuff
| root_url | ||
| end | ||
| end | ||
| helper_method :next_path |
There was a problem hiding this comment.
Oh interesting I'm not familiar with this. Does this expose the method for use in the templates?
|
|
||
| private | ||
|
|
||
| def set_cbv_flow |
There was a problem hiding this comment.
I wonder if this should be on the model? Maybe returns a symbol or something. I guess my concern is it works by inference - we infer the step based on the presence of an id. Wondering if there's a way to capture that behavior closer to a model, without the controller context
bin/update
Outdated
| def stub_environment_variable(variable, value, &block) | ||
| previous_value = ENV[variable] | ||
| ENV[variable] = value | ||
| block.call | ||
| ENV[variable] = previous_value | ||
| end | ||
|
|
||
| around do |ex| | ||
| stub_environment_variable("ARGYLE_API_TOKEN", "foobar", &ex) | ||
| end |
# Conflicts: # app/controllers/providers_controller.rb # config/routes.rb
|
Fixed! Tested it on my machine, the Link stuff works. |
CbvFlow.This controller may get big; we can break it up if it does.
bin/updatescript to run after yougit pullto make sure youhave all the new dependencies and everything.