Using the clever concurrency testing borrowed from SubscriptionPlacementJob, but I thought a shorter pause time (just 100ms) would be sufficient.
I considered doing this with a new 'state' field (upcoming/open/close), but decided to keep it simple.
Although we won't be allowing multiple in the this PR, we certainly plan to in the future.
The migration helper add_reference couldn't handle the custom column name, so I had to put it together manually.
This will store the URL for each user that wants a notification.
We probably don't need URL validation (it's not done on Enterprise for example). It could be validated by browser input, and anyway will be validated if the webhook actually works or not.
Inspired by Keygen: https://keygen.sh/blog/how-to-build-a-webhook-system-in-rails-using-sidekiq/
With product validations being prioritized, custom and
insightful error messages were being overwriten by
general and ambiguous model validations, which
were confusing users
Before if a shipping method was shared between multiple distributors it could only be disabled/enabled on that order cycle for all the distributors which have that shipping method e.g. you couldn't select that shipping method for one distributor but disable it for another.
Before the OrderCycleShippingMethod had a validation which checked the shipping method belonged to the order cycle distributor. Instead of this validation this just ignores shipping methods which don't belong to one of the order cycle's distributors when they are being attached in the OrderCycleForm service. This pattern is already being used in the OrderCycleForm service for ignoring Schedules that the person doesn't own.
Co-authored-by: Maikel <maikel@email.org.au>
Instead we will make sure the order cycle is not available on the shopfront if it is doesn't have valid shipping methods. This will preven the issue where if one distributor deletes their shipping method, we don't want to invalidate the order cycle for all other distributors.
Co-authored-by: Maikel <maikel@email.org.au>
It's a clearer name because 'preferred' implies there could be other unpreferred shipping methods available as well.
Co-authored-by: Maikel <maikel@email.org.au>
It makes things much simpler if we return all shipping methods by default without needing OrderCycleShippingMethod records to be added to the database.
Co-authored-by: Maikel <maikel@email.org.au>
Specs should test public methods. Private methods should be able to
change without specs breaking.
The specs were also just mirroring the code and didn't really have any
meaning. So I wrote new ones which are hopefully clearer.
There's actually a case which I don't quite understand. Why don't we
require a customer for a new complete order? It means that we don't have
a customer creating it but then it's added when updating the order.
Seems inconsistent.
