So that when the admin gets notifications, the template they return
also has a "redact_personalisation" boolean attached to it. Note, it
won't do the redacting on the api - that'll be part of the admin.
Under the hood, this uses an association_proxy, which is essentially
black magic. But it proxies the `redact_personalisation` property of
`TemplateRedacted` onto the `Template` object, so that Marshmallow
can pick it up.
Note: NOT currently added to NotificationWithTemplateHistory
* Alter config so an error will be raised if you forget to mock out a
celery call in one of your tests
* Remove an unneeded exception type that was masking errors
currently, they're made by creating a one-line job, but we want to
reduce task/csv file noise so we're moving them to persist in the
same vein as API usage. However, we can't just call through to that
since there are some differences:
* no api keys
* tighter control over API format
* no scheduling
* no client references
etc.
So, re-using as much of the v2 validation stuff as possible, I've
created this file that just does basic validation, and then calls
through to persist_notification and schedules a task. Woo.
We need this for the two way stuff in the admin app.
We already have this as a public endpoint, but the admin app can’t use
it, because the admin app auths with its own key, not that of the
service it’s acting on behalf of.
This endpoint makes sure that a request originating from one service
can’t be used to see notifications belonging to another service.
This allows us to reference it across the API code base and return it in the API.
But not currently attached to the service DB model - a static method on the class.
Marshmallow validates and deserialises - BUT, when it deserialises,
it explicitly sets `sms_sender=None`, even when you haven't passed
sms_sender in. This is problematic, because we wanted to take advantage
of sqlalchemy's default value to set sms_sender to `GOVUK` when the
actual DB commit happens.
Instead, still use marshmallow for validating, but manually carry out
the json deserialisation in the model class.
This fixes a bug that only manifested when the database was upgraded,
but the code hadn't updated. 🎉
