Replace labels by adding a key kwarg in the model for status.
We still need this as sqlalchemy attmempts to look for `notification_status`
on the model (Notification/NotificationHistory). To achieve true ORM mapping
(map status -> notification_status) we need the key kwarg.
More here:
http://docs.sqlalchemy.org/en/latest/core/metadata.html#sqlalchemy.schema.Column#key
There are a variety of ways a notification can be created - via the
API, via CSV, via one-off messages, via the same but sent as a test -
the point is, there are lots of entry points, and lots of inconsistency
about how personalisation may be sent in. If there are no
personalisation options in the template, we may get either `None` or
`{}` - if the value is None, to avoid an error in our encryption lib,
we just store None in the database.
THIS ENDS NOW!
We've had some problems on the front-end that is caused by
notifications having the `None` value. This commit changes the
personalisation property getter and setter to store/return `{}`
rather than None.
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
If passing in `redact_personalisation` to the template update endpoint,
we should mark that template permanently as redacted - this means that
we won't ever return the personalisation for any notifications for it.
This is to be used with templates containing one time passwords, 2FA
codes or other sensitive information that you may not want service
workers to be able to see.
This is implemented via a separate table, `template_redacted`, which
just contains when the template was redacted.
Including:
- url to push the inbound sms to
- bearer_token to be added to the header of the request.
The services will be expected to manage these properties.
A job only ever has one notification type.
This is the first deploy, where the columns are added and populated.
Next a data migration will happen to populate these new columns for the older jobs that do not have the values set.
Then we stop populating the old columns and remove them.
This refactoring of the table structure will make the queries to the table much easier to handle.
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.
two endpoints:
* get all inbound sms for a service (you can limit to the X most
recent, or filter by user's phone number [which will be normalised])
* get a summary of inbound sms for a service - returns the count of
inbound sms in the database, and the date that the most recent was
sent
Added a validation method that always fails for scheduled notifications.
Comment out config for scheduled task.
The schedule notifications will be turned on once we can invite services to use it.
Waiting for the service permission story, must commit this in order to keep things from going stale.
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. 🎉
