this gets tests passing for now, but we'll need to make sure that
notifications.serialize handles letters with personalisation that
doesn't include underscores in future - for example, address_line_1
may also be addressline1 if someone uploaded a CSV
> For get all or get one letter the response needs to be updated so that
> it looks similar to admin app.
>
> status: created|sending --> received letter
> new column: `estimated delivery date`: derived from created at date.
> (see how the admin app is doing it)
>
> NOTE:
> At the moment we only have 2 statuses for a letter created and
> sending, but we will want to have other internal statuses that make
> sense to the Notify team but not our services. When we know those
> statuses the status map will be updated at that point.
– https://www.pivotaltracker.com/story/show/150512525
This commit implements the date (not status) part of this story.
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.
* Ensure we dont raise exception if e.cause does not contain a message
* Ensure we handle case where e.path may be empty
* Refactor existing tests to conform to new format
Created a new schema that accepts request parameters for the
get_notifications v2 route.
Using that to validate now instead of the marshmallow validation.
Also changed the way formatted error messages are returned because
the previous way was cutting off our failing `enum` messages.
The "cost" value was flawed for a couple of reasons.
1. Lots of messages are free, so in those instances the "cost"
doesn't tell you anything
2. The query to get the rate was expensive and we don't have
an obvious way to get it back very efficiently for large numbers
of notifications.
So we scrapped it.
Our previous test ws returning a notification without a `sent_by`
attribute, which meant that cost was always 0.
Unfortunately, this meant that returning a real value for cost was
untested and (whaddya know) it broke immediately.
Old test scenario:
- billable_units=1, sent_by=None, cost=0
New scenarios
- billable_units=0, sent_by='mmg', cost=0
- billable_units=1, sent_by='mmg', cost=1
The new 'v2' API wants to return less data than the previous one,
which was sending back tons of fields the clients never used.
This new route returns only useful information, with the JSON
response dict being built up in the model's `.serialize()` method.
Note that writing the test for this was a bit painful because of
having to treat loads of keys differently. Hopefully we think this
is a good way to write this test, because if we don't, we should
start thinking of a better way to check the values are what we
expect.
