One small change in behavior is that this creates an array with all the
row_objects at once, rather than creating them 1000 at a time.
That should be fine, given that the client batches these in units of
10000 anyway, and so we're just creating 10K rows of a relatively
small data structure in Python code here.
This provides a clean warning and 40x error, rather than a 500, for
this corner case which is very likely user error.
The test here is awkward because we have to work around
https://github.com/zulip/zulip/issues/12362.
This adds a new API for sending basic analytics data (number of users,
number of messages sent) from a Zulip server to the Zulip Cloud
central analytics database, which will make it possible for servers to
elect to have their usage numbers counted in published stats on the
size of the Zulip ecosystem.
Now that we allow multiple users to have registered the same token, we
need to configure calls to unregister tokens to only query the
targeted user_id.
We conveniently were already passing the `user_id` into the push
notification bouncer for the remove API, so no migration for older
Zulip servers is required.
Previously, Zulip did not correctly handle the case of a mobile device
being registered with a push device token being registered for
multiple accounts on the same server (which is a common case on
zulipchat.com). This was because our database `unique` and
`unique_together` indexes incorrectly enforced the token being unique
on a given server, rather than unique for a given user_id.
We fix this gap, and at the same time remove unnecessary (and
incorrectly racey) logic deleting and recreating the tokens in the
appropriate tables.
There's still an open mobile app bug causing repeated re-registrations
in a loop, but this should fix the fact that the relevant mobile bug
causes the server to 500.
Follow-up work that may be of value includes:
* Removing `ios_app_id`, which may not have much purpose.
* Renaming `last_updated` to `data_created`, since that's what it is now.
But none of those are critical to solving the actual bug here.
Fixes#8841.
There are several situations in which we want to create a Customer and
stripe.Customer object before we really have a billing relationship with a
customer. The main one is giving non-profit or educational discounts.
We've had this sort of logic for GCM for a long time; it's worth
adding for APNS as well.
Writing this is a bit of a reminder that I'm not a fan of how our unit
tests for push notifications work.
This completes the separation of our logic for managing Stripe
customers from the view code for the billing page.
As we add more features to our Customer model and to our Stripe
integration, we might further separate those two things; but for now
they're nearly synonymous and there's no problem in them being mixed
together.
Pull the code that talks to Stripe out into its own functions.
In a followup commit we'll move these to a separate file, as well
as the error-handling logic that remains in the view function
for now.
Also fix the translation markings: the translated string must be a
constant (e.g. a format string), or else translation is impossible.
Viewing with `-b` shows the few changes that happen in the logic
as it moves out of the view function; viewing without shows the
few changes in the rest of the view function.
Several changes:
* De-duplicate code for different error types.
* No need to list lots of error subtypes where we aren't treating
them differently; StripeError is the base class of them all.
* Unexpected, non-Stripe-related, exceptions we can handle in the normal
way. Just make them show up in the billing-specific log too.
* The Stripe client library already logs type, code, param, and message
before raising an error, so we don't need to repeat those; just add the
HTTP status code (because it's not there already and sure why not),
and the Python exception type the client library chose to raise
in case that makes things a bit easier to interpret.