We use the same strategy Zulip already uses for starred messages,
namely, creating a new UserMessage row with the "historical" flag set
(which basically means Zulip can ignore this row for most purposes
that use UserMessage rows). The historical flag is ignored, however,
in determining which users' browsers to notify about new reactions,
and thus the user will get to see the reaction appear when they click
a message (and any reactions other users later add, as well!).
There's still something of a race here, in that if some users react to
a message while the user is looking at the unsubscribed stream but
before the user reacts to that message, those reactions will not be
displayed to that user (so counts will be a bit lower, or something).
This race feels small enough to ignore for now.
Fixes#3345.
If `render()` is called from middleware that runs before the
authentication middleware, then this code path will be called with a
request object where request.user is not yet set. Handle this by
providing a reasonable error message.
In aa880b0419, we used the raw
do_set_realm_description method rather than calling the API, which
meant that the API success path wasn't actually tested.
This adds an organization description field to the Realm model, as well as
an input field to the organization settings template. Added three tests.
Set the max length of the field to 100 characters.
Fixes#3962.
An empty narrow (ie, the home view) can be represented in code as either
`None` or `[]` but we had incorrect handling that failed to fully
properly deal with either case.
(1) In `get_stream_name_from_narrow`, we failed to deal with `None` by
trying to always iterate over `narrow`.
(2) In several other places, we failed to deal with `[]` by explicitly
checking `if narrow is None` or `if narrow is not None`. Changing these
to truthiness checks should work for both the `None` and `[]` cases.
A previous commit changed a `get` (which can throw `DoesNotExist`) to use an
existing object, but kept the `try` / `except` block:
4bf3ace444
Removing this unused code path allows us to achieve 100% test coverage.
Changing assert_in_success_response to require List[Text] instead of
Iterable[Text] prevents the following misuse:
self.assert_in_response_success("message", response)
Currently, this will check whether 'm', 'e', 's', 'a', and 'g' separately
appear in the response, which is probably not the intended behavior. The
correct usage is as follows:
self.assert_in_response_success(["message"], response)
This of course only works in the 2 minute window where missed-message
emails are planned, but nonetheless likely avoids common cases of
emailing users with deleted messages.
Fixes: #3873.
Tim Abbott
hollywoodno
sinwar
Rishi Gupta
Umair Khan
Yago González
Ayush Jain
adnrs96
Steven Ganz
Sarah Stringer
Jacob Hurwitz
K.Kanakhin
Elliott Jin
Susan Salituro
dattatreya303