Just recently, I started using g_return_val_if_fail as a brief assertion
checker. It'll also exit the function with a specified return value.
But actually this introduces some weird behavior. It's configurable by
environment variables and it'll print out a log message, if the
expression didn't validate properly. But some of these assertions are
actually ment to be silent.
Using a simple macro makes it simple to structure the assertions and its
return values in a block at the start of a function or anywhere else.
These aren't used anymore. Any notification which gets moved from
waiting to displayed, will have set the start field with the current
monotonic time.
So testing start == 0 won't ever succeed in
queues_notification_is_finished, as the tested notification is contained
in the displayed queue. So the start field never will be 0.
Also there's no semantics for start being 0 in dunst actually
implemented.
In previous releases when a replace request came in with an id that
doesn't exist we created a new notification with it anyway.
This is used by some to imitate the behaviour of `stack_tag` and while
not recommended (as it will break if another notification gets assigned
that id) we want to avoid such subtle breakages without consideration.
This bug was introduced in d879d70da060ea78fe735d62249a0afdf3e61bc8.
When the display queue had is full and a new notification would come
in, new notification wouldn't get shown until a currently displayed
notification would timeout.
Even if the notification would have been shown on top of the displayed
queue. So e.g. if the displayed queue would have been filled with
"normal" urgency notifications, an incoming "urgent" notification would
have been delayed.
To let those more important notifications through, the tail of displayed
and head of waiting are swapped on every update if necessary.
For notifications without any timeout, ttl can't be positive.
So when, show_age_threshold is active, dunst won't wake up, if there
are only notifications displayed without any timeout.
On Linux, CLOCK_MONOTONIC has got a bug. It does not count
onwards during sleep, albeit required by POSIX. This behavior
is reasoned with the requirement for poll().
Also the GLib people are sticking to this behavior in their
g_get_monotonic_time() function.
So we have to use a drop in replacement, which respects CLOCK_BOOTTIME
on Linux, as this is the clock, what would be CLOCK_MONOTONIC on POSIX
systems.
The signal_notification_closed function is called even when a
notification has been pulled from history and re-emits the
NotificationClosed signal, this can confuse clients since multiple such
signals are not accounted for in the spec.
Notification spec prohibits to reuse notification IDs (unless uint32 is
exhausted). Therefore returning the same ID twice must not happen.
Sending a signal, that the old notification timed out makes most sense.
It wasn't closed by the user, nor by a CloseNotification call either.
When we stack notifications, no interaction happened (the equivalent
of timing out).
