This is a workaround for performance issues with the current implementation
when adding many regions to the selection one at a time.
If the Selection implementation was to change at some point and adding regions
to the selection only takes a small constant amount of time, then this
optimization may no longer be necessary.
Related to: #7274
This allows the clear methods to be used before calling ::add(), to avoid the
emission of a signal saying "there are no <foo> selected right now".
There should be no side-effects from this commit.
Note that correct use of this new API is complex, and requires avoiding the use
of wrapper methods like clear_objects().
Each MidiRegionView(MRV) is connected to the Selection::ClearMidiNoteSelection
signal that is used to notify the all MRV instances to clear their note
selection.
The MRV class also has a private static SelectionCleared signal that is used to
signal other MRV instances when their selection has been cleared. When the
Selection::ClearMidiNoteSelection signal is emitted it causes each MRV to also
emit the SelectionCleared signal. So the emission takes quadratic time.
With 1500 MRV instances emission takes about 2.2 seconds on my machine, and
some operations like track selection cause it to be emitted 3 times(another
issue).
The Selection class in the Editor knows which MRV instances have note
selections, as it is notified by MidiRegionView whenever the selection count
becomes zero or becomes non-zero. Clearing the Note selection should then just
be O(N) and direct calls can be used rather than signals.
This change removes both the signals and uses the existing references between
Selection and MRV class to control note selection. There should be no
behavioural changes in Midi note selection with this change.
- disallow simultaneous events via ControlList::editor_add ()
- clicking on an automation line selects the points that define it.
- don't 'flash' a region selection when using mousedraw mode.
- cp click selection resembles region selection.
- region gain points respect snap modifier (a la automation points).
Fix several other cases where a single mouse click could cause several
(not nested) selection ops.
Fix missing selection memento for midi notes and midi commands.
Rename some variables.
Fix random style issues.
The user can now replay *all* earlier selection operations until the next
session undo/redo command, or the completion of a new operation.
Nothing relating to selection ops is stored, and selection operation history
is begun on first idle.
Selection operation history is fundamentally different from the history of
operations which act on a selection in terms of both their viewport and the
amount of information required to replay them.
WRT undo, the user of a selection op doesn't care about the viewport state
at the beginning of an op, but rather that at the end of the previous one.
combines selection related editor properties with the current editor selection.
The related editor properties are:
mouse mode,
zoom setting,
left frame of the canvas,
y origin of the canvas.
Selection state now includes region views (storing the underlying region id)
and time.
This patch also fixes a region mute undo bug.
Among other things, this means that automation controls/lists have the actual
min/max/normal/toggled of parameters, and not those inferred from the Parameter
ID, which is not correct for things like plugin parameters.
Pushing things down to the Evoral::ParmeterDescriptor may be useful in the
future to have lists do smarter things based on parameter range, but currently
I have just pushed down the above-mentioned currently used attributes.
The idea here is to do the reasonable thing, and copy objects of some
type (e.g. MIDI region, gain line) to tracks with a matching type. The user
can override this with a track selection, which will be used straight-up.
Lost: ability to copy/paste lines across types, e.g. gain to pan. This is
often questionable, but sometimes useful, so we will need to implement some
sort of "greedy mode" to make it possible. Implementation simple, but not sure
what to do. Perhaps this should only be possible if one automation track is
explicitly (i.e. via track selection) involved, and the types are at least
compatible-ish?
This cleans up a lot of false-positives in static analysis
and also helps compilers to optimize code paths in general.
(tagging the fatal stingstream operator as ‘noreturn’ is
far less trivial)
