Without this, any resize of the GLContextWidget might leave the label
somewhere it shouldn't be. Toggling fullscreen is a single example of
that behavior.
The breadcrumbbar in here serves exactly the same purpose as the one in
File Manager. Given that, I believe it's worth to keep these two
visually consistent.
Removed the old custom checkbox selection code in the Visualization
menu and replaced them with GUI::ActionGroup with set_exclusive
enabled instead :^)
This way we can feed it the values if we wanted to change an existing
Guide and handle the default as before.
That we have to pass a String here is a bit ugly.
The height of the bottom color container was cut off at the bottom.
This adjusts the heights and also the primary/secondary color
widget so that it's in line.
Add a new Color category to the Filter menu. Add the Grayscale
filter under Filter->Color to turn the selected layer to grey colors.
Created GrayscaleFilter.h. Modify Filter.h to allow filters
without parameters.
The prior commits add the `DrawMode` enum to keep track of where
the shape is being drawn from. With this addition, the prior `Mode`
enum name is confusing, so this commit renames it to `FillMode` to
be more explicit :^)
Essentially the same logic as the prior commit, but now for the
`RectangleTool` class. Pressing `alt` lets you draw the rectangle
with the starting position as the center.
Like other common image editing applications, now if you press
`alt` while drawing an ellipse, it uses the starting position as
the center of the ellipse as opposed to one of the corners of the
bounding rect.
The EllipseTool class now keeps track of a `DrawMode`, which is
either `DrawMode::FromCorner` (default), or `DrawMode::FromCenter`
(the option added by this commit).
The `draw_using()` function was modified to now take in the start
and end positions and construct the `ellipse_intersecting_rect`
itself, since we need to construct it differently based on the
drawing mode.
This change removes the manually created model class
in order to use a generic GUI::ItemListModel.
Besides from code reusability, it also makes the list
searchable as you type.
This is the first step in transitioning Piano to a full LibDSP backend.
For now, the delay effect is replaced with a (mostly identical)
implementation in LibDSP.
The new ProcessorParameterSlider attaches to a LibDSP::Processor's
range parameter (LibDSP::ProcessorRangeParameter) and changes it
automatically. It also has the ability to update an external GUI::Label.
This is used for the three delay parameters and it will become useful
for auto-generating UI for Processors.
As Piano will later move to the RollNote defintions of LibDSP, it's a
good idea to already insert velocity and pitch support, even though it's
currently not used.
The default action (shown in bold) indicates what would you get
by double-clicking on file. Since it's a default option, I think it
deserves to be on top (together with alternative launch options). :^)
Also they're not task actions like "Extract .zip here" or "Add to
bookmarks".
No previous model index will be invalidated afterwards, so avoid
invalidating them.
Also fixes an issue where committing to an edit with the inline cell
editor makes the focused cell switch to A0.
Fixes#9677.
Previously, if you used one of the keyboard shortcuts to select a
different tool, it didn't change the cursor to the corresponding
one till you clicked somewhere / did something else to trigger an
update. This was pretty jarring since there's no indication as to
whether the tool change was successful or not.
This patch just calls `set_override_cursor()` when a valid active
tool is set to immediately update it.
In general, I think `opt == x` looks much nicer than
`opt.has_value() && opt.value() == x`, so I'm updating
the remaining few instances I could find with some regex
magic in my search.
1) The Sound Player visualizer couldn't deal with small sample buffers,
which occur on low sample rates. Now, it simply doesn't update its
buffer, meaning the display is broken on low sample rates. I'm not too
familiar with the visualizer to figure out a proper fix for now, but
this mitigates the issue (and "normal" sample rates still work).
2) Piano wouldn't buffer enough samples for small sample rates, so the
sample count per buffer is now increased to 2^12, introducing minor
amounts of (acceptable) lag.
All audio applications (aplay, Piano, Sound Player) respect the ability
of the system to have theoretically any sample rate. Therefore, they
resample their own audio into the system sample rate.
LibAudio previously had its loaders resample their own audio, even
though they expose their sample rate. This is now changed. The loaders
output audio data in their file's sample rate, which the user has to
query and resample appropriately. Resampling code from Buffer, WavLoader
and FlacLoader is removed.
Note that these applications only check the sample rate at startup,
which is reasonable (the user has to restart applications when changing
the sample rate). Fully dynamic adaptation could both lead to errors and
will require another IPC interface. This seems to be enough for now.
Some tools (e.g. ZoomTool) doesn't need layer to work. This commit
makes mouse events fire even if there is no layer. This fixes
a bug that ZoomTool didn't work when there is no layers.
This commit adds a Tool::MouseEvent struct, which contains events that
may be needed by tools: layer-relative, image-relative and raw (editor-
relative) event.
The raw event is used by ZoomTool to properly pan the view. This fixes
a bug which caused image to snap out of sight.
