As now the RgbMap is generated correctly depending on the current layer
(if it’s opaque or transparent), we’ve fixed
LockAlpha/Transparent/Merge/Blur/Replace/JumbleInkProcessing for
Indexed images with a background layer, when a transparent layer is the
active one, and when a non-mask index has same RGB values than the mask
index.
Now the remap is calculated from the old palette to the new palette when
the "Remap" button is pressed. In this way we can remap after loading
a new palette (fix#737), or after generating an optimized palette, or
changing colors (fix#563).
With this change now we add a "mask" image/parameter in rotation
functions. In this way we can identify which specific pixels are
inside the original mask/selection, and in opaque mode we can
include/scale/rotate all those pixels inside the mask, whatever
value they are, even if they are the mask color.
Fixes#730
- Correct usage of disposal methods (and possibility to generate records
with "restore previous frame" disposal method)
- Create color maps for each frame when necessary
- Remove options to dither RGB images (to simplify encoder)
- Better palette creation depending on local colormaps
- Reuse palette colors from previous frames
- Use frame transparent index correctly
- Convert to RGB automatically if the image has more than 256 colors
- Remove onPostLoad()
- Add progress bar feedback
With this change we save the new palette chunk when it's necessary, and
the old color2 chunk will be always present only for backward
compatibility. (So we can open a .ase file saved with v1.1 in v1.0.)
Instead of calling findBestfit() for all RGBA colors in
RgbMap::regenerate(), we mark all entries as invalid, and then we
validate them only when the user require one specific entry from
RgbMap::mapColor(). With this we avoid a lot of unnecessary computations
each time the palette changes.
This include several changes:
- Color::getIndex() can return palette values with alpha != 255
- Fix Transparent and Blurs ink for indexed images to make better use
of palette entries with alpha values
- Fix bilinear resize algorithm for indexed images with alpha
- New RgbMap with four parameters: R, G, B, A
- Add one extra color scale function used in the alpha channel of the
new RgbMap
- Fix color curve, convolution matrix, invert color, and median filters
to take care of this new alpha channel on indexed images
- Fix ordered dithering and quantization
Related to #286
We've added a new chunk type in .ase files to save palettes with alpha
channel, color name, and palettes with more than 256 colors.
Related to #668, #467, and #286
- Added doc::BlendMode enum and doc::BlendFunc type
- Renamed LayerImage::getBlendMode() -> blendMode()
- BLEND_MODE_COPY is BlendMode::SRC now
- BLEND_MODE_NORMAL is BlendMode::NORMAL now
- Added app::cmd::SetLayerBlendMode
This problem was introduced in d732f5b05f
Now image_iterator.h and image_bits.h use forward declarations. And if
someone want to use the Image implementation, ImageBits, LockImageBits,
ImageIterators, or some function inside primitive_fast.h, they just have
to include image_impl.h file.
Now the StatusBar observes changes in the cel opacity to update the
slider (e.g. when we change the cel opacity through Cel > Properties
menu, the StatusBar refresh its slider).
This commit includes the new doc::PalettePicks class. It's a helper
class to identify which entries are selected in the ColorBar.
We've removed the Copy/Paste buttons from the Palette Editor window.
When we create a CompressedImage from a Brush image, we want to know what
pixels are != the mask color, just to generate the PointShape (we don't
care the specific color of the scaneline/each pixel, that information is
used in the "ink processing" step).
Now the ContextBar contains a set of brushes. The ChangeBrushCommand
supports a new "slot" parameter and "change" = "custom" to select a
specific custom brush from the ContextBar. Alt+1, Alt+2, etc. are mapped
to this ChangeBrushCommand (see changes in gui.xml).
Also, as the ButtonSet that represent different brushes in the ContextBar
uses icons generated from the brush, we don't need the skin parts that
represent each brush type (we can generate those icons from some standard
brushes). Those skin parts were removed.
Compiler could not find notifyActiveDocumentChanged.
/home/aurelien/src/aseprite/src/./doc/test_context.h: In instantiation of ‘void doc::TestContextT<Base>::onRemoveDocument(doc::Document*) [with Base = app::Context]’:
/home/aurelien/src/aseprite/src/app/document_api_tests.cpp:66:1: required from here
/home/aurelien/src/aseprite/src/./doc/test_context.h:44:58: error: ‘notifyActiveDocumentChanged’ was not declared in this scope, and no declarations were found by argument-dependent lookup at the point of instantiation [-fpermissive]
notifyActiveDocumentChanged(m_activeDoc = nullptr);
This is a first step to include the current selected range of layers
and frames inside the doc::Site structure. So in the future we'll not need
to know about the Timeline in commands that handle those ranges.
Changes:
* Add DataRecoveryView
* Split SubObjectsIO into an interface and SubObjectsFromSprite
* Add a "setId" parameter to some doc::read_* functions (we need this
functionality from session recovery)
Changes:
* Paint FrameTags in Timeline with labels
* Add app::ui::FrameTagWindow
* Fix FrameTag::m_aniDir initialization
* Add warning for files that doesn't support frame tags
* Remove document preferences related to the active loop
* Created base::write24bits() function (and detect endianness with a
special base/config.h.cmakein)
* Use _WIN32 instead of ALLEGRO_WINDOWS
* Use _DEBUG instead of DEBUGMODE
* Replaced AL_CONST with "const" in bmp_format.cpp
* Replace stricmp() with base::utf8_icmp()
Images must be stored independently of cels. So read/write_cel() just
read/write the Cel information (with image ID only), and the image data
must be read/written by the caller.
Undoers were little objects to swap/revert an action. They didn't
execute the action itself, they just revert its previous state. Now
undoers were replaced with cmds: A cmd is an object that
executes/undoes/redoes just one action.
Changes:
* Remove old undo library and app/objects_container_impl.cpp
(now we use the doc::ObjectId directly to store undo info)
* Remove all Undoers from app/undoers/
* Replace DocumentApi impl with little Cmds in app/cmd/, these
cmds handle execute/undo/redo of each action at the logic layer
* Remove doc::Dirty object
* Remove doc::Settings: all undo configuration is in the app side
* Move undo options from app:ISettings to app::Preferences
* Rename UndoTransaction to Transaction
* Create a CmdSequence to store a sequence of Cmds (as now the new
undo library doesn't support open/close groups)
* Add doc::get<T>(ObjectId) function to get any kind of object
from the doc library by its ID
* Add Cel::document() and Sprite::document() members
* Add Sprite::cels(frame_t) to get all cels in the given frame
* Add Layer::displaceFrames() member function
* Move the "allow non-linear history" flag from undo2::UndoHistory
to app::DocumentUndo
Changes:
* Add doc::ImageRef to count references to the same image between Cels
(at this moment we cannot generate linked cels anyway)
* Remove doc:Stock class and doc::Sprite::m_stock member variable
* Remove app::undoers::Add/RemoveImage
* Add doc::SubObjectsIO and app::undoers::ObjectIO to
replace doc::LayerSubObjectsSerializer
Changes:
* Create render library (move util/render.cpp to render/render.cpp)
* Move app::Zoom class to render::Zoom
* Remove doc::Image::merge() member function
* Add gfx::Clip helper class (to clip dst/src rectangles before a blit)
* Move doc::composite_image() to render::composite_image()
* Remove doc::Sprite::render()
* Replace Sprite::getPixel() with render::get_sprite_pixel()
* Remove doc::layer_render() function
* Convert DitheringMethod to a enum class
* Add AppRender to configure a render::Render with the app configuration
* Move checked background preferences as document-specific configuration
* Add doc::Sprite::layer() and palette() member functions
* Add doc::Layer::cel() member function
* Add doc::Palette::entry() member function()
* Add doc::frame_t type
* Move create_palette_from_rgb/convert_pixel_format to render library
* ExportSpriteSheet doesn't need a temporary image now that we can specify
the source rectangle in the render routine
Here we avoid copying and clearing pixels that will not be used
in the whole tool loop process.
Changes:
* Add several member functions in ToolLoop to validate/invalidate regions
of source/destination images so we know what regions are safe to use
by inks and can be shown in the editor
* Add new DocumentObserver::onExposeSpritePixels() member to validate
pixels that will be displayed in the editor
* Add Ink::needs/createSpecialSourceArea() member functions to validate
extra areas for inks like blur or jumble
* Add undoers::ModifiedRegion to save the undo information about the
modified region
* Add ShowHideDrawingCursor class
* Change "blur" tool policy from overlap to accumulate
(This is a real fix for issue #239)
This avoids to open a document two times on Mac OS X
from the command line (the file is loaded when we
process the command line, and by Finder, which send us
an application:openFile: message).
Changes:
* Add doc::Sprite::bounds()
* Use gfx::Point to store the doc::Cel position
* Replace "int x, int y" with a "const gfx::Point&" param
* Fix Mask::intersect(const gfx::Rect&) and Mask::crop(const Image*)
Fix a problem selecting one cel in the timeline, using Ctrl+A and then
Ctrl+C to copy the whole image area. (This indicates that the user want
to copy the cel content, not the timeline cel.)
The objective of these changes is to create a more testable doc/logic API.
Included changes:
- Added doc::Context, doc::Documents (the old app::Documents),
and doc::Sprites (and observers for each)
- Added raster::Sprite::createBasicSprite()
- Added doc::ColorMode (equal to raster::PixelFormat)
- Added some methods to doc::Document: context(), sprites(), sprite()
(to replace app::Document::getSprite()), width/height(), colorMode(),
name(), and close()
- Moved app::DocumentObserver/Event to doc::DocumentObserver/Event
- Replaced app::ContextObserver with doc::DocumentsObserver and a couple
of signals.
- Renamed app::Context::getActiveDocument() with
doc::Context::activeDocument()
- Renamed app::Context::getActiveLocation() with
app::Context::activeLocation()
- Removed app::ContextObserverList
- Removed app::DocumentId (now we can use doc::ObjectId)
- Removed app::Context::getSettings()
- Add doc::ExportData with information about the last export operation.
- Add RepeatLastExport command.
- Add SaveFileBaseCommand as base Command class, to accept "filename"
param.
- Add parameters to ExportSpriteSheet so we can call it without UI
from "repeat last export".
