This would fail building with Visual Studio 2017, at least downstream in Godot
where we undefine old Windows compilers' non-standard `min`/`max` macros (see
`minmax.h`/`NOMINMAX`).
Change id() to identifier().
error log
```
../src/lib/gl_engine/tvgGlRenderer.cpp:333:19: error: ‘const class tvg::Fill’ has no member named ‘id’
333 | switch (fill->id()) {
```
The reference node doesn't have to be placed inside the defs block,
so when refering to it, it is necessary to search the entire node tree.
In case of invalid svg file this can lead to circular references.
Added prevention against such a sutuation.
Some gcc versions and configurations was giving an error:
error: "comparison of integer expressions of different signedness: 'const SwCoord'
{aka 'const long int'} and 'uint32_t' {aka 'unsigned int'} [-Werror=sign-compare]"
This patch should fix the error
For the performance reasons, the LumaMask is used only when
it's necessary - when the mask color is other than white.
Otherwise the AlphaMask is enough.
In an svg file the mask-type attribute can be specified.
It takes one of the two values: Luminosity or Alpha.
After the LumaMask is introduced into TVG, this attribute can be
properly read by the svg loader.
The SvgCompositeNode is replaced by the SvgMaskNode and SvgClipNode.
This is needed for using Luma/AlphaMask in the svg loader and in the future,
when we introduce other mask's features.
Introduced CompositeMethod::LumaMask that converts the source pixels to the
grayscale (luma value) before alpha blending. Thanks to it, mask works more like
typical mask in graphics editor software.
Grayscale is calculated with weighted method:
(0.0721*B + 0.7154*G + 0.2125*R) * A
Introduced surface->blender.lumaValue function
Basically, this Accessor is a utility to debug the Scene structure,
by traversing the scene-tree by users.
You can search specific nodes to read the property information,
figure out the structure of the scene tree and its size.
We actually don't recommend you to touch the property unless you really
know the each paint's position and role because it's not visible, difficult to
understand its anatomy.
Also, You must underatnd that modifying the nodes of the scene will be going
well with both the art-design structure and the prorgram logic.
In this first version, Accessor only supports for the Picture class.
@example:
auto picture = tvg::Picture::gen();
picture->load("test.svg");
//The callback function from lambda expression.
//This function will be called for every paint nodes of the tree.
auto f = [](const tvg::Paint* paint) -> bool
{
if (paint->identifier() == Shape::identifier()) {
//override properties?
uint8_t r, g, b, a;
paint->fillColor(&r, &g, &b, &a);
paint->fill(r / 2, g / 2, b / 2, a);
}
//You can return false, to stop traversing immediately.
return true;
};
auto accessor = tvg::Accessor::gen();
picture = accessor->access(move(picture), f);
...
@Issue: https://github.com/Samsung/thorvg/issues/693
Handle correctly duplicated spans from the multiple y span data.
Previous logic only expected the one single y span data from the rle.
However rle might have multiple y span data if the anti-aliasing is applied.
This patch also removed the bad design of the common engine
which handles the anti-alising option to ignore the anti-aliasing rle generation.
Just realized, it's difficult to control that condition due to scene-composition.
