We have an incorrect value comparison when the stroke corner center
is adjusted. All coordinates in the stroke raster have been scaled by
the given values (sx, sy), while the stroke width remains at its
original size due to it not being clear how it should be scaled.
This brings wierd rendering results at times. Now fix it.
@Issue: https://github.com/thorvg/thorvg/issues/1336
Svgs without any viewbox and width/height information
have to be loaded before any other action is taken.
This is necessary to get the valid sive/viewbox info.
When reload is called from tvgPicture,
clear() of external_jpg loader is called and 'data' is freed.
In the past, there was no case where reload was called until bitmap() was called after load.
Currently I have to consider reload() for a none-viewbox svg case
So change the color myself because loader can't call open() again.
In case the "stroke" attrib is set to "none",
the width of the stroke is set to zero. Thanks
to that it isn't taken into account while
establishing the bounds of the shape.
For svgs with the width and/or height value set to zero
rendering was disabled - the load api return Result:Unknown
and draw - Result::InsufficientCondition.
Now an empty scene is added, so that both, load and draw,
return Result::Success.
Set colorspace to ABGR when colortype of lodepng is LCT_RGBA.
Since an image without an alpha channel becomes an ARGB colorspace with LCT_RGB,
it is the same as the default colorspace.
Since the color space is set at the time of specifying the target buffer of the canvas,
there is no way to know the color space when the picture is loaded.
So, check the color space applied to SwCanvas at the time of reload()
and change the color space.
There is an issue of BGR color space support for each loader.
The external_jpg loader resets the TJPF color space and calls read() to get a new buffer.
In the case of external_png, we need to change the color value directly
because it have to start over from begin_read_*.
This solution can affect performance as much as it access again image buffer
that have already been `read()` done. However, this only happens once.
../src/lib/gl_engine/tvgGlRenderer.cpp: In member function ‘virtual bool GlRenderer::renderShape(tvg::RenderData)’:
../src/lib/gl_engine/tvgGlRenderer.cpp:171:30: warning: ‘a’ may be used uninitialized in this function [-Wmaybe-uninitialized]
171 | drawPrimitive(*sdata, r, g, b, a, i, RenderUpdateFlag::Stroke);
| ~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
../src/lib/gl_engine/tvgGlRenderer.cpp:171:30: warning: ‘b’ may be used uninitialized in this function [-Wmaybe-uninitialized]
../src/lib/gl_engine/tvgGlRenderer.cpp:171:30: warning: ‘g’ may be used uninitialized in this function [-Wmaybe-uninitialized]
../src/lib/gl_engine/tvgGlRenderer.cpp:171:30: warning: ‘r’ may be used uninitialized in this function [-Wmaybe-uninitialized]
../src/lib/gl_engine/tvgGlRenderer.cpp: In member function ‘virtual void* GlRenderer::prepare(const tvg::RenderShape&, tvg::RenderData, const tvg::RenderTransform*, uint32_t, tvg::Array<void*>&, tvg::RenderUpdateFlag, bool)’:
../src/lib/gl_engine/tvgGlRenderer.cpp:215:21: warning: ‘alphaS’ may be used uninitialized in this function [-Wmaybe-uninitialized]
215 | uint8_t alphaF, alphaS;
re-design the shape data structure so that render backends
are able to access them directly.
This also let us remove tvgShape member data from the Shape::Impl.
To achieve this, migrate shape/stroke/path
from the canvas interface to the render interface.
If a paint is used as a clipper, it must be determined in the paint behavior.
Propagate its decision to the immediate derived classes so that
not only shapes but also scenes must be dealt as a clipper properly.
This revised this change 0de3872be3
for better a solution.
When we do 1.0 release, we clean up the coding style of all codes.
Currently, there are limitations in our coding style and lint settings.
This results in unnecessary bot comments.
According to the svg specs clip's fill and opacity
should be ignored. Till now setting the alpha/opacity
value to zero resulted in the shape's rendering abort.
@Issue: https://github.com/Samsung/thorvg/issues/1192
According to the svg standard the clips opacity
doesn't affect the final rendering. In order to
avoid any confusion the opacity values are
overwritten by the max possible value.
