RenderRootTest has tests for verifying behavior of getRenderRoot() method when dirty region is an empty rect i.e. (0, 0, -1, -1)
tests: emptyDirtyRegion1, emptyDirtyRegion2, invalidDirtyRegionOutsideOpaqueRegion
See RectBounds.makeEmpty() and RectBounds.isEmpty() for definition of an empty rect.
Current behavior is that when an empty dirty region rect is passed to root.getRenderRoot(), it returns the root itself in NodePath, which means that entire scenegraph tree of the root shall be rendered.
Other behavior could be that no rendering should happen when the dirty rect is empty.
We need to choose a behavior and make related changes.
As of now we think current behavior is more correct and changing it is more likely to cause regressions. If we go with this approach then it will only be test bug.
tests: emptyDirtyRegion1, emptyDirtyRegion2, invalidDirtyRegionOutsideOpaqueRegion
See RectBounds.makeEmpty() and RectBounds.isEmpty() for definition of an empty rect.
Current behavior is that when an empty dirty region rect is passed to root.getRenderRoot(), it returns the root itself in NodePath, which means that entire scenegraph tree of the root shall be rendered.
Other behavior could be that no rendering should happen when the dirty rect is empty.
We need to choose a behavior and make related changes.
As of now we think current behavior is more correct and changing it is more likely to cause regressions. If we go with this approach then it will only be test bug.
- relates to
-
JDK-8233362 Evaluate ignored graphics tests
- Resolved
-
JDK-8234077 Evaluate ignored unit tests in RenderRootTest
- Resolved