These are chat archives for bigdataviewer/bigdataviewer-core
@tpietzsch What is the rationale behind choosing
bufferedImages in the
In principle, you could have all of these as
ArrayImg< ARGBType, IntArray >, and just copy into a new
BufferedImage before drawing to the screen, right? Was your decision to use the classes as above to avoid this additional copy? Or are there other effects that I am not aware of?
I am currently trying to write a
ViewerPanelFX that is essentially the
ViewerPanel in JavaFX, and I ran into the problem that the JavaFX
WritableImage does not share data when you pass an
int (instead, the data is copied), and thus I cannot just c&p
ARGBScreenImage and replace occurences of
WriableImage. In particular, this line would fail:
I found a (well hidden, since non api) interface
PlatformImage, which I could probably implement to get around this. I am debating with myself, if just using
ArrayImg< ARGBType, IntArray > as described above, and just create a
WritableImage before drawing onto the screen.
GraphicsContext2D(all javafx), bummer.
PlatformImageis off-heap probably