>How does one get Macsbug to trap when CopyDeepMask is called, then
>disassemble that instruction?
>
>I need to see if it is calling OffsetRect at some point depending on OS
>8.1, versus 8.6
>
>I am getting some strange behavior with CopyDeepMask in 8.1 vs 8.6, and the
>only thing I can think is that in OS 8.1, CopyDeepmask is offsetting the
>Mask rect to full left if it isn't. This results in some bad behavior when
>I drag to the left of my image and adjust the source and mask rects.
>
>I don't know for sure other than it looks like my mask rect is getting
>offset something serious behind my back.
>
>?
>
>rc


I do believe the problem I have been experiencing is that CopyDeepMask has
a bug whereby if the mask GWorld is a different depth from the Source, and
only if the mask rect is offset from full left....(not at  0), then it will
skew the copy all over the place.

The solution unfortunately is to make the mask GW 32 bit also, instead of
the 8 bit I was using.

I think this will probably solve my OS 9 issues too I bet.

What a bummer. Anyhow... nice to know the problem.

Funky that it works -fine- in OS 8.6 at any depths, but not in 8.1, or 9
unless source and mask GWorlds are the same. Then again, SeedFill is
screwed up in OS 8.6.


Beware. :)


rc