Ok, I just checked in code to do this. (And moved most of the
relevant code from fv_View.cpp to fg_GraphicRaster.cpp).
As I was testing, I noticed it would assert sometimes at the top of
fp_ImageRun::findPointCoords. It wasn't reliable; I couldn't get it
to happen while I was running AbiWord in gdb. Maybe I was doing
something slightly differnt though. It may be related to bug #407
too. I don't think I changed anything which would cause it.
> This will not address the real problems being raised on this thread,
> but it will effect the original [arguably broken] behavior which
> was intended.
I think the above behavior with the possibility of resize from within
AbiWord should be sufficient. Oh, it should probably look at pHYs
first too.
-- Matt Kimball mkimball@xmission.com