Hi, > > but we had all hells job getting it to work correctly on little endian > > machines (i.e. we needed a way to define _BIG_ENDIAN correctly). On the > > I chose the test > #if X_BYTE_ORDER == X_BIG_ENDIAN > because it was already present in the code, in > gdk-pixbuf-xlib-private.h, so the macros seemed to be supposed to be > defined. Do you mean this can test positive on a little-endian machine? > (I cannot try it out, I have no such beast).
On the tests I've done (not sure which ones Franz has done), I was finding that my test machines were using the BIG_ENDIAN code and then the SMALL_ENDIAN code and getting totally confused by everything! > > testbed MacOSX box I was finding problems getting the patch to work > > correctly either. > > What problems were these? I tried it on several MacOSX machines with > several different versions of X11 or xfree86 installed. The main problem was that I was remotely doing the work (it's on a friends machine who has allowed me remote access). After about 2 hours of twiddling with his machine, everything was fine. He had screwed quite a few things up... The patch works fine under OSX. TTFN Paul -- One OS to fool them all One browser to find them One email client to bring them all And through security holes, blind them...
