[vwnc] Spell Checker doesn't port between platforms

[Forgot to send this to vwnc.]

I have an image with the new highlighting capabilities loaded, but with slightly older versions than Martin’s final ones, specifically with SpellChecker2 v14.  I built it on a Windows box.  In it I have a workspace (also copied) with spell checking enabled.  I’ve copied that image to a Sun box.  When I start the image, I had to close the pane to the workspace file and reopen it.  (So much for file name portability.)  When the contents are spell checked, I get notices of errors in the Transcript.  If I disable ignoring of errors on the Highlighters settings page and apply, I get the notifier for the exception.  It’s a subscript out of bounds error following the letter chain in the vocabulary.  Assuming this might be an endianness problem, I re-initialized the SpellChecker2.Default vocabulary, but this did not clear the problem.

Is accessing of these offsets or other values within the vocabulary platform endianness specific?  Intel (Windows) is little-endian and the Sun is big-endian.

Cheers!