Re: Fix for 687364 rangecheck in buildshading2 from Quark XPress PS file

Dan Coby wrote:

 > I also worried about the other functions mentioned in TN 5044.
 > However I do not have any examples of files which will show a problem.
 > If you have any such files that you can send then please send them.

I know for sure that Amiable generates these operators but I don't have
sample files right now. Even if the program doesn't break on the
incomplete implementation, painting an vector object with a spot color
but a sample image with a process color is unacceptable.

>>(2) It is only a matter of time before we find a file that doesn't work
>>on the systems with findcmykcustomcolor, etc.

> I do not understand what you mean by this statement.  It seems to be
> a repeat of your first issue.  If not then please explain.

We have an file that requires findcmykcustomcolor.
It is possible to write a file that breaks in the opposite case: when
findcmykcustomcolor is defined. Murphy law and the prior experience
indicate that every imaginable PostScript abuse does take place.

Idiom recognition was designed to fix bugs (and inefficiencies) in
PostScript programs. I suggest to check whether it can be used in
this case.