warren@UUNET.UU.NET (12/27/89)
I have GhostScript 1.0 of Aug 10 88. I wrote a ega/vga driver for Xenix. Would someone like it? One thing lacking in the gx_device_procs struct is a function that restores the display to the mode that was in effect before you ran gs. I added a new field for this purpose. The code gave Xenix's C compiler (which is really Microsoft C 5.something) a headache by typedefing structs that weren't defined. In the end I managed to get around this by removing all the "typedef struct foo_s foo" and putting "struct" in front of every "foo". It wasn't pretty and I may have caused the following bugs in the process. Some example .ps files cause core dumps. I think there are two problems. The first is that if you try to show a space (e.g. book14.ps), igs->path->first_subpath is null and as a result gfill dumps. I verified this by removing the blanks and seeing that it worked. The other dump happens in let[qv].ps. It's also due to a null first_subpath but I can't quite figure out what is happening. If there have been any fixes that deal with this, or a more recent release of ghostscript, please let me know. If someone cares to take the time to explain how the /space logic is suppose to work I would be very grateful.