steve@fnord.umiacs.umd.edu (Steve D. Miller) (08/06/90)
Archive-name: troff-fix/06-Aug-90 Original-posting-by: steve@fnord.umiacs.umd.edu (Steve D. Miller) Original-subject: Re: Does troff work? Archive-site: fnord.umiacs.umd.edu [128.8.120.3] Archive-directory: /pub Reposted-by: emv@math.lsa.umich.edu (Edward Vielmetti) Interesting. This fix is the one I wanted to use, but I couldn't figure out the format of the ASCII font width tables without sources, and our sources hadn't yet arrived, so... My approach was to punt the Ultrix troff, replacing it with the BSD version, plus some really important bug fixes. I then hacked some Makefiles in TranScript to run a COFF-to-a.out mangler I wrote (which is sufficient only for font width tables, folks, since they don't have to look like real binaries). Once I did all that, we could troff just fine. After I had things working, my motivation to change my fix dropped dramatically; I'd really rather do almost anything than work on text processing software, and if I have to work on text processing software, I'd really rather work on almost anything other than troff. In any case, if you'd like, anonymous FTP to fnord.umiacs.umd.edu and grab the files psfixfont.c and trofffix.shar from /pub. The former is the COFF->a.out font width table converter, and the latter is the required set of fixes to the 4.3BSD troff that will make it run on a DECstation. Actually, DEC didn't modify the font width table format -- the DECstation Ultrix troff sources all say, 'copyright MIPS Computer Systems' on them. I suspect that MIPS modified the format, though perhaps someone else (AT&T?) did so even before they did. I tried really hard (two or three times!) to get RISC troff fixed during the 4.0 field test, and failed miserably. Argh. -Steve -- Spoken: Steve Miller Domain: steve@umiacs.umd.edu UUCP: uunet!mimsy!steve Phone: +1-301-454-1808 USPS: UMIACS, Univ. of Maryland, College Park, MD 20742 New phone (August 6thish, 1990): (301) 405-6736