rcomr@koel.co.rmit.oz (Mark Rawling) (02/01/91)
Here are some bug fixes to cap50 lwsrv to fix some some problems
with the procset saving feature of lwsrv.
(Apologies if these bugs have already been reported, but we couldn't
find a fix at any of our usual sources)
In file simple.c ("$Header: simple.c,v 1.16 88/05/21 14:02:19 cck Rel $")
function scantoken(pf,echo,tptr)
Change the lines (385-6)
} else if ((tracefile != NULL) || echo)
dumpout(tokstr, i);
to
} else {
if ((tracefile != NULL) || echo)
dumpout(tokstr, i);
if (procsetfile != NULL)
fprintf(procsetfile, "%s%c", tokstr, crtolf? c: '\n');
}
This ensures that comments in Apple LaserPrep are saved into the FoundProcSet
files. Appledict #70 0 uses a comment (%endeexec) as a delimiter for the
stretch and smooth eexec hexstrings.
Change the lines (408-11)
}
if (procsetfile)
putc(c, procsetfile);
}
to
}
if (crtolf && c == '\r')
c = '\n';
if (procsetfile)
putc(c, procsetfile);
}
This fixes the "-T crtolf" option so that CRs do in fact get translated to
NLs in the FoundProcSet files.
While I'm on the subject of lwsrv, has anyone looked at adding an option to
defeat all this FoundProcSet stuff all together. We prep our laserwriter
with macs directly and only use lwsrv for off site "laser faxes".
We don't appreciate lwsrv prepending 35k of Laserprep to our 300 byte
faxes. Naturally we have fudged dummy FoundProcSet files that are
empty except for an appledict version header but this is pretty ridiculous
and will stuff up when a new version of Laserprep comes along. We will
probably end up doing it ourselves, but shouldn't it be an official option??
============================================================================
Mark Rawling, CSIRO Division of Information Technology,
High Performance Computation Group,
c/o Royal Melbourne Institute of Technology,
email: rcomr@koel.co.rmit.oz{.au}, phone: (+ 61 3) 660 2726