[comp.archives] [sun] Re: Pageview on ps-files produced by dvi3ps

primer@math.harvard.edu (Jeremy Primer) (12/13/90)

Archive-name: tex/dviware/dvips/1990-12-12
Archive: labrea.stanford.edu:/pub/dvips54.tar.Z [36.8.0.47]
Original-posting-by: primer@math.harvard.edu (Jeremy Primer)
Original-subject: Re: Pageview on ps-files produced by dvi3ps
Reposted-by: emv@ox.com (Edward Vielmetti)

In article <6650@emory.mathcs.emory.edu> cheung@mathcs.emory.edu (Shun Yan
Cheung) writes:

   The output of pageview on a ps file produced by dvi3ps is not a pretty
   sight. All the pages are display on top of each other and pageview
   also kindly tells you that it is displaying the first page of a
   one page document. Is there a dvi-to-ps program that produces
   pageview-able ps output ? Or maybe is there a fix in pageview
   available that can display dvi3ps outputs ? Many thanks.

The author of pageview has said that the next release will have an option
to allow for viewing non-conforming PostScript files.  But the main
problem is with the PostScript from dvi3ps, and not with pageview.

We have just been through all this here.  A very good answer, perhaps even
the best answer, is to use T. Rokicki's dvips program.  Ftp all four dvips
tar files from labrea.stanford.edu or neon.stanford.edu, and read the
32-page manual.  dvips will do everything dvi3ps does, with the possible
exception of chtex (chinese TeX).  It should be possible to make that work
too.  Note that you have to change your PSlatex macros (psfonts.sty) to
the ones in the dvips distribution, using the virtual font system.  Do not
expect dvips to process old .dvi files which used any Postscript features.

The thorniest point is that dvips DOES support psfig/TeX inclusion of
PostScript files (as well as just about everything else), but the version
of the psfig macro needed is different (psfig.tex itself, and not merely
some PostScript header file).  The solution is to get the dvips version of
psfig (from upenn dist site), which was designed for Arbortext's
commercial "dvips" but works for this one as well, but to forget about the
PostScript header provided with it because Rokicki's "special.pro" header
has all the psfig macros defined already.  Then trash the dvi2ps-li
(a.k.a. dvi3ps) version of psfig and tell users to reconstruct any .dvi
files which make use of PostScript features or fonts.  Do not really try
to run both TeX drivers on the same system.

If you are not the system administrator, please show this to your system
administrator and inform him that dvips produces *conforming* PostScript
files which are 30-80% shorter than those from dvi2ps-li, that it is
blindingly fast, that it solves numerous other problems.  It allows
separate configuration files for each printer (to deal with, for example,
old Apple LaserWriters with almost no Virtual Memory) and in fact has the
most impressive and consistent set of configuration options that I have
*ever* seen in a Unix program.  (The MS-DOS version is also the subject of
rave reviews.)  The only adjustment we may be making for our users is to
wrap it in a shell script which will recognize the "-f" and "-t" options
they are accustomed to for specifying which pages to print (these have
different names in dvips) and a couple of other nifty options.
--
Jeremy Primer, Department of Mathematics, 1 Oxford Street, Cambridge MA 02138
primer@math.harvard.edu     ...!harvard!zariski!primer    primer@huma1.bitnet

chiu@brahms.amd.com (Timothy Chiu) (12/13/90)

Archive-name: tex/dviware/dvitops/1990-12-13
Archive: june.cs.washington.edu:/tex/dvitops.tar.Z [128.95.1.4]
Original-posting-by: chiu@brahms.amd.com (Timothy Chiu)
Original-subject: Re: Pageview on ps-files produced by dvi3ps
Reposted-by: emv@ox.com (Edward Vielmetti)

Sorry, typo in that last followup, should have read:


I'm currently using a package called dvitops written by James Clark.
It works with pageview (i.e. I don't get overlapping pages, but
even so it's limited to Sun's resolution).  It's available
anonymous ftp from june.cs.washington.edu (128.95.1.4)
in ./tex/dvitops.tar.Z


Timothy Chiu		Advanced Micro Devices
chiu@amd.com		P.O.Box 3453 M/S 167 Sunnyvale,CA 95133