[tex-k] dvips Z option and Zandt's alt-rule.pro revisited

Michael Shell list1 at michaelshell.org
Sat Mar 6 14:53:35 CET 2004

On Fri, 05 Mar 2004 14:59:01 -0800
"Tomas G. Rokicki" <rokicki at CS.Stanford.EDU> wrote:

> And just because the bugs are in old interpreters doesn't mean we
> can ignore them; there are still an awful lot of old LaserJets
> around, with ancient interpreters, that need to generate the
> correct output.

I agree totally with this as I for one still admire and use the LaserJet
II/III series.

Perhaps Zandt can be asked what type of bug reports he has received
(if any) concerning alt-rule.pro?

Nevertheless, in my tests, his algorithm produces noticably superior
results on my LaserJets (including a more modern Level 2 HP 4si).
I have also run some tests with very wide (like 100pt) rules and
Zandt's algorithm still worked wonderfully.

What irks me is that -Z loads texc.pro and I seem to have no control
over this in the config file. I can specify tex.pro or texc.pro as
what is loaded by default in the absence of any options, but -Z pulls
in texc.pro at the end regardless of what is in the config file. This
is very unexpected and confusing - and limits flexibility too.

It would be nice to be able to specify the "-Z" .pro file in the
config file. In this way, dvips can be shipped with *four* primary
.pro files:



and one could specify which of the two sets are used via the config
file. One could even make the Zandt versions the standard versions
and provide the old versions as tex_classic.pro, etc. to let the
user "roll back" to the old versions if problems develop - and
we would find out soon enough if there are any interpreters which
choke on Zandt's code.

Another approach would be to change the behavior of -Z to load
right after the first loaded .pro file (or just after tex.pro,
or load texc.pro in place of tex.pro - identifying tex.pro by
name rather than by its order in the config file) instead of simply
loading texc.pro at the end - who knows what other tex.pro patch
code users might want to try in the future?

 Mike Shell

More information about the tex-k mailing list