lout-users
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: lout output in ghostscript


From: Valeriy E. Ushakov
Subject: Re: lout output in ghostscript
Date: Fri, 1 Oct 1999 21:26:38 +0400

On Fri, Oct 01, 1999 at 06:36:43PM +0200, Hanu? Adler wrote:

> > ... my guess is that your printer has Times-Roman _with_ necessary
> > diacritic characters, while ghostscript has Times-Roman _without_.
> > You can try prfont.ps from ghostscript distribution: send it to
> > printer and ask it to print a Times-Roman for you (copy that file
> > and add /Times-Roman DoFont at the end).
> 
> You are probably right, but I must presume that my printer is an
> exception and people who have other printers will not be able to read
> postscript created by lout.

Yes, unless fonts are included in the document.  See below.


> I'd suggest that some latin2 ps fonts should be included in the lout
> distribution ...

Several people from .pl and .cz that I talked to use QuasiPalladio
family with Lout.  I guess this font is sort of standard for Latin2
users.  I don't think that the fonts themselves belong to the lout
distribution.  It makes more sense to package them separately, so that
lout only provides an LCM and a separate package provides the AFMs and
fontdefs.  (What's the license status of QuasiPalladio, BTW?)

Ideally (but that's only my humble opinion), Lout should only provide
AMFs/fontdefs for Adobe's base 14 fonts, because that's what is
guaranteed to be available in every PostScript system.  This might
seems to be overly minimalistic, but people periodically come with
questions like: I use Palatino in my document and it doesn't actually
print Palatino.  If Palatino metrics are not distributed in the
standard set, then people have to make a conscious decision of using
Palatino (if they happen to have it) or using, say, URW Palladio or
QuasiPalladio.  I feel that this should really be an add-on packages.

I recall someone recently proposed (volunteered? :-) to prepare such
add-on packages for popular font families, so doing this for Latin2 or
Cyrillic will be in line with that proposal and, I think, with the Tao
of Lout.  I was somewhat cold towards the idea at the time, but giving
it a second thought, I now think that's good to have a packaging
scheme like that.

What others think about it?


> ... and whenever a latin2 document is being created, the font
> definition should be placed inside the postscript document.  Just as
> it is done by dvips (from LaTeX I get perfect documents readable
> just everywhere).

Lout adds %%IncludeResource comments for fonts that the document uses
and I think this is the correct and desired behaviour.

Spooler can use a postprocessor, like includeres(1) from PSUtils, to
actually include resources reffered to by the document.  Or, if you
have, say, QuasiPalladio family downloaded to your printer, then you
can configure your spooler to not include those resources
(QuasiPalladio in this case) that are already printer resident.

Since Lout emits %%IncludeResource comments on a per-page basis
(except for the first page), blindly running includeres(1) could
increase the size of the file *significantly*.  While this is ok when
you send the file to printer (and even desirable, since fonts won't be
cluttering printer's memory), this is inappropriate for posting the
document online.  So a simple script can be written to move
%%IncludeResource comments to the document setup section and then
includeres(1) will only include one copy of each font.

The %%IncludeResource approach provides, in my opinion, greater
flexibility and clear separation of concerns where Lout do what it
have to and document management system do what it have to.  The
document management system is this case could be either a spooler or
you (posting document online) or something else.


> Or do you have some other solution?

I hope I addressed your concerns.


> > Re quotedblbase.  For some reason it's not included into the map file.
> > Just add it at any free place in LtLatin2.LCM (C0 (01-31) or C1
> > (128-159) areas are fine).
> 
> This has helped, thanks.

This should be fixed in the shipped LtLatin2.LCM.


SY, Uwe
-- 
address@hidden                         |       Zu Grunde kommen
http://www.ptc.spbu.ru/~uwe/            |       Ist zu Grunde gehen


reply via email to

[Prev in Thread] Current Thread [Next in Thread]