Pstricks and OzTex Again

Denis Girou Denis.Girou at idris.fr
Fri Jul 10 22:04:35 CEST 1998


-----------------------------------------------------------------------------
This is the PSTricks mailing list, devoted to discussions about computational
graphics in (La)TeX using the PSTricks package from Timothy van Zandt.
For help using this mailing list, see instructions at the end of message.
-----------------------------------------------------------------------------

>>>>> "Jean-Francois.Mertens" == MERTENS Jean-Francois <jfm at core.ucl.ac.be> writes:

    Jean-Francois.Mertens> I was curious about this, and tried pst-tex on my own machine
    Jean-Francois.Mertens> _ confirming essentially Denis Girou's mail, except that
    Jean-Francois.Mertens> 1) apparently the test file needs a \usepackage[T1]{fontenc}
    Jean-Francois.Mertens> _ a.o. for Apollinaire's poem _ (the test probably dates
    Jean-Francois.Mertens> from a period where people still believed in the Latex Team's
    Jean-Francois.Mertens> promise to make T1 the default as soon as the EC fonts would
    Jean-Francois.Mertens> be out ) , and

  I perhaps understand the problem you notice, but I think that you make
a confusion between input and output encodings. The test file is not at all
linked to T1 encoding (in fact the default mode on my installation is OT1,
that I use nearly all the time).

  Nevertheless, the real point you notice is that the input encoding mode must 
be changed, as there are 8 bits characters in the file. So, the missing
command is: \usepackage[latin1]{inputenc}

  In fact I forget it because it is the default on my system (which is based 
on MlTeX - multi-lingual TeX), but of course it must be there for portability...

    Jean-Francois.Mertens> 2) curiously in this poem the term "eglise" comes out quite
    Jean-Francois.Mertens> differently (and I would say worse) than in the LGC, both when viewed
    Jean-Francois.Mertens> under ghostview, when viewed under Acrobat, and when printed.

  I don't really see what is the problem here, outside the fact that accents
are positioned differently in cmr and ec fonts. I remember that there was
discussions about this point on the french TeX lists, as some people doesn't
like at all the way ec fonts position french accents, but it is a matter of
taste and in any case outside of our subject here.

    Jean-Francois.Mertens> Also the terminal T of "maintenant" in the sc line drops much lower
    Jean-Francois.Mertens> on the page. It is hard to believe
    Jean-Francois.Mertens> the coding has changed _ might anything have changed in PStricks
    Jean-Francois.Mertens> causing this? Can this be reproduced elsewhere, or is there
    Jean-Francois.Mertens> something wrong with my installation ?

  Two points here:

    1/ It is true that when I prepared the test-pst.tex file, I made silently
various changes in the LGC files themselves (something like 15 or 20), as they
are available in CTAN/info/lgc. It was to solve different problems, mainly
the fact that some of the files could only run by themselves as a complete job 
(with a \documentclass, etc.) and was not planned to be locally inserted in a
document. The other cases were to correct the size of the drawings (pspicture
dimensions) which were not accurate (but the pspicture environment is
generally not shown in the LGC examples). Nevertheless, if you compare
doc/lgc/4-5-13.inl with the code in LGC, you will see that I changed the first 
value from -0.1 to -0.2. Why?! Just because if you make the test with -0.1,
you will see that the last T will be completely outside the complete drawing,
as it place will be after the end of the path defined in \pstextpath. But this
did not occured for LGC, just because another font than cmr was used, with
tiny differences in the width of the letters.

    2/ In your case, you see an analog problem because you change the encoding
to T1, which has also tiny differences in the width of the letters, which
explain that the last letter doesn't fit on the curve. The values must be
extended also for this case (in fact 2mm more seems required).


  In any case, I correct the two files to add the `inputenc' package loading
and to avoid the problem you mention in T1 encoding:

pstricks/doc:
------------

  * 98/07/10 test-pst.tex: This file require \usepackage[latin1]{inputenc}
definition as one example (calligramme of 4-5-13.inl) use 8 bits characters.
(reported by Jean-Francois Mertens <jfm at core.ucl.ac.be>)

  * 98/07/10 lgc/4-5-13.inl: Change the values for one verse (adding 1mm on
both left and rigth sides) to allow correct typesetting in T1 encoding
as well as OT1.
(reported by Jean-Francois Mertens <jfm at core.ucl.ac.be>)

  Thanks for the report!

D.G.

-----------------------------------------------------------------------------
The list interface (subscription, information, access to the archives) is on:
http://www.tug.org/cgi-bin/lwgate/pstricks
Otherway to unsubscribe, send mail to pstricks-request at mail.tug.org
with a blank subject and in body the line unsubscribe <email-address>
-----------------------------------------------------------------------------



More information about the PSTricks mailing list