Too much betas with languages (long)

Too much betas with languages (long)

[Balazs Nagy]

Hi,

I have two versions of teTeX.  At work I have a 1.0.6 and at home there is a
1.0.7.  Each has Hungarian patch (by me), which contains an

- enco-pdf.tex addition:
	\defineaccent  H U ^^db
	\defineaccent  H u ^^fb
	\defineaccent  H O ^^d5
	\defineaccent  H o ^^f5

- an enco-whu.tex based on enco-win.tex with the changes of

	\~o -> \H o
	\~O -> \H O
	\^u -> \H u
	\^U -> \H U

- Hungarian additions to lang-ura.tex and

- mult-sys.tex addition:
	\defineconstant {hungarian}   \defineconstant {hu}

- cont-usr.tex addition:
	\definesysnonym [lang-hu.pat]  [huhyph.tex]

Then, I generated cont-en.efmt.

With the old engine I can get good results, but with the new one I get weird
effects (maybe I neither do background colors but I didn't try it with the
new install): just before a macro places some text, it prints a beta symbol. 
Ie. the page numbers look like <beta><beta><number>, other fix running
labels are in the <beta><label> format, the cover has a
\midaligned{\tfa\currentdate} which has the same beta sign and so on.  The 
tex file generator starts with

        print "% interface=english paper=a4\n";
        print "\\doublesidedprinttrue\n";
        print "\\setupoutput[pdftex]\n";
        print "\\mainlanguage[hu]\n";
        print "\\useencoding[pdfdoc,whu]\n";
        print "\\input ../setup\n";
        print "\\def\\modulename{$title}\n";
        print "\\def\\moduletype{\\typenovel}\n";
        print "\\starttext\n";
        print "\\startstandardmakeup\n";
        print "\\headline\n";
        print "\\blank[6*big]\n";
        print "\\midaligned{\\bfd{$title}}\n";
        print "\\blank[3*big]\n";
        print "\\midaligned{\\tfd{\\moduletype}}\n";
        print "\\vfill\n";
        print "\\midaligned{\\tfa{$written: $author \\email{$email}}}\n";
        print "\\midaligned{\$\\star\$}\n";
        print "\\blank[small]\n";
        print "\\midaligned{\\tfa\\currentdate}\n";
        print "\\stopstandardmakeup\n";

The setup.tex contains

\def\headline{
  \switchtobodyfont[ss,10pt]
  \midaligned{\tfc i~g~e~n~~k~ö~n~n~y~ű~~s~z~e~r~e~p~j~á~t~é~k}
}

\define[1]\email{\useURL[#1][mailto:#1][][\ss $<$#1$>$]\from[#1]}

Which can be a problem?
-- 
Nagy Balazs, LSC
http://www.lsc.hu/

Re: Too much betas with languages (long)

[Hans Hagen]

At 01:03 PM 10/24/00 +0200, Balazs Nagy wrote:
>Hi,

>- enco-pdf.tex addition:
>	\defineaccent  H U ^^db
>	\defineaccent  H u ^^fb
>	\defineaccent  H O ^^d5
>	\defineaccent  H o ^^f5

If this does not conflict with any other language I can add it permanently. 

>- an enco-whu.tex based on enco-win.tex with the changes of
>
>	\~o -> \H o
>	\~O -> \H O
>	\^u -> \H u
>	\^U -> \H U

Same for this. 

>- Hungarian additions to lang-ura.tex and
>
>- mult-sys.tex addition:
>	\defineconstant {hungarian}   \defineconstant {hu}

i've added this 

>- cont-usr.tex addition:
>	\definesysnonym [lang-hu.pat]  [huhyph.tex]

and this

>Then, I generated cont-en.efmt.

are all the hungarian labels okay? 

>With the old engine I can get good results, but with the new one I get weird
>effects (maybe I neither do background colors but I didn't try it with the

to what extend is it related to backgrounds? 

>new install): just before a macro places some text, it prints a beta symbol. 
>Ie. the page numbers look like <beta><beta><number>, other fix running
>labels are in the <beta><label> format, the cover has a
>\midaligned{\tfa\currentdate} which has the same beta sign and so on.  The 
>tex file generator starts with
>
>        print "% interface=english paper=a4\n";
>        print "\\doublesidedprinttrue\n";

What is this? Looks pretty low level to me: \doublesidedprinttrue is not a
user command

>        print "\\setupoutput[pdftex]\n";
>        print "\\mainlanguage[hu]\n";
>        print "\\useencoding[pdfdoc,whu]\n";

Pdfdoc encoding is never to be activated since it is only needed for low
level pdf where it happens automatically! So, no [pdftex] 

>        print "\\input ../setup\n";

you may try \environment setup 

>        print "\\def\\modulename{$title}\n";
>        print "\\def\\moduletype{\\typenovel}\n";

no lowercase names since it may clash! 

>        print "\\starttext\n";
>        print "\\startstandardmakeup\n";
>        print "\\headline\n";

same for \headline

>        print "\\blank[6*big]\n";
>        print "\\midaligned{\\bfd{$title}}\n";
>        print "\\blank[3*big]\n";
>        print "\\midaligned{\\tfd{\\moduletype}}\n";
>        print "\\vfill\n";
>        print "\\midaligned{\\tfa{$written: $author \\email{$email}}}\n";
>        print "\\midaligned{\$\\star\$}\n";
>        print "\\blank[small]\n";
>        print "\\midaligned{\\tfa\\currentdate}\n";
>        print "\\stopstandardmakeup\n";
>
>The setup.tex contains
>
>\def\headline{
>  \switchtobodyfont[ss,10pt]
>  \midaligned{\tfc i~g~e~n~~k~ö~n~n~y~û~~s~z~e~r~e~p~j~á~t~é~k}
>}

\def\stretched% slightly adapted to handle spaces
  {\ifvmode\hbox to \hsize\else\ifinner\else\hbox\fi\fi
   \processtokens\relax\hss\relax{\hss\hss}}

\hbox to 5cm{\stretched{Hi There}}

there is also: 

\stretcheduppercase{Hi There}

and in the next release: \stretchedlowercase and \stretchednormalcase

>\define[1]\email{\useURL[#1][mailto:#1][][\ss $<$#1$>$]\from[#1]}
>
>Which can be a problem?

I don't know. Your file runs ok here. In order to check it I need a minimal
example + log + pdf file. 

Hans
-------------------------------------------------------------------------
                                                  Hans Hagen | PRAGMA ADE
                      Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
 tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com
-------------------------------------------------------------------------

Re: Too much betas with languages (long)

[Balazs Nagy]

On Tue, 24 Oct 2000, Hans Hagen wrote:

> At 01:03 PM 10/24/00 +0200, Balazs Nagy wrote:

> >- enco-pdf.tex addition:
> >	\defineaccent  H U ^^db
> >	\defineaccent  H u ^^fb
> >	\defineaccent  H O ^^d5
> >	\defineaccent  H o ^^f5
> 
> If this does not conflict with any other language I can add it permanently. 

It is OK because only Hungarians use hungarumlaut (or double accent).

> >- an enco-whu.tex based on enco-win.tex with the changes of
> >
> >	\~o -> \H o
> >	\~O -> \H O
> >	\^u -> \H u
> >	\^U -> \H U
> 
> Same for this. 

This is tricky because Latin-1 and Latin-2 differs (in a Hungarian's point
of view) in these changes, thus until there isn't a real Latin-2 support, an
enco-hu.tex or enco-whu.tex can be a good choice.

[...]
> i've added this 
[...]
> and this

Thanks.

> are all the hungarian labels okay? 

AFAIK yes (I'll send you my latest lang-ura.tex file).

> >        print "\\useencoding[pdfdoc,whu]\n";
> 
> Pdfdoc encoding is never to be activated since it is only needed for low
> level pdf where it happens automatically! So, no [pdftex] 

You meant [pdfdoc].  Yes, it was the problem.  Sadly the old engine simply
swallows it and doesn't do any errors.

> >        print "\\input ../setup\n";
> you may try \environment setup 

Nifty!  Where can I find a short list of these useful commands?

> no lowercase names since it may clash!

Reasonable.  Thanks.
> \def\stretched% slightly adapted to handle spaces
>   {\ifvmode\hbox to \hsize\else\ifinner\else\hbox\fi\fi
>    \processtokens\relax\hss\relax{\hss\hss}}

My version is a bit older.  It contained \normalspace or so.

> and in the next release: \stretchedlowercase and \stretchednormalcase

Yeah,  go on!

Thank you for the fast answer.  I am a satisfied context user now ;-)
-- 
Nagy Balazs, LSC
http://www.lsc.hu/

Re: Too much betas with languages (long)

[Hans Hagen]

At 08:49 PM 10/24/00 +0200, Balazs Nagy wrote:
>On Tue, 24 Oct 2000, Hans Hagen wrote:
>
>> At 01:03 PM 10/24/00 +0200, Balazs Nagy wrote:
>
>> >- enco-pdf.tex addition:
>> >	\defineaccent  H U ^^db
>> >	\defineaccent  H u ^^fb
>> >	\defineaccent  H O ^^d5
>> >	\defineaccent  H o ^^f5
>> 
>> If this does not conflict with any other language I can add it
permanently. 
>
>It is OK because only Hungarians use hungarumlaut (or double accent).

ok, added. 

>> >- an enco-whu.tex based on enco-win.tex with the changes of
>> >
>> >	\~o -> \H o
>> >	\~O -> \H O
>> >	\^u -> \H u
>> >	\^U -> \H U
>> 
>> Same for this. 

eh, are you sure about enco-win? 

>This is tricky because Latin-1 and Latin-2 differs (in a Hungarian's point
>of view) in these changes, thus until there isn't a real Latin-2 support, an
>enco-hu.tex or enco-whu.tex can be a good choice.

then, why not extent enco-il2? 

>You meant [pdfdoc].  Yes, it was the problem.  Sadly the old engine simply
>swallows it and doesn't do any errors.

that is because it is a legal encoding, but just not meant to be used this way

>> >        print "\\input ../setup\n";
>> you may try \environment setup 
>
>Nifty!  Where can I find a short list of these useful commands?

in the manuals 

\environment somefile 

has the advantage that it also searches in lower paths, which makes sense
in the case of structured file management

>> no lowercase names since it may clash!
>
>Reasonable.  Thanks.
>> \def\stretched% slightly adapted to handle spaces
>>   {\ifvmode\hbox to \hsize\else\ifinner\else\hbox\fi\fi
>>    \processtokens\relax\hss\relax{\hss\hss}}
>
>My version is a bit older.  It contained \normalspace or so.

indeed, but it was not really meant for spaced text in a larger font -) 

Hans
-------------------------------------------------------------------------
                                                  Hans Hagen | PRAGMA ADE
                      Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
 tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com
-------------------------------------------------------------------------

Re: Too much betas with languages (long)

[Balazs Nagy]

On Wed, 25 Oct 2000, Hans Hagen wrote:

> >> >- enco-pdf.tex addition:
> >> >	\defineaccent  H U ^^db
> >> >	\defineaccent  H u ^^fb
> >> >	\defineaccent  H O ^^d5
> >> >	\defineaccent  H o ^^f5

This is missing from enco-il2.tex too.

> eh, are you sure about enco-win? 
> 
> >This is tricky because Latin-1 and Latin-2 differs (in a Hungarian's
> >point of view) in these changes, thus until there isn't a real Latin-2
> >support, an enco-hu.tex or enco-whu.tex can be a good choice.
> 
> then, why not extent enco-il2? 

Because when I write Latin-2's 0xd5 character, I want TeX to convert it to
\H O.
-- 
Regards: Kevin (Balazs)
IKSZ: szabad szerepjatek barmilyen helyszinre. http://www.iksz.hu/

Re: Too much betas with languages (long)

[Hans Hagen]

At 05:48 PM 10/26/00 +0200, Balazs Nagy wrote:
>On Wed, 25 Oct 2000, Hans Hagen wrote:
>
>> >> >- enco-pdf.tex addition:
>> >> >	\defineaccent  H U ^^db
>> >> >	\defineaccent  H u ^^fb
>> >> >	\defineaccent  H O ^^d5
>> >> >	\defineaccent  H o ^^f5
>
>This is missing from enco-il2.tex too.

added. 

Hans
-------------------------------------------------------------------------
                                                  Hans Hagen | PRAGMA ADE
                      Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
 tel: +31 (0)38 477 53 69 | fax: +31 (0)38 477 53 74 | www.pragma-ade.com
-------------------------------------------------------------------------