Re: Latin 2 support

[Valeriy E. Ushakov]

On Mon, Nov 27, 2000 at 08:21:33 +0100, dr.Eduard Werner (Edward Wornar) wrote:

> So what's the point in @SysInclude{ latin2 } then if it does *not*
> make latin-2 characters available? I think I'm missing something ...

The point is that it does. ;-)

It gets Latin2 fonts defined with those additional composite glyphs
defined in "plus" files.


> But coming back to the accent problem. Say I want an s with a ^
> accent.  The obvious address@hidden circumflex}s puts the s beneath the
> circumflex and address@hidden circumflex}|{-10p}s puts it even farther
> away. What should I do? When talking about accented characters
> occurring frequently, the clean solution would be a soft font, but
> changing the font for an occasional character is not a very economic
> solution.

Is it for some specific language that uses some specific encoding?  S
with cirumflex is in 8859-3, so why not define a 8859-3 mapping file
and properly declare the language used and then use extra composites
files to get both hyphenation and correct ghlyphs.


On Mon, Nov 27, 2000 at 11:50:04 +0100, Ian Carr-de Avelon wrote:

> > Esperanto but I don't know where to start.   

> I also thought of this one. I assume it can be done with glif files
> like the ones for Latin-2, but I'm not clear whether it would also
> need changes in the C code.

No, C code doesn't need to be changed.

You need an LCM file for the encoding (8859-3, I assume), TeX
hyphenation patterns and rules for extra composites ("plus" font
files).

> All I have worked out so far is in a line like
> 
>     C -1 ; WX 389 ; N sacute ; B 51 -10 348 678 ;
> 
> 389 can be found from s in the other font files, as can 51 -10 348.
> 678 is common to all the acutes.

Well, basicly - yes, accented chars are very stereotypic.  The width
of the composite (WX) is the same as the width of the base glyph,
accent positioning is usually the same across all base glyphs and
the upper-right corener of the bounding box is usually determined by
that and is also common for all accented glyphs.

You can use an existing program like accfont or a2ac to make an
accented font and peek at the metrics of accented chars and write
"plus" font file by looking at them.

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