Re: Formatting DocBookXML with Lout

[Ludovic Courtès]

Hi,

Yves Forkl <address@hidden> writes:

> I'm facing another problem: While my text consists mainly of characters
> from Latin-1, I also need to use 6 characters from Latin-2 (always in
> the same words), e.g. U+010D "LATIN SMALL LETTER C WITH CARON".
>
> I know that this is not an easy matter, but nevertheless, I would like
> to ask naively: How can I trick Lout into putting these Latin-2
> characters into the PS as well? (Change the font, the encoding? Do I
> need to acquire additional fonts which I would need to embed when
> generating PDF?)

That may be doable, but requires hackery.

For instance, if you look at the AFM files for the "Times" font, under
`fonts' in the Lout source tree, you'll find `Ti-Rm' and `Ti-Rm+': the
former contains all Latin-1 characters, while the latter contains
Latin-2 (and perhaps some more) characters.

Then you can look at `data/fontdefs.ld' and `data/latin2.ld': the former
contains Latin-1 font definitions while the latter contains Latin-2
definitions.

Finally, there are "character maps" under `maps', namely `Latin1.LCM'
and `Latin2.LCM', that map 8-bit character codes to glyph names.

So, one possibility for you would be to have a custom character map that
maps some unused 8-bit character (you should even be able to use one
lower than 32) to "ccaron".  You would then create your own font
definition file, based on `fontdefs.ld', with an address@hidden' field
pointing to, say, a trimmed down version of `Ti-Rm+' that only contains
`ccaron'.  This is all untested, but sounds plausible.  :-)

(Alternatively, you can port Lout to FreeType so that we can get out of
this quagmire. ;-))

Hope this helps,
Ludovic.