> Below (and at
> http://snipabacken.dyndns.org/~grahn/tmp/groff-se.patch) is an
> update which, as far as I can tell, brings sv.tmac to the same level
> as fr.tmac.
Thanks! Applied (with minor modifications). Note that I need a
copyright assignment from you in case you want to contribute more
stuff to groff...
> - .hcodes map the accented letters 'e' to plain 'e' -- I hope
> that is proper.
Why that? First, there isn't an `è' in the Swedish hyphenation
patterns, so you don't need to mention it. Second, there are patterns
which use `é', so both `É' and `é' should map to `é'.
> - I don't really understand the escapes for national characters.
> Will \[a ao] really end up as 'å'?
Yes. \[x y ...] defines a composite character. `ao' as a non-first
argument within \[...] is mappable by `.composite' requests, mainly to
map glyph names of spacing accents to non-spacing ones. In the file
composite.tmac you can find
.composite ao u030A
so we have glyph name \[a u030A], which groff internally maps to
u0061_030A, and this is what you find in the font files (see the
groff_char man page).
> I would have expected \[oa].
You can use this too; it is also mapped internally to glyph name
u0061_030A.
> - What about the '.ss 12 0' line here and in fr.tmac? It makes
> sense to use it in all documents IMHO, but should these files
> activate it?
I think yes. Compare this with LaTeX's babel mechanism -- the
decision whether to use \frenchspacing or not is also handled in the
language files. AFAIK, Swedish doesn't have an additional space after
a fullstop. Why do you think that it should not be there?
Werner
_______________________________________________
Groff mailing list
Groff@gnu.org
http://lists.gnu.org/mailman/listinfo/groff
