[NTG-context] ec encoding and tcaron
mojca.miklavec.lists at gmail.com
Sat Jun 10 00:08:51 CEST 2006
On 6/9/06, Michal Kvasnicka wrote:
> Hi Richard.
> > I'm sure the EC encoding contains the 'tcaron' character (see the
> > lm-ec.enc file for example).
> > I have ConTeXt on top of TeXLive 2005.
> > I can find:
> > texmf/fonts/enc/dvips/base/ec.enc
> > texmf/fonts/enc/dvips/lm/ec-lm.enc
> > texmf/fonts/enc/dvips/lm/lm-ec.enc
> > I use EC normally for typesetting Czech documents. So I would suggest
> > to use EC ;-)
> I'm sorry you're not right. The lm-ec.enc really includes tcaron, but
> neither ec.enc nor EC.enc does, at least at my teTeX 3.0. If you use
> only LatinModern, it works, because lm-ec.enc is used. But I doubt it
> works well for other fonts. Does it? I was unsuccessful. Can you send me
> your ec.enc file please?
Approximately a year ago there was a discussion on tex-fonts mailing
list about fixing ec.enc (for example there's a glyph called "dbar",
which is agains any standards: in Unicode it's called "d with stroke"
and according to Adobe standards it should be called "dcroat"; and
many more inconsistencies).
The result of the discussion was something like: "No, we may not
change this since it may break functionality of some fonts which used
that standard years ago. Googling for this and that reveales that some
fonts still use those old glyph names ..." (perhaps Google found 4
hits or so ...) The explanation/excuse was that everyone using modern
glyph names should create his own "ec.enc" (just as it was done for
I would recommend you to use lm-ec.enc or tex256.ec ("fixed" version
of ec.enc which should be present in the same folder as ec.enc). The
Polish guys did their best to follow the standards, so the ec file of
Latin Modern should be OK for most fonts. If some glyph is still
missing, you can manually change the encoding definition (and then
note that specific encoding in map file as well, in the same was as
it's done for LM).
You can test the resulting font with
\showfont[yourfontname] % where yourfountname should preferrably
In that way you'll spot any missing glyphs pretty easily.
About xl2.enc: it is incomplete and neiter supported by ConTeXt (it
could be, but there's no real benefit) nor by most popular fonts (you
have to create metrics and everything by yourself, so your files might
be highly unportable). I dropped the idea about using it pretty soon.
It's not impossible to use it, but probably not worth it.
More information about the ntg-context