ntg-context - mailing list for ConTeXt users
 help / color / mirror / Atom feed
* Greek font
@ 2003-09-06  8:06 Thomas A.Schmitz
  2003-09-06 15:49 ` Willi Egger
  0 siblings, 1 reply; 14+ messages in thread
From: Thomas A.Schmitz @ 2003-09-06  8:06 UTC (permalink / raw)


Please excuse this long post. I was/am fiddling with fonts, "probably 
the most complicated aspect of TeX," to quote Hans... I feel I'm almost 
there, but I still need some help and would be very grateful if anybody 
could give me a hint.
Thanks to Giuseppe Bilotta's help, I managed to typeset classical Greek 
with the CB-Greek fonts. Everything works, the accents are in the right 
places etc., yet the original CB-Greek doesn't look very good in pdfs. 
I have a very beautiful Greek font that comes with the LaTeX 
"psgreek"-package; it works in LaTeX and is, according to the manual, 
encoded with the same map as CB-Greek (in LaTeX, the encoding is called 
lgrenc.def, but I couldn't find any useful imformation in that file). I 
wrote a typescriptfile to use it in ConTeXt. The font works, all the 
glyphs are there, as I can see with a \showfont, but they're not in the 
right slots, which means that it doesn't build the correct ligatures 
for displaying accents etc. I assume I need to write a new encoding 
file to reassign the glyphs to their new position. I'd be willing to do 
this by hand (there aren't too many of them, after all), but how do I 
go about that?
What I have got about this font (it's called "GreekOxonia," and in my 
eyes, it's the most beautiful typeface that's out there for classical 
Greek):
greeoxon.pfb
greeoxon.sfd
greeoxon.sfd~
greeoxon.afm
greeoxon.tfm
a map file psgreek.map
and a couple of accompanying .vf, .ovf and .ofm-files (all of them 
containing just a few characters of gibberish when I try to open them).
Can anybody give me a hint where I could start? Any help would be most 
appreciated!
Best
Thomas

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Greek font
  2003-09-06  8:06 Greek font Thomas A.Schmitz
@ 2003-09-06 15:49 ` Willi Egger
  2003-09-06 16:13   ` Thomas A.Schmitz
  2003-09-06 18:03   ` Henning Hraban Ramm
  0 siblings, 2 replies; 14+ messages in thread
From: Willi Egger @ 2003-09-06 15:49 UTC (permalink / raw)


Hi Thomas,

May be that I am not the right person to reply. Never the less I have
installed quite some fonts in ConTeXt.

As you show in your message, you have the pfb and afm files.
With afm2tfm you can generate a tfm file with the required encoding (Could
one use e.g. texnansi?)
By means of texfont you can generate the fontfiles needed for ConTeXt in the
required encoding including a mapfile which you can include in the
pdftex.cfg.

Willi
----- Original Message ----- 
From: "Thomas A.Schmitz" <thomas.schmitz@uni-bonn.de>
To: <ntg-context@ntg.nl>
Sent: Saturday, September 06, 2003 10:06 AM
Subject: [NTG-context] Greek font


> Please excuse this long post. I was/am fiddling with fonts, "probably
> the most complicated aspect of TeX," to quote Hans... I feel I'm almost
> there, but I still need some help and would be very grateful if anybody
> could give me a hint.
> Thanks to Giuseppe Bilotta's help, I managed to typeset classical Greek
> with the CB-Greek fonts. Everything works, the accents are in the right
> places etc., yet the original CB-Greek doesn't look very good in pdfs.
> I have a very beautiful Greek font that comes with the LaTeX
> "psgreek"-package; it works in LaTeX and is, according to the manual,
> encoded with the same map as CB-Greek (in LaTeX, the encoding is called
> lgrenc.def, but I couldn't find any useful imformation in that file). I
> wrote a typescriptfile to use it in ConTeXt. The font works, all the
> glyphs are there, as I can see with a \showfont, but they're not in the
> right slots, which means that it doesn't build the correct ligatures
> for displaying accents etc. I assume I need to write a new encoding
> file to reassign the glyphs to their new position. I'd be willing to do
> this by hand (there aren't too many of them, after all), but how do I
> go about that?
> What I have got about this font (it's called "GreekOxonia," and in my
> eyes, it's the most beautiful typeface that's out there for classical
> Greek):
> greeoxon.pfb
> greeoxon.sfd
> greeoxon.sfd~
> greeoxon.afm
> greeoxon.tfm
> a map file psgreek.map
> and a couple of accompanying .vf, .ovf and .ofm-files (all of them
> containing just a few characters of gibberish when I try to open them).
> Can anybody give me a hint where I could start? Any help would be most
> appreciated!
> Best
> Thomas
>
> _______________________________________________
> ntg-context mailing list
> ntg-context@ntg.nl
> http://www.ntg.nl/mailman/listinfo/ntg-context
>

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Greek font
  2003-09-06 15:49 ` Willi Egger
@ 2003-09-06 16:13   ` Thomas A.Schmitz
  2003-09-08 12:13     ` Jens-Uwe Morawski
  2003-09-06 18:03   ` Henning Hraban Ramm
  1 sibling, 1 reply; 14+ messages in thread
From: Thomas A.Schmitz @ 2003-09-06 16:13 UTC (permalink / raw)


Willi,

thanks for your reply! The problem is: I do have proper tfm-files, and 
the font is installed properly (since it works). But given that 
classical (polytonic) Greek demands some unusual features, I think I 
will have to adapt the font by hand. On the /showfont map, for example, 
I can see that a certain "ligature" is in, say, slot 165. Combinations 
of accents [+ breathing] + vowel are produced by the ligature 
mechanisms (let's say that's the character for "alpha with acute accent 
and smooth breathing"). What I would need to do is:
1) tell TeX that the combination ">'a" should be considered a ligature
2) tell my driver that this ligature can be found in slot 165.
Feasible? Or am I shooting for the impossible? Well, I got one step 
further and converted the tfm and vf-files into vp-files that I can 
read--and maybe edit, if only I knew how...
The thing is, the font does work in LaTeX, so it shouldn't be 
impossible to make it work in ConTeXt, right?
Best, Thomas

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Greek font
  2003-09-06 15:49 ` Willi Egger
  2003-09-06 16:13   ` Thomas A.Schmitz
@ 2003-09-06 18:03   ` Henning Hraban Ramm
  2003-09-06 21:10     ` Thomas A.Schmitz
                       ` (2 more replies)
  1 sibling, 3 replies; 14+ messages in thread
From: Henning Hraban Ramm @ 2003-09-06 18:03 UTC (permalink / raw)



Am Samstag, 06.09.03, um 17:49 Uhr (Europe/Zurich) schrieb Willi Egger:
> With afm2tfm you can generate a tfm file with the required encoding 
> (Could
> one use e.g. texnansi?)

No, texnansi encoding is a western european roman encoding.
But I guess one could use a real greek encoding with texfont,
perhaps lgr-tex (contains only numbers, I don't know what they mean)
or one of the numerous kerkis encodings? (gpkerkis looks good)
(It looks you could actually use Kerkis fonts; it's a greek
Bookman.)
But don't ask me more... ;-)

Grüßlis vom Hraban!
-- 
http://www.fiee.net/texnique/

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Greek font
  2003-09-06 18:03   ` Henning Hraban Ramm
@ 2003-09-06 21:10     ` Thomas A.Schmitz
  2003-09-06 21:35     ` Thomas A.Schmitz
  2003-09-07 22:07     ` Thomas A.Schmitz
  2 siblings, 0 replies; 14+ messages in thread
From: Thomas A.Schmitz @ 2003-09-06 21:10 UTC (permalink / raw)


Yes, I actually tried Kerkis, it may be appropriate for modern Greek, 
but I find it ugly for classical texts. When I do this
texfont --encoding=gpkerkis  --vendor=me --collection=Oxonia --makepath 
--install
and try the font, I don't get any Greek glyphs - strange.

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Greek font
  2003-09-06 18:03   ` Henning Hraban Ramm
  2003-09-06 21:10     ` Thomas A.Schmitz
@ 2003-09-06 21:35     ` Thomas A.Schmitz
  2003-09-07 22:07     ` Thomas A.Schmitz
  2 siblings, 0 replies; 14+ messages in thread
From: Thomas A.Schmitz @ 2003-09-06 21:35 UTC (permalink / raw)


OK, another step (in the right direction?). lgr-tex is an encoding for 
monotoniko Greek (i.e. has only the acute accent, the only one used in 
moderne dimotiki Greek), so not suitable. However, it refers to a file 
of the CB-Greek family "cblig.mf," which contains just the information 
I was looking for - defines all the ligatures, like so:
"ligtable ">":                                               % smooth 
breathing
                 "a" =: oct"202", "e" =: oct"342", "h" =: oct"232", "r" 
=: oct"374"
I can actually understand what this file does: if TeX finds the 
combination >a, it will produce a ligature mapped to slot o202. Does 
this help me in any way? Is there a way I could use a similar table in 
ConTeXt?

--
Thomas A. Schmitz
Philologisches Seminar
Universitaet Bonn
Am Hof 1 e
53113 Bonn
Tel.: 0228/73-7747
Fax: 0228/73-7748

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Greek font
  2003-09-06 18:03   ` Henning Hraban Ramm
  2003-09-06 21:10     ` Thomas A.Schmitz
  2003-09-06 21:35     ` Thomas A.Schmitz
@ 2003-09-07 22:07     ` Thomas A.Schmitz
  2003-09-08 13:52       ` Willi Egger
  2 siblings, 1 reply; 14+ messages in thread
From: Thomas A.Schmitz @ 2003-09-07 22:07 UTC (permalink / raw)


Well, it's me again. I spent the entire weekend trying to set this up, 
but so far, I haven't had any real success. I tried a couple of things 
like different encodings with pfaedit, I tried texfont, fontinst, 
t1utils... I thought that defining the ligatures in a .vf-file would be 
the best approach. Well, it did work, but only kind of: I could create 
a .vpl-file via afm2tfm and edit it, introducing the ligatures I 
needed. These ligatures would then work as expected. The only problem 
was that when I recoded this .vpl to a .vf and a .tfm and installed the 
.tfm, it would miss a couple of characters--they were just gone (like 
uppercase D, G and P), their slots empty. When I actually installed the 
.vf-file in the "texmf/fonts/vf/"-folder, I would get no glyphs at all, 
sometimes a bunch of warnings about duplicate definitions.
Please bear with me if I'm getting on your nerves--this my third or 
fourth attempt to use ConTeXt. I just love it, it's wonderful and makes 
so many things so easy. But every time, using Greek is the stumbling 
block for me. unfortunately, writing in classical Greek is part of my 
job, so if I can't make it work in acceptable quality, I will have to 
go back to LaTeX. So pretty please--can anybody help me? Or even point 
me to people with professional knowledge who might be able to help for 
a fee?

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Greek font
  2003-09-06 16:13   ` Thomas A.Schmitz
@ 2003-09-08 12:13     ` Jens-Uwe Morawski
  0 siblings, 0 replies; 14+ messages in thread
From: Jens-Uwe Morawski @ 2003-09-08 12:13 UTC (permalink / raw)


On Sat, 6 Sep 2003 18:13:21 +0200
Thomas A.Schmitz <thomas.schmitz@uni-bonn.de> wrote:

> What I would need to do is:
> 1) tell TeX that the combination ">'a" should be considered a ligature

in LaTeX this is implemented using some tricks (language support)
from the 'babel' package. Thus, this has _nothing_  to do with fonts
and their encoding. It is neither a remapping using virtual fonts
nor a reencoding of fonts. It _is_ transliteration. Therefore, it has to
be implemented in ConTeXt.

> 2) tell my driver that this ligature can be found in slot 165.

it's already done in the tfm and vf files

Jens

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Greek font
  2003-09-07 22:07     ` Thomas A.Schmitz
@ 2003-09-08 13:52       ` Willi Egger
  2003-09-08 14:40         ` Thomas A.Schmitz
  2003-09-10 23:01         ` Hans Hagen
  0 siblings, 2 replies; 14+ messages in thread
From: Willi Egger @ 2003-09-08 13:52 UTC (permalink / raw)


Hi Thomas,

This matter is much more complicated than I thought in first instance. - I
must admit, that am of no further help. I do hope, dat Hans is following
this tread. He is the guru to get this matter moving.

Cheers Willi

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Greek font
  2003-09-08 13:52       ` Willi Egger
@ 2003-09-08 14:40         ` Thomas A.Schmitz
  2003-09-10 13:08           ` Re[2]: " Giuseppe Bilotta
  2003-09-10 15:51           ` Siep Kroonenberg
  2003-09-10 23:01         ` Hans Hagen
  1 sibling, 2 replies; 14+ messages in thread
From: Thomas A.Schmitz @ 2003-09-08 14:40 UTC (permalink / raw)


Willi,
thanks for being supportive. In the meantime, I have taken another 
step: by using tftopl, I produced a .pl-file of my font and edited it. 
I inserted a "LIGTABLE" with ligature (and kerning) information. After 
using pltotf and using the new .tfm, I had partial success: I get a lot 
of the ligatures I want! However, a few problems remain:
1) The "space"-glyph has vanished. I assume it must somehow have been 
mismapped. Does anybody have an idea how I could remap this?
2) One Greek accent is called by using the "~"-character. However, I 
have no idea to which slot it is mapped; the same is true for the 
"|"-character. How can I find out? And, by what Giuseppe wrote me a 
while ago, I gather that I have to make at least the "|" inactive. Is 
this done with \catcode`|=\other ?
I'm hopeful that these problems can be solved. And I am looking forward 
to that. I feel a lot like the monkey in a story by Wilhelm Busch: it 
has a coconut, it knows there must be something delicious inside, but 
it doesn't know how to reach it. that's the way i feel about context!
Best
Thomas

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re[2]: Greek font
  2003-09-08 14:40         ` Thomas A.Schmitz
@ 2003-09-10 13:08           ` Giuseppe Bilotta
  2003-09-10 15:51           ` Siep Kroonenberg
  1 sibling, 0 replies; 14+ messages in thread
From: Giuseppe Bilotta @ 2003-09-10 13:08 UTC (permalink / raw)


Monday, September 8, 2003 Thomas A.Schmitz wrote:

> Willi,
> thanks for being supportive. In the meantime, I have taken another
> step: by using tftopl, I produced a .pl-file of my font and edited it.
> I inserted a "LIGTABLE" with ligature (and kerning) information. After
> using pltotf and using the new .tfm, I had partial success: I get a lot
> of the ligatures I want! However, a few problems remain:
> 1) The "space"-glyph has vanished. I assume it must somehow have been
> mismapped. Does anybody have an idea how I could remap this?

Can't help you here.

> 2) One Greek accent is called by using the "~"-character. However, I
> have no idea to which slot it is mapped; the same is true for the
> "|"-character. How can I find out? And, by what Giuseppe wrote me a
> while ago, I gather that I have to make at least the "|" inactive. Is
> this done with \catcode`|=\other ?

Yes, and you need to do the same for ~; or, you can use the =
instead of the ~ for the circumflex accent.

-- 
Giuseppe "Oblomov" Bilotta

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Greek font
  2003-09-08 14:40         ` Thomas A.Schmitz
  2003-09-10 13:08           ` Re[2]: " Giuseppe Bilotta
@ 2003-09-10 15:51           ` Siep Kroonenberg
  2003-09-10 16:26             ` Thomas A.Schmitz
  1 sibling, 1 reply; 14+ messages in thread
From: Siep Kroonenberg @ 2003-09-10 15:51 UTC (permalink / raw)


On Mon, Sep 08, 2003 at 04:40:31PM +0200, Thomas A. Schmitz wrote:
> Willi,
> thanks for being supportive. In the meantime, I have taken another 
> step: by using tftopl, I produced a .pl-file of my font and edited it. 
> I inserted a "LIGTABLE" with ligature (and kerning) information. After 
> using pltotf and using the new .tfm, I had partial success: I get a lot 
> of the ligatures I want! However, a few problems remain:
> 1) The "space"-glyph has vanished. I assume it must somehow have been 
> mismapped. Does anybody have an idea how I could remap this?

TeX doesn't use a space glyph. Instead, there is a fontdimen
space and there may be boundarychar kerns and ligatures to replace
space kerns and -ligs.

If you want more complete documentation: get a TeX source tree, run
weave on pltotf.web or vptovf.web and compile the resulting TeX
file.

-- 
Siep Kroonenberg
siep@elvenkind.com

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Greek font
  2003-09-10 15:51           ` Siep Kroonenberg
@ 2003-09-10 16:26             ` Thomas A.Schmitz
  0 siblings, 0 replies; 14+ messages in thread
From: Thomas A.Schmitz @ 2003-09-10 16:26 UTC (permalink / raw)


Thanks again for helping. I'm happy to say that I'm almost home free. 
The trick was to go via .vpl-files and edit them. The missing glyphs 
were somewhere in the table, I could simply copy and paste them in the 
right position; I then added ligatures for all the accents/breathings 
etc and finally kerning information. I then used vptovf to produce a 
matching tfm, and everything worked as expected; even the "space" has 
micraculously reappeared. Strangely, when I put the resulting .vf-file 
into my "fonts/vf"-folder, TeX would go haywire; this means that I 
couldn't use any remapping in the fonts. I had to cut a few corners, it 
is just a "quick-and-dirty" fix, but it works!!! I've learnt more about 
fonts than I ever wanted, but I now have (almost) real control; I even 
managed to replace one character I didn't like with its counterpart 
from another font (and then edit this new character to perfection in 
pfaedit, a really amazing tool). One last (little) problem; maybe one 
of you has an idea: since I couldn't remap, I have trouble producing a 
single breathing in front of a capital letter. I helped myself by 
defining \define\<{\getglyph{greeoxon}{96}}, so I could at least use 
\<A to obtain capital alpha with rough breathing. The problem is that 
after these commands, I can't get any kerning to work. Any ideas?
Maybe one day I'll know enough about TeX to do all of this the proper 
way. Maybe one day I'll be able to use Omega (btw, how can the new 
e-Omega/Aleph be obtained? And where can I dload Omega 1.15?). But for 
the time being, I'm a happy camper again. Should any of you be 
interested, I'd be glad to share my stuff.
Best
Thomas

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: Greek font
  2003-09-08 13:52       ` Willi Egger
  2003-09-08 14:40         ` Thomas A.Schmitz
@ 2003-09-10 23:01         ` Hans Hagen
  1 sibling, 0 replies; 14+ messages in thread
From: Hans Hagen @ 2003-09-10 23:01 UTC (permalink / raw)


At 15:52 08/09/2003 +0200, you wrote:
>Hi Thomas,
>
>This matter is much more complicated than I thought in first instance. - I
>must admit, that am of no further help. I do hope, dat Hans is following
>this tread. He is the guru to get this matter moving.

sure, but this week i'm in dante+gutenberg meeting mode as well as project 
deadline mode

Hans
-------------------------------------------------------------------------
                                   Hans Hagen | PRAGMA ADE | pragma@wxs.nl
                       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
-------------------------------------------------------------------------
                        information: http://www.pragma-ade.com/roadmap.pdf
                     documentation: http://www.pragma-ade.com/showcase.pdf
-------------------------------------------------------------------------

^ permalink raw reply	[flat|nested] 14+ messages in thread

end of thread, other threads:[~2003-09-10 23:01 UTC | newest]

Thread overview: 14+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2003-09-06  8:06 Greek font Thomas A.Schmitz
2003-09-06 15:49 ` Willi Egger
2003-09-06 16:13   ` Thomas A.Schmitz
2003-09-08 12:13     ` Jens-Uwe Morawski
2003-09-06 18:03   ` Henning Hraban Ramm
2003-09-06 21:10     ` Thomas A.Schmitz
2003-09-06 21:35     ` Thomas A.Schmitz
2003-09-07 22:07     ` Thomas A.Schmitz
2003-09-08 13:52       ` Willi Egger
2003-09-08 14:40         ` Thomas A.Schmitz
2003-09-10 13:08           ` Re[2]: " Giuseppe Bilotta
2003-09-10 15:51           ` Siep Kroonenberg
2003-09-10 16:26             ` Thomas A.Schmitz
2003-09-10 23:01         ` Hans Hagen

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).