On Sun, 23 Jan 2005 22:47:57 +0100, h h extern wrote:

Mikael Persson wrote:

...

I am looking for a way of writing some home exercises in my Russian course in ConTeXt. However, so far I have not succeed with it. Should it work out of the box without installing any new fonts (I am having the fonts from a full install of TeXLive2003 now)? I have looked a bit in the source files for context, but with no luck.

well, i just took a look in the latest tex tree, and i cannot find a trace of the russian fonts that are defined in abundance in context's type-enc.tex; one problem with russian is that nobody seems to be responsible to get all the right stuff on tex live;

Based on the definitions, the following should work (and is reported to work):

\starttext

\definetypeface [latin] [rm] [serif] [latin-modern] [default] [encoding=texnansi]

\definetypeface [russian] [rm] [serif] [computer-modern] [default] [encoding=cyr] % t2a

\latin \dostepwiserecurse{`a}{`z}{1}{\char\recurselevel} \endgraf \russian \dostepwiserecurse{`a}{`z}{1}{\char\recurselevel} \endgraf

\stoptext

Hello, and thank you for your answer. However this small test file does not work here. Not in TeXLive2003 or with a fresh install of TeXLive2004. If I run it as it is, I get the bluesky cmr12 and cmr10. If cyr is changed to t2a, then it does not run anymore. It complains about larm1000 not found. I guess this is the cm-super files. In type-enc.tex there is a line \definefontsynonym [cmr10] [larm1000] [encoding=t2a] and I have the cm-super fonts installed, however no file called larm1000.tfm. In cm-super-t2a.map I see: larm1000 SFRM1000 "T2AEncoding ReEncodeFont"

In order to get it working i suggest those using russian to do the following:

- determine what encodings are *really* in use

This, I shall ask one russian guy I know.

- collect the tfm/vf/pfb files that are needed to get that done - build me the right map files

(i probably involves some fonts from cmsuper; so ... what is the reasonable subset that we need; if needed we can stick to a few reasonable sized (as we do with greek); it's mostly the math that is sensitive to design sizes)

we can then make a proper subset for (1) tex live and (2) the minimal context distribution; even better would be to have those glyohs needed to be added to the latin modern fonts [i can look into that later]

That sounds great!

...

I guess there are different ways to achieve this, however, since I also want some swedish text in the same documents, the best would probably be if one of the following options worked:

the simpliest is to switch font and language at the same time (but i cannot test anything if i have no fonts running on my sytstem]

...

1) Typing everything in utf8, and get the different languages automatically when compiling.

currently the utf handler does not change fonts, but since language switches are to be given anyway, it can be hooked into the language handler if needed

...

2) Typing everything with western letters, having some kind of \startrussia Boris \stoprussia and get Boris typeset with Cyrillic letters.

Just language+font switches using the typeface macros.

Hans

OK, thank you for clarifying this. Best Regards, Micke P

On Tue, 25 Jan 2005 16:10:20 +0100, h h extern wrote:

Mikael Persson wrote:

...

larm1000 SFRM1000 "T2AEncoding ReEncodeFont"

and I got sfrm1000.pfb and cm-super-t2a.enc files... So, in order to have the cm-super fonts working, do I have to generate some files or do something more?

you need tfm (larm1000.tfm) files but i dunno where to find them, maybe in some cmsuper zip

...

This, I shall ask one russian guy I know.

good, we can repackage the tfm's if needed; (i'm surprised how often fonts / names / etc change)

Hans

Thank you Hans! I did not find the tfm files anywhere on the Internet (strange), so I generated them myself. Now it works very well. I put a description of how I did, and some comments on what encoding that worked (windows 1251), and what encodings that did not (KOI8R, KOI8-R, ISO-8859-5), and a small test file at http://www.math.chalmers.se/~mickep/russian/ Hope this helps any other users in the future. And please feel free to tell me if there are better solutions. Micke P

Mikael Persson wrote:

http://www.math.chalmers.se/~mickep/russian/

it looks like russian indeed are all those fonts in the cmsuper path useful? we should make one map file with all the relevant mappings that we can add to the context distribution; also, we need a zip with prebuild tfm files; i can then add that to the minimal context file set Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------

Mikael Persson said this at Wed, 26 Jan 2005 22:38:50 +0100:

I also thought more about the utf. Is there anyway, now, to use utf8 encoding of the file, and somehow (from enco-uc.tex or how it would work) get the russian letters (that is for example \uchar4{"11} to be \cyrillicB, as it is written in enco-uc.tex and then from that get the right letter)?

Yeah, there is a way to use UTF-8 input, but it's not ready for Cyrillic quite yet--It'll take some work (from you? :D ) to get it going. UTF-8 input depends on the unic-* files. The Cyrillic vector, unic- 004.tex, hasn't been done yet. All it takes is putting the named glyphs (that you saw in enco-uc, for example) in the right order, using unic- 001.tex as a model. Interested? I can get you started if you are... adam -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Adam T. Lindsay, Computing Dept. atl@comp.lancs.ac.uk Lancaster University, InfoLab21 +44(0)1524/510.514 Lancaster, LA1 4WA, UK Fax:+44(0)1524/510.492 -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

Adam Lindsay wrote:

Yeah, there is a way to use UTF-8 input, but it's not ready for Cyrillic quite yet--It'll take some work (from you? :D ) to get it going.

indeed, it would be godo to have that vector, yesterday i (partially) made the greek one Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------

On Thu, 27 Jan 2005 10:11:51 +0100, Hans Hagen wrote:

Adam Lindsay wrote:

...

Yeah, there is a way to use UTF-8 input, but it's not ready for Cyrillic quite yet--It'll take some work (from you? :D ) to get it going.

indeed, it would be godo to have that vector, yesterday i (partially) made the greek one

Hans

OK, so here they are. Hope I understood correctly how it works... :) enco-004.tex and enco-uc-add.tex I used the names from gucharmap (char map app for Gnome) where they were not yet defined. The names is not always so short as one maybe would like. /Micke P

Mikael Persson wrote:

OK, so here they are. Hope I understood correctly how it works... :) enco-004.tex and enco-uc-add.tex I used the names from gucharmap (char map app for Gnome) where they were not yet defined. The names is not always so short as one maybe would like.

This one probably needs to be updated (Adam: what to do with names, we have Greeka GreekA etc maybe it should be greeka greekA, but maybe that will clash some day; same for cyrillic) Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------

Mikael Persson said this at Thu, 27 Jan 2005 14:50:13 +0100:

...

(Adam: what to do with names, we have Greeka GreekA etc maybe it should be greeka greekA, but maybe that will clash some day; same for cyrillic)

Hans

Ack! You're asking me? I always thought you had a logic to it... I just picked up the (\cyrillicBLAH) pattern from what was already in ConTeXt. But if you were to push me on my opinion, the greek characters should probably have a \greek prefix, rather than \Greek, because of your usual lowercase namespacing conventions. I'd say leave cyrillic as it is? adam -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Adam T. Lindsay, Computing Dept. atl@comp.lancs.ac.uk Lancaster University, InfoLab21 +44(0)1524/510.514 Lancaster, LA1 4WA, UK Fax:+44(0)1524/510.492 -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

I uploaded a new alpha release with the adapted / extended cyrillic/greek encoding vectors (enco-* & unic-*) Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------

Mikael Persson wrote:

\enableregime[utf]

\definetypeface [russian] [rm] [serif] [computer-modern] [default] [encoding=cm-super-t2a] % neither with encoding=t2a nor cyr

there is no cm-super-t2a encoding, just t2a should do

\setupbodyfont[russian]

\starttext Мама и Папа % Some russian characters \stoptext

well, it looks like russian in amy mailer -) but shouldn't you enter utf8 code ? Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------

Hello all, I am reading a beautifully designed book called 'Boodbye Gutenberg", and it gave me inspiration to change the design of my self-published novel. However, I wonder if with my limited knowledge in Contet I can make it happen. Here it is: Version 1: The text is standard, justified both sides. When the reader opens the book I want him/her to see a background light-gray picture centered in the spine of the book from the top of the page to the bottom. (The picture is a guitar with the strings along the spine), I would like half of the guitar in the left page, and half in the other page. In other words, there would be a bleeding picture to the right of the left page, and a bleeding picture to the left of the right page. The text would be black, and the guitar would be seen as background. I know it is difficult to follow my explanation, but I hope you understand. Version 2: The text is wrapped around the figure, that is, each line in the left page is left justified, but ends at the border of the picture. And each line in the right page is right justified, but begins at the border of the picture. I am hoping to get the version 1. The other version is like an exercise for me, to see how it would be done. I have tried with layers for the background picture, and it works, but I don't know how to tell Context to use alternative pictures, one for the left pages, another for the right pages. My apologies for the length ... thank you ciro ===== ================== Ciro A. Soto Author of "The Guitar Maker. An Exploration of Wisdom, Design and Love." Pub. Date: Dec. 2005. "All problems are at the interface. Each one of them has a solution."

Ciro A. Soto wrote:

Sorry, I made a typo error. The title of the book I am reading is "Goodbye Gutenberg" by Valerie Kirschenbaum. A book I recommend for any person interested in book design, typography, history, reading, fonts, etc.

hm, i dunno that one; sounds kind of sad, this 'goodbye' -) i wonder if it makes sense to add a biblography corner to the wiki, where users can enter/describe - good books about typography - books with special (nice) typography so that ww have a nice 'example' bibliographic test file as well (xml version as well as bibtex one) Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------

Mikael Persson wrote:

Hm, I am entering utf8 code. I am not sure how it got converted in the email, however, when I type my .tex file I am sure that it is in utf-8. But you mean that this example should work?

I attatch the testfile and the resulting pdf and logs and so on...

tracing shows that the utf part is okay: \enableregime[utf] % \definetypeface [russian] % [rm] [serif] [computer-modern] [default] [encoding=t2a] % % \setupbodyfont[russian] \starttext \startbuffer Мама и Папа % Some russian characters \stopbuffer \chardef\utfunicodetracer=0 \blank \getbuffer \blank \chardef\utfunicodetracer=2 \blank \getbuffer \blank \chardef\utfunicodetracer=7 \blank \getbuffer \blank \stoptext but i cannot test the font part, can you send me a zip with your t2a tfm files and map file and enc file? Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------

concerning cyrillic ... we should also take care of hebrew, then we have latin. greek, cyrillic and hebrew covered. Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------

Rob Ermers wrote:

Hi,

I would be interested in a solution for Arabic... Presently \ works, but otherwise support for Arabic (\startarabic) is not optimal. If one of you could make a set-up for that too ...

arabic a bit more tricky since it has those ligatures, so that needs aleph or arabtex (of course we can make an encoding with named glyphs) Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------

Mikael Persson wrote:

Hm, I don't think so. But that is a question for everyone to consider. However, if you ask me I would be happy to have (snipped from the README file):

rm: Computer Modern Roman sl: Computer Modern Slanted ti: Computer Modern Italic cc: Computer Modern Caps and Small Caps ui: Computer Modern Unslanted Italic

is this oen really used?

sc: Computer Modern Slanted Caps and Small Caps ci: Computer Modern Classical Serif Italic

is that one used?

bx: Computer Modern Bold Extended bl: Computer Modern Bold Extended Slanted bi: Computer Modern Bold Extended Italic xc: Computer Modern Bold Extended Caps and Small Caps oc: Computer Modern Bold Extended Slanted Caps and Small Caps rb: Computer Modern Roman Bold bm: Computer Modern Roman Bold Variant ss: Computer Modern Sans Serif si: Computer Modern Sans Serif Slanted sx: Computer Modern Sans Serif Bold Extended so: Computer Modern Sans Serif Bold Extended Slanted tt: Computer Modern Typewriter st: Computer Modern Typewriter Slanted it: Computer Modern Typewriter Italic tc: Computer Modern Typewriter Caps and Small Caps vt: Computer Modern Variable Width Typewriter vi: Computer Modern Variable Width Typewriter Italic

"Each font shape comes in 14 font sizes ranging from 5pt to 35.83pt (or 11 font sizes for typewriter fonts ranging from 8pt to 35.83pt)."

I don't think all sizes are necessary.

if we stick to 10pt as base ... (see type-cbg.tex)

and moreover it would be nice to have

sform5 .. sform10: Computer Modern Concrete Roman sfosl5 .. sfosl10: Computer Modern Concrete Slanted sfoti10: Computer Modern Concrete Italic sfocc10: Computer Modern Concrete Caps and Small Caps

ok, but a different set

(maybe not all sizes here either) and

sfbmr{8,9,10,17}: Computer Modern Bright Roman sfbmo{8,9,10,17}: Computer Modern Bright Oblique sfbsr{8,9,10,17}: Computer Modern Bright Semibold sfbso{8,9,10,17}: Computer Modern Bright Semibold Oblique sfbbx10: Computer Modern Bright Bold Extended sfbtl10: Computer Modern Typewriter Light sfbto10: Computer Modern Typewriter Light Oblique

used? keep in mind that cmsuper is far from perfect

Hm, well, this is all but ~18 (times 13 or 14 sizes)... so maybe it is worth to have them all? What do one loose? speed? work? If work, then I am ready to write what shall be written if you only show me for one font.

i don't want 75 meg font files; i can live with 2 meg extra in the minimal distribution, so a base of 10pt ones sounds ok to me so: - we need a koi vector (enco-koi) - a list of the 10pt font files that make sense

About the encodings: According to my russian friend koi8-r is the most common now, but utf is coming more and more. This is what one person said, so if someone else think it is different, they may very well be right (my friend is mostly TeX:ing on UNIX systems, and from what I read from search results, the koi8r and koi8-r (which seems to be the same?) are mostly used on UNIX and www. So maybe the windows 1521 encoding is still used by Windows users?)

However, I can't get it working with koi8-r. It works in LaTeX (tried with the russian "Not so short introduction to LaTeX" document, and it seemed to use t2a and koi8-r). Under ConTeXt, the document I try (the rexample.tex saved in koi8-r instead of windows 1521) compiles, I get russian letters, but the letters are at wrong places. I am using \enableregime[koi8-r], and saving the document in koi8-r encoding. Is there anything else I should do?

so where does this t2a then fit in? we can make koi the main one and move the t2's to some additional typescript [so that it no longer slows down things] that users can 'load' in their cont-sys.tex Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------