Re: [NTG-context] Example ConTeXt document: PS output
On 11-7-2010 7:39, luigi scarso wrote: 2010/7/11 Ivo Solnickýivo.solni...@gmail.com: Now (not addressed to Mojca), please don't be a smartarse and tell me (again) that this is all volunteer work and to send in patches, a new Device Independent file definition and implementation for Unicode. I'm full of admiration for you guys with the giant IQs who have created ConTeXt in tea breaks and at night, never taken a second of time from the employer, publishing house or university or hairdressing salon proprietor that pays you. But I don't pretend to be in that class. I'm just a mental midget. And I have other things to do. It ain't my project. Something else is. Maybe LuaLaTeX should work for you. But even so there is a dvi-to-ps step to do, which is pratically the same of pdf-to-ps in context mkiv. one of the reasons behind pdf is that it takes the programming language out of ps which means (1) better portability, (2) faster rendering in addition to that pdf provides a couple of goodies irrelevant to printers (annotations and such) dvi as outputformat is fine, but you then need to inject extra code for each (fundamentally) different backend tex + specials - dvi - ps - pdf tex + specials - dvi - pdf tex + literals - pdf (pdftex) now, as ps is seldom needed as final format it makes no sense generate it (and going from pdf - ps is easy) using specials (inlined code) is no fun but has alway sbeen supported in mkii (spec-* files) using an intermediate layer in mkiv we have tex - pdf so no specials and hardly any literals at the tex which is much cleaner; of course there can be more backends but a ps one is unlikely although in principle mkiv could be made to provide dvi to be fed in some backend driver i see no advantage in that Hans - Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl - ___ If your question is of interest to others as well, please add an entry to the Wiki! maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://tex.aanhet.net archive : http://foundry.supelec.fr/projects/contextrev/ wiki : http://contextgarden.net ___
[NTG-context] How to sort in Lua using locale?
Hello,
I would like to create a document using some functionality that is
pretty close to creating an index, but the built-in index creation
code doesn't suffice. My main question is: how do I sort strings using
the built-in sorting rules for the current language?
Here is an example with a table of names:
names = {
{first=Hans, last=Hagen},
{first=Hartmut, last=Henkel},
{first=Taco, last=Hoekwater},
{first=Caron, last=Haček}
{first=Ančka, last=Car}
{first=Ana, last=Car}
{first=Anita, last=Car}
{first=Matija, last=Čop}
}
and I want to print out the following:
\section{C}
\person{Car}{Ana}
\person{Car}{Ančka}
\person{Car}{Anita}
\section{Č}
\person{Čop}{Matija}
\section{H}
\person{Haček}{Caron}
\person{Hagen}{Hans}
\person{Henkel}{Hartmut}
\person{Hoekwater}{Taco}
The following code works for English and needs a fix to respect local
sorting rules. I would be grateful for any hint before reinventing the
wheel (the hint can also be a pointer to the right line of ConTeXt
source code that does the sorting; I was a bit lost in the code).
\mainlanguage[sl]
\setuphead[section][style=\bf,before=\blank,after=,number=no]
\def\person#1#2{\hbox{{\sc #1}, #2}}
\starttext
Code that needs fixing:
\startluacode
names = {
{first=Hans,last=Hagen},
{first=Hartmut, last=Henkel},
{first=Taco,last=Hoekwater},
{first=Caron, last=Haček},
{first=Ančka, last=Car},
{first=Ana, last=Car},
{first=Anita, last=Car},
{first=Matija, last=Čop},
}
-- this should sort with local sorting rules; how?
function sort_names(a,b)
if (a.last == b.last) then
return a.firstb.first
else
return a.lastb.last
end
end
-- this is extremely ugly and needs to be fixed
-- with some UTF-8 library
function first_letter(str)
if (string.byte(str,1)127) then
return string.sub(str,1,2)
else
return string.sub(str,1,1)
end
end
table.sort(names, sort_names)
for k,v in pairs(names) do
-- nested tables would also be an option; but are not required
if (k==1) or (k1 and (first_letter(names[k-1].last) ~=
first_letter(v.last))) then
tex.print(string.format('\\section{%s}', first_letter(v.last)))
end
tex.print(string.format('\\person{%s}{%s}', v.last, v.first))
end
\stopluacode
\crlf\crlf
Desired result:
\section{C}
\person{Car}{Ana}
\person{Car}{Ančka}
\person{Car}{Anita}
\section{Č}
\person{Čop}{Matija}
\section{H}
\person{Haček}{Caron}
\person{Hagen}{Hans}
\person{Henkel}{Hartmut}
\person{Hoekwater}{Taco}
\stoptext
Thanks a lot,
Mojca
___
If your question is of interest to others as well, please add an entry to the
Wiki!
maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage : http://www.pragma-ade.nl / http://tex.aanhet.net
archive : http://foundry.supelec.fr/projects/contextrev/
wiki : http://contextgarden.net
___
Re: [NTG-context] How to sort in Lua using locale?
On 12-7-2010 12:30, Mojca Miklavec wrote:
Hello,
I would like to create a document using some functionality that is
pretty close to creating an index, but the built-in index creation
code doesn't suffice. My main question is: how do I sort strings using
the built-in sorting rules for the current language?
\startluacode
document.whatever = { names = { } }
function document.whatever.sortnames()
local names = document.whatever.names
local split = sorters.splitters.utf
for i=1,#names do
local name = names[i]
name.split = {
split(name.last),
split(name.first),
}
end
sorters.sort(names, sorters.comparers.basic)
end
function document.whatever.flushnames()
local names = document.whatever.names
local previous = false
for i=1,#names do
local name = names[i]
local letter, current = sorters.firstofsplit(name)
if previous ~= current then
previous = current
context.section(letter)
end
context.person(name.last,name.first)
end
end
document.whatever.names = {
{ first = Hans,last = Hagen },
{ first = Hartmut, last = Henkel },
{ first = Taco,last = Hoekwater },
{ first = Caron, last = Haček },
{ first = Ančka, last = Car },
{ first = Ana, last = Car },
{ first = Anita, last = Car },
{ first = Matija, last = Čop },
}
sorters.setlanguage(sl)
document.whatever.sortnames()
document.whatever.flushnames()
\stopluacode
-
Hans Hagen | PRAGMA ADE
Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com
| www.pragma-pod.nl
-
___
If your question is of interest to others as well, please add an entry to the
Wiki!
maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage : http://www.pragma-ade.nl / http://tex.aanhet.net
archive : http://foundry.supelec.fr/projects/contextrev/
wiki : http://contextgarden.net
___
Re: [NTG-context] parsing issue with \math{} inside natural table
On 9-7-2010 10:12, Brian R. Landy wrote:
Hi, I ran into a case where a natural table has an issue when it
contains inline math and an alignment character is used. Here's a small
example which triggers the problem:
\starttext
\bTABLE
\bTABLEbody
\bTR
\bTD[alignmentcharacter={.},aligncharacter=yes]{\le 4.0\math{-}}\eTD
\eTR
\eTABLEbody
\eTABLE
\stoptext
However, if I replace \math{-} with $-$, the file works. Am I correct
that $-$ and \math{-} should be interchangeable?
side effect of alignmentcharacter that triggers expansion of cell
content, so just use $ here
-
Hans Hagen | PRAGMA ADE
Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com
| www.pragma-pod.nl
-
___
If your question is of interest to others as well, please add an entry to the
Wiki!
maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage : http://www.pragma-ade.nl / http://tex.aanhet.net
archive : http://foundry.supelec.fr/projects/contextrev/
wiki : http://contextgarden.net
___
Re: [NTG-context] Better support for Croatian in lang-sla.tex (patch included)
On 11-7-2010 11:46, Vedran Miletić wrote: Here is a better patch, with new strings translated and v!appendix commented as in lang-ger.tex. I could also add support for Serbian latin, but I have no idea how to add new language that will allow both latin and cyrilic. we need to define what that means then ... as i can imagine only combining multiple languages with respect to hyphenation in that case it's not so much a new language as more a matter of defining patterns in mkii this is not feasible due to the relationship between fonts and languages etc but in mkiv it's quite doable i actually have a mkiv prototype for that but as it breaks a few things i have to redo it so it it will show up some time soon Hans - Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl - ___ If your question is of interest to others as well, please add an entry to the Wiki! maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://tex.aanhet.net archive : http://foundry.supelec.fr/projects/contextrev/ wiki : http://contextgarden.net ___
Re: [NTG-context] Serbian: Latin vs. Cyrillic
Datuma 11. srpnja 2010. 13:24 Mojca Miklavec mojca.miklavec.li...@gmail.com je napisao/la: It would be great if ConTeXt supported: \mainlanguage [serbian] [script=latin | cyrillic] as well as \mainlanguage [english] [variant=american | british] etc. This is what Polyglossia does and I find it nice. +1. How hard would it be to implement that? -- Vedran Miletić ___ If your question is of interest to others as well, please add an entry to the Wiki! maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://tex.aanhet.net archive : http://foundry.supelec.fr/projects/contextrev/ wiki : http://contextgarden.net ___
Re: [NTG-context] Serbian: Latin vs. Cyrillic
On 12-7-2010 4:00, Vedran Miletić wrote: Datuma 11. srpnja 2010. 13:24 Mojca Miklavec mojca.miklavec.li...@gmail.com je napisao/la: It would be great if ConTeXt supported: \mainlanguage [serbian] [script=latin | cyrillic] as well as \mainlanguage [english] [variant=american | british] etc. This is what Polyglossia does and I find it nice. +1. How hard would it be to implement that? i have no clue what it means but we have already inheritance in place since ages so effectively we have: \installlanguage [usenglish] [en-us] \installlanguage [ukenglish] [en-gb] \installlanguage [english] [en-us] and adding \installlanguage [brittish][en-uk] is no big deal i see no advantage in adding yet another mechanism for that with regards to script ... not sure how that relates as the language mechanism does two things - labels and conventions (time, date, numbering, quoting, ...) - hyphenation control - Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl - ___ If your question is of interest to others as well, please add an entry to the Wiki! maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://tex.aanhet.net archive : http://foundry.supelec.fr/projects/contextrev/ wiki : http://contextgarden.net ___
Re: [NTG-context] Better support for Croatian in lang-sla.tex (patch included)
2010/7/12 Hans Hagen pra...@wxs.nl: On 11-7-2010 11:46, Vedran Miletić wrote: Here is a better patch, with new strings translated and v!appendix commented as in lang-ger.tex. I could also add support for Serbian latin, but I have no idea how to add new language that will allow both latin and cyrilic. we need to define what that means then ... as i can imagine only combining multiple languages with respect to hyphenation In case of Serbian it's in fact about dealing with two completely different languages in terms of TeX. Latin script and Cyrillic script require different patterns, different font encodings and different labels. The \mainlanguage[serbian][script=latin] and [script=cyrillic] would only be a more elegant/proper way as opposed to \mainlanguage[serbian-latn] and \mainlanguage[serbian-cyrl] (or serbian-latin or serbianl or sr-latn or ...). In reality the two should be two different languages, the second parameter would only be a more elegant way, like for German. Polyglossia has [german][spelling=old/new] (or something similar). If you simply choose two names for language (for example sr-latn and sr-cyrl), it should be straightforward to add both languages to MKII as well. Mojca ___ If your question is of interest to others as well, please add an entry to the Wiki! maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://tex.aanhet.net archive : http://foundry.supelec.fr/projects/contextrev/ wiki : http://contextgarden.net ___
Re: [NTG-context] Better support for Croatian in lang-sla.tex (patch included)
On 12-7-2010 6:01, Mojca Miklavec wrote:
2010/7/12 Hans Hagenpra...@wxs.nl:
On 11-7-2010 11:46, Vedran Miletić wrote:
Here is a better patch, with new strings translated and v!appendix
commented as in lang-ger.tex.
I could also add support for Serbian latin, but I have no idea how to
add new language that will allow both latin and cyrilic.
we need to define what that means then ... as i can imagine only combining
multiple languages with respect to hyphenation
In case of Serbian it's in fact about dealing with two completely
different languages in terms of TeX. Latin script and Cyrillic
script require different patterns, different font encodings and
different labels.
The \mainlanguage[serbian][script=latin] and [script=cyrillic] would
only be a more elegant/proper way as opposed to
\mainlanguage[serbian-latn] and \mainlanguage[serbian-cyrl] (or
serbian-latin or serbianl or sr-latn or ...). In reality the two
should be two different languages, the second parameter would only be
a more elegant way, like for German. Polyglossia has
[german][spelling=old/new] (or something similar).
so that would eventually expand to
\mainlanguage[serbian-\languageparameter{script}-\languageparameter{spelling}-..-\languageparameter{whoknowswhat}]
the danger of that approach is that we end up with as many keys as
languages because most are only used (ok, they are probably then also
misused) for one case (and no one knows what's used for what then)
in mkiv i'd like to follow another route also because we might want to
bring more typographical issues in the game
for instance, although not yet interfaced there is a difference between
labellanguage
headlanguage
mathlabellanguage
that takes care of labels but we migth need something country related
for dates, maybe
conversioncountry
and for patterns we can just use
patternset
when setting up a specific language we then choose a preset i.e. a
combination of settings and serbian-latin is quite ok then
this all is something i'm working on and is also influenced by
discussions in the oriental tex team
If you simply choose two names for language (for example sr-latn and
sr-cyrl), it should be straightforward to add both languages to MKII
as well.
in mkii one can mix font encodings and multiple font encodings can be
set up per language; so, it should be possible to define a language with
two encodings and then when a font switch takes place the right ones
will be chosen. But anyhow, mkii will not be extended and certainly not
in areas were there was never any demand.
Hans
-
Hans Hagen | PRAGMA ADE
Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com
| www.pragma-pod.nl
-
___
If your question is of interest to others as well, please add an entry to the
Wiki!
maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage : http://www.pragma-ade.nl / http://tex.aanhet.net
archive : http://foundry.supelec.fr/projects/contextrev/
wiki : http://contextgarden.net
___
Re: [NTG-context] parsing issue with \math{} inside natural table
Thanks for the reply. I'm actually generating TeX from an application
that interprets $ in it's own code and occasionally this causes me
trouble in passing the $ through to TeX to use in math mode (I think
they have a bug and am hoping I can get that fixed). So I tend to use
\math{} to avoid complications with escaping $, but was stuck in this
case because I couldn't use $ or \math{}.
Anyway, I just now solved my problem in the other application and am
able pass through $ appropriately so it's no longer an issue for me.
Thanks,
Brian
On Mon, 12 Jul 2010, Hans Hagen wrote:
On 9-7-2010 10:12, Brian R. Landy wrote:
Hi, I ran into a case where a natural table has an issue when it
contains inline math and an alignment character is used. Here's a
small
example which triggers the problem:
\starttext
\bTABLE
\bTABLEbody
\bTR
\bTD[alignmentcharacter={.},aligncharacter=yes]{\le 4.0\math{-}}\eTD
\eTR
\eTABLEbody
\eTABLE
\stoptext
However, if I replace \math{-} with $-$, the file works. Am I correct
that $-$ and \math{-} should be interchangeable?
side effect of alignmentcharacter that triggers expansion of cell
content, so just use $ here
-
Hans Hagen | PRAGMA ADE
Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com
| www.pragma-pod.nl
-
___
If your question is of interest to others as well, please add an entry
to the Wiki!
maillist : ntg-context@ntg.nl /
http://www.ntg.nl/mailman/listinfo/ntg-context
webpage : http://www.pragma-ade.nl / http://tex.aanhet.net
archive : http://foundry.supelec.fr/projects/contextrev/
wiki : http://contextgarden.net
___
___
If your question is of interest to others as well, please add an entry to the
Wiki!
maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage : http://www.pragma-ade.nl / http://tex.aanhet.net
archive : http://foundry.supelec.fr/projects/contextrev/
wiki : http://contextgarden.net
___
Re: [NTG-context] parsing issue with \math{} inside natural table
On Mon, 12 Jul 2010, Brian R. Landy wrote:
Thanks for the reply. I'm actually generating TeX from an application that
interprets $ in it's own code and occasionally this causes me trouble in
passing the $ through to TeX to use in math mode (I think they have a bug and
am hoping I can get that fixed). So I tend to use \math{} to avoid
complications with escaping $, but was stuck in this case because I couldn't
use $ or \math{}.
With expansion problems, sometimes using \noexpand\math{...} works.
Aditya
___
If your question is of interest to others as well, please add an entry to the
Wiki!
maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage : http://www.pragma-ade.nl / http://tex.aanhet.net
archive : http://foundry.supelec.fr/projects/contextrev/
wiki : http://contextgarden.net
___
[NTG-context] \frac with Cambria
\usetypescript[cambria]
\setupbodyfont[cambria,8pt]
\starttext
\inlinemath{\displaystyle\frac{(1-0.1) -
0.89803921568627}{1-0.1}}%
{\switchtobodyfont[modern,8pt] \inlinemath{\displaystyle\frac{(1-0.1) -
0.89803921568627}{1-0.1}}}
\stoptext
IMO the numerator with cambria is too much distant from fraction's
line: the same with LatinModern looks better.
Denominator looks ok.
--
luigi
___
If your question is of interest to others as well, please add an entry to the
Wiki!
maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage : http://www.pragma-ade.nl / http://tex.aanhet.net
archive : http://foundry.supelec.fr/projects/contextrev/
wiki : http://contextgarden.net
___
Re: [NTG-context] Better support for Croatian in lang-sla.tex (patch included)
On 12.07.2010 21:23, ntg-context-requ...@ntg.nl wrote: we need to define what that means then ... as i can imagine only combining multiple languages with respect to hyphenation Much anticipated! -- Best Regards, Vyatcheslav Yatskovsky ___ If your question is of interest to others as well, please add an entry to the Wiki! maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://tex.aanhet.net archive : http://foundry.supelec.fr/projects/contextrev/ wiki : http://contextgarden.net ___
