- ntg-context - NTG Mailing Lists

ConTeXt Meeting #7
by Leo Arnold 13 May '13

13 May '13

Hi there, will the registration open before the end of early bird discount? Cheers Leo

3 4

[***SPAM***] unwanted space and error in combination of writebetweenlist and placecontent[criterium= local]
by Zenlima 12 May '13

12 May '13

Hi, I ran into one error and one problem during building a part as two columns into the toc. 1. Extra line: I am getting an unwanted extra empty line at the bottom of two columns with an even number of total lines - with an odd number of total lines no extra empty line appears. How can I get rid of it? (see attached example) 2. This error only appears only with \placecontent[criterium=local] in combination with a beginning line of \writebetweenlist; with a global \placecontent and \writebetweenlist it works fine. To reproduce this uncomment the line "%\placecontent[criterium=local]" in the attached example. (Sorry that I was not able to make the example shorter, I needed some "meat" to fill the toc for the showcases) Thanks, H. --8<---------------------------- \usemodule[subsub] \definehead[paragraphHead][alternative=myheader,section=section-8,number=no] \defineheadalternative [myheader] [alternative=horizontal, renderingsetup=setup:myheader] \startsetups [setup:myheader] \inouter{\headtextcontent} \stopsetups \setuplist[chapter][style=bold,before=\blank,alternative=b,width=1.5em,color=HeadColor] \setuplist[section][margin=1.5em,width=2em] \setuplist[subsection][margin=3.5em,width=3em] \setuplist[paragraphHead][alternative=c,style=\tfx] \def\startParagraphTocColumns{% \writebetweenlist[paragraphHead][location=here]{\setupnarrower[left=2em]\startnarrower[left]\startcolumns[n=2,distance=0mm,balance=yes]}% \ignorespaces} \def\stopParagraphTocColumns{% \writebetweenlist[paragraphHead][location=here]{\stopcolumns\stopnarrower}% \ignorespaces} \starttext \setupcombinedlist[content][list={chapter,section,subsection,textdividerHead,paragraphHead}] \completecontent \startParagraphTocColumns \dorecurse{5}{\paragraphHead{Hasselt makes headlines} \input tufte} \stopParagraphTocColumns \chapter{Lorem ipsum} \section{Sed eram} \chapter{Lorem ipsum} %\placecontent[criterium=local] \section{Sed eram} \section{Sed eram} \startParagraphTocColumns \dorecurse{5}{\paragraphHead{Hasselt makes headlines} \input tufte} \stopParagraphTocColumns \section{Sed eram} \section{Sed eram} \startParagraphTocColumns \dorecurse{4}{\paragraphHead{Hasselt makes headlines} \input tufte} \stopParagraphTocColumns \section{Sed eram} \startParagraphTocColumns \dorecurse{3}{\paragraphHead{Hasselt makes headlines} \input tufte} \stopParagraphTocColumns \section{Sed eram} \section{Sed eram} \stoptext

1 0

[luatex-fonts] +mkmk; +mark results in call to undefined onetimemessage()
by Philipp Gesang 12 May '13

12 May '13

Hi, in luatex-plain, the example below \font\grk="file:LinLibertine_R.otf:+mark;+mkmk;mode=node" at 14.4pt {\grk μῆνιν ἄειδε θεὰ Πηληϊάδεω Ἀχιλῆος\hfill\break οὐλομένην, ἣ μυρί' Ἀχαιοῖς ἄλγε' ἔθηκε} \bye triggers a call to onetimemessage() which is nil at the time when the upvalue is created. Can we have calls to fonts.loggers.onetimemessage() instead, so we can override that one with our own logger? On a related note, we would like to reduce verbosity. In luatex-basics-gen.lua, could we have: logs = logs or { new=dummyreporter, reporter=dummyreporter, messenger=dummyreporter, report=dummyfunction, } So we can provide our own “logs” table? Thanks Philipp

3 2

Re: [NTG-context] Arabic typesetting, one more
by "H. Özoguz" 12 May '13

12 May '13

> \usemodule[fnt-20,art-01] > > \starttext > > \def\sample{اللهِ\quad اللَّـهِ} > > \setvariables > [otftracker] > [font=name:traditionalarabicnormal, > size=24pt, > features=arabic, > direction=-1, > title=Feature Check, > sample=\sample] > > \page > > \setvariables > [otftracker] > [font=arabtype, > size=24pt, > features=arabic, > direction=-1, > title=Feature Check, > sample=\sample] > > \stoptext Yes, again, kassra is set too low, in both fonts. But with traditionalarabic it is worse than with arabtype. What could be a way to fix this? Huseyin

4 3

python parse/formatter example on the wiki
by seasoul 11 May '13

11 May '13

I was trying the example of "writing a parse/formatter" under "Verbatim text". The python formatter runs successfully. But when I compile the text file with all the definitions included, an error occurs. <inserted text> } <to be read again> \endgroup \buff_verbatim_type_block_verbatim_indeed ...ines \dostoptagged \csname #2\e... It could be implemented forcefully, but definitely not an excuse to ignore this error. Any solutions?

2 2

Re: [NTG-context] Formulas in ConTeXt are different from those in plain TeX
by Hans Hagen 11 May '13

11 May '13

On 5/11/2013 12:16 PM, Tim Li wrote: > Thanks, Hans, > > can you give me a small example or show me how to do that? I really > don't know how to do that. :-( see plain.tex and the tex book but you can bet on interference ... deep down, math in context is implemented in a rather special way and that might conflict with more traditional approaches 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 -----------------------------------------------------------------

1 0

Formulas in ConTeXt are different from those in plain TeX
by Tim Li 11 May '13

11 May '13

Hi, I have asked this question on tex.stackexchange.com http://tex.stackexchange.com/questions/113592/formulas-in-context-are-diffe… as the title said, formulas typeset by ConTeXt are different from those by plain TeX. The minimal examples are here: % plain luatex in ConTeXt standalone $$\int_a^bf(x)dx=F(b)-F(a)$$ \bye % context \starttext \startformula \int\nolimits_a^bf(x)dx=F(b)-F(a) \stopformula \stoptext Marco said to me that that's because the fonts used by LuaTeX and ConTeXt are different. This can be seen from the pdf produced by them pdffonts context.pdf name type emb sub uni object ID ------------------------------------ ----------------- --- --- --- ---------OLAZTU+LMRoman12-Regular CID Type 0C yes yes yes 18 0 RXHAAY+LatinModernMath-Regular CID Type 0C yes yes yes 19 0JTHTZJ+LatinModernMath-Regular CID Type 0C yes yes yes 20 0 pdffonts plain.pdf name type emb sub uni object ID------------------------------------ ----------------- --- --- --- --------- APMHZE+LMMathExtension10-Regular Type 1 yes yes no 4 0 AWBTPY+LMMathItalic7-Regular Type 1 yes yes no 5 0 ITITNI+LMMathItalic10-Regular Type 1 yes yes no 6 0 DMNVHT+LMRoman10-Regular Type 1 yes yes no 7 0 MGKPYK+LMMathSymbols10-Regular Type 1 yes yes no 8 0 even if using the `modern-designsize-virtual` typescript(advice from Marco), I still get the different outputs. Now, I get confused about this qustion. If ConTeXt uses LuaTeX as its engine, how can I tune ConTeXt so that it can produce the same formulas with those in plain TeX? Tim

2 1

[luatex-fonts] font cache behavior
by Philipp Gesang 10 May '13

10 May '13

Hi all, this document: \font\foo="file:LinLibertine_R.otf:+onum;mode=node;" \foo foo 12345 whatever \bye works correctly if the font is uncached but leads to garbled output in firefox’s pdf plugin if the font is loaded from cache. Afaics none of the other viewers are affected. Also I haven’t found another font that does this. Libertine fonts: http://sourceforge.net/projects/linuxlibertine/files/linuxlibertine/5.3.0/ (First observed by Luigi.) What is going on here? Philipp

2 1

Re: [NTG-context] [luatex] font feature with new luaotfload
by Philipp Gesang 10 May '13

10 May '13

···<date: 2013-05-10, Friday>···<from: Hironori KITAGAWA>··· > I tried the following source with TeX Live 2013 pretest. > > --- > \documentclass{article} > \usepackage{fontspec} > \begin{document} > \fontspec[RawFeature=+pwid]{KozMinPr6N-Regular} > A…B > > \fontspec[RawFeature=+fwid]{KozMinPr6N-Regular} > A…B > > \end{document} > --- > > But I got unintentional result from LuaLaTeX with luaotfload v2.2. > In concrete terms, '…' in the second line should be replaced by > its fullwidth variants CID #668 (as in koz-xetex.pdf), > but it doesn't so ('…' is not replaced) in koz-luatex.pdf. Confirmed. This is the code translated to Context: \definefontfeature[pwid][mode=node,script=latn,language=dflt,pwid=yes] \definefontfeature[fwid][mode=node,script=latn,language=dflt,fwid=yes] \starttext {\definedfont[file:kozminpr6n_regular.otf*pwid]A…B} {\definedfont[file:kozminpr6n_regular.otf*fwid]A…B} \stoptext Below is the patch translated for font-otf.lua. Best Philipp --- font-otf.lua.orig 2013-05-10 13:24:16.252748027 +0200 +++ font-otf.lua 2013-05-10 13:25:00.830697179 +0200 @@ -644,6 +644,9 @@ if not unicode or unicode == -1 or unicode >= criterium then unicode = cidunicodes[index] end + if unicode and descriptions[unicode] then + unicode = -1 + end if not unicode or unicode == -1 or unicode >= criterium then if not name then name = format("u%06X",private)

1 0

Re: [NTG-context] Installing an arabic font
by "H. Özoguz" 10 May '13

10 May '13

> I uploaded a beta that might fix it, but it needs more testing. The > issue might have been introduced when support was added for more complex > ligature building, compare for instance this: Thanks Hans! Indeed, it was a fix, but there are still problems I find in further testing, see attachment for this code: \definefont[amiri][file:UthmanTN1Ver10.ttf*arabic at 17 pt] \define[1]\Arab{\setupalign[r2l]\amiri#1} \starttext \Arab{ لرَّحْمَـٰنِ } \stoptext In context the two signs are set one upon the other ("sukun" and "fatha"), see the second attachment (word) for the correct setting. But now this has nothing todo with ligatures, deactivating them does not help now. Huseyin

2 1