- ntg-context - NTG Mailing Lists

new upload / math
by Hans Hagen 11 Jun '21

11 Jun '21

Hi, In addition the the mentioned integral trickery we can now do this \setupbodyfont[newcomputermodern-book] \enableexperiments[math.extensibles] \starttext % \startformula % x = \integral^1_2 \mathinner{\vrule height 10mm depth 10mm width 5mm} = % x = \integral^1_2 \mathinner{\vrule height 20mm depth 20mm width 5mm} = % x = \integral^1_2 \mathinner{\vrule height 30mm depth 30mm width 5mm} = % x = \integral^1_2 \mathinner{x} = % x = \integral^1_2 x % \stopformula % \startformula % x = \int\Umathadapttoright^1_2 \mathinner{\vrule height 10mm depth 10mm width 5mm} = % x = \int\Umathadapttoright^1_2 \mathinner{\vrule height 20mm depth 20mm width 5mm} = % x = \int\Umathadapttoright^1_2 \mathinner{\vrule height 30mm depth 30mm width 5mm} = % x = \int\Umathadapttoright^1_2 \mathinner{x} = % x = \int\Umathadapttoright^1_2 x % \stopformula % $ % x = \int\Umathadapttoright^1_2 \mathinner{\vrule height 10mm depth 10mm width 5mm} = % x = \int\Umathadapttoright^1_2 \mathinner{\vrule height 20mm depth 20mm width 5mm} = % x = \int\Umathadapttoright^1_2 \mathinner{\vrule height 30mm depth 30mm width 5mm} = % x = \int\Umathadapttoright^1_2 x = % $ \def\rab{\Umathchar"1"00"007C } \startTEXpage[offset=3cm] $ \color[green]{\rab\Umathadapttoright^{\black 1}_{\black 2}^^3__4} \color[blue] {\mathinner{\vrule height 30mm depth 30mm width 5mm}} \color[red] {\rab\Umathadapttoleft ^5_6^^{\black 7}__{\black 8}} $ \stopTEXpage \startTEXpage[offset=10pt] % \showglyphs $ \rab\Umathadapttoright^1_2^^3__4 \mathinner{\vrule height 30mm depth 30mm width 5mm} \rab\Umathadapttoleft^5_6^^7__8 $ \stopTEXpage \stoptext So, we can have left and right sided adaptation. The \enableexperiments[math.extensibles] is a cheat that makes the extensible into one character (cut and paste related). Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl -----------------------------------------------------------------

2 1

upload / some new math stuff
by Hans Hagen 11 Jun '21

11 Jun '21

Hi, I played a bit with these newcomputermodern fonts. They are extended versions of latin modern and come in: - 10pt (serif, sane, mono, math) - book (serif, sane, mono, math) - 8pt (partial) I made a typescript for the 10pt regular and book (as I see no reason to configure 8pt). So, assuming that one installs these fonts: \setupbodyfont[newcomputermodern-book] \setupbodyfont[newcomputermodern] should work ok. The book variant is quite nice. I didn't look into the provided math alternates (context supports that but I didn't define them for this font ... input from math lovers is welcome). Now, this font also implements, as cambria does, upright integrals and as that has comes on my radar every year when I ponder these integrals I decided that it was about tiem to deal with it properly. It is no big deal to implement them gropwing (after all one can see them as left delimiter) but the main issue are these super and subscripts. \setupbodyfont[newcomputermodern-book] % \setupbodyfont[cambria] \starttext \startformula x = \integral^1_2 \mathinner{\vrule height 10mm depth 10mm width 5mm} = x = \integral^1_2 \mathinner{\vrule height 20mm depth 20mm width 5mm} = x = \integral^1_2 \mathinner{\vrule height 30mm depth 30mm width 5mm} = x = \integral^1_2 \mathinner{x} = x = \integral^1_2 x \stopformula \startformula x = \int\Umathadaptsize^1_2 \mathinner{\vrule height 10mm depth 10mm width 5mm} = x = \int\Umathadaptsize^1_2 \mathinner{\vrule height 20mm depth 20mm width 5mm} = x = \int\Umathadaptsize^1_2 \mathinner{\vrule height 30mm depth 30mm width 5mm} = x = \int\Umathadaptsize^1_2 \mathinner{x} = x = \int\Umathadaptsize^1_2 x \stopformula \def\rab{\Umathchar"1"00"007C } $ x = \int\Umathadaptsize^1_2 \mathinner{\vrule height 10mm depth 10mm width 5mm} = x = \int\Umathadaptsize^1_2 \mathinner{\vrule height 20mm depth 20mm width 5mm} = x = \int\Umathadaptsize^1_2 \mathinner{\vrule height 30mm depth 30mm width 5mm} = x = \int\Umathadaptsize^1_2 x = x = \rab\Umathadaptsize^1_2^^1__2 \mathinner{\vrule height 30mm depth 30mm width 5mm} $ \stoptext After playing with several solutions (delimiters, fences, radicals etc) I just settled for a few line extension to the engine (there was already plenty in place in luametatex for that anyway) and it seem to work ok. This \Umathadaptsize is a bit like the limits related suffixes (and uses the same scanner logic). It basically flags an operator as something that should grow cf what follows. I can add some fine tuning if needed (can also be dealt with at the macro end, as with other mechanisms, but this is just a first test of useability). (I might also extend another mechanism to deal with it but putting scripts on top or after is already handles pretty well by regular mechanisms, so ...) Of course only a few fonts provide these upright growing integrals (makes me wonder about summation and product). Concerning the newcm fonts: for some reason the book variants (otf files) are about twice as large (in filesize) but what we use (cached) is the same size (only bboxes differ) so there is something fishy in the otf. I can imagine that at some point we add the 10pt/book ones to the standard mkiv distribution and maybe only the subset to the lmtx distribution but for now these book ones are too big for that, but it all depende on them being used. Happy testing (hopefully there are no side effects). Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl -----------------------------------------------------------------

5 6

Incorrect sorting in \completelistofabbreviations for numbers
by Markus Brandt 10 Jun '21

10 Jun '21

Hi everyone, I think I've found a sorting bug in \completelistofabbreviations in the latest LMTX (and probably also MKIV). When using this minimal example \abbreviation[OTT]{123}{One, Two, and Three} \abbreviation{ABC}{A, B, and C} \abbreviation{ZYX}{Z, Y, and X} \starttext \completelistofabbreviations[criterium=all] \stoptext I get ABC A, B, and C 123 One, Two, and Three ZYX Z, Y, and X Numbers should go before letters. Do you know any workaround/fix for that? Best, Markus

3 7

new in upload
by Hans Hagen 10 Jun '21

10 Jun '21

\starttext \enabletrackers[localanchor] \startoverlayMPgraphic{mp:whatever-6} draw anchorcell (1,1) withpen pencircle scaled 1pt withcolor "blue" ; fill anchorspan (2,2) (3,5) withpen pencircle scaled 1pt withcolor "yellow" ; draw anchorcell (2,2) withpen pencircle scaled 1pt withcolor "green" ; draw anchorcell (3,3) withpen pencircle scaled 1pt withcolor "red" ; \stopoverlayMPgraphic \starttabulate[synchronize=background,background=mp:whatever-6,format={|l|c|r|}] \NC test \NC test \NC test \NC \NR \NC test \NC test test \NC test \NC \NR \NC test test \NC test \NC test \NC \NR \NC test \NC test \NC test \NC \NR \NC test \NC test \NC test test \NC \NR \stoptabulate \stoptext -- ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl -----------------------------------------------------------------

2 2

Calculating Best Box Width?
by Michael Urban 09 Jun '21

09 Jun '21

I am not especially facile with ConTeXt, and it has been years since I did anything complex with TeX, so I am hoping someone could help me with this. The following is probably not great code, but it shows what I am trying to do. The problems are twofold: a benign problem is that I get underfull hbox errors (\dontcomplain notwithstanding). But the other problem is that with text of a certain size, the final line is much shorter than the others. Is there some clever way to repeatedly typeset the text (e.g., put it in an hbox, measure the width, and repeatedly try .5, .333, .25 of the hbox width until it fits in the allocated space)? How to go about this? \defineblank[EpigraphDistance][3pt] \definestartstop[EpigraphText][ style=\ssa, before={\blank[EpigraphDistance] \setupnarrower[left=.25\textwidth,right=0pt] \startnarrower[left,right] \setupalign[flushright,nothyphenated,broad] \dontcomplain \noindent}, after={\stopalignment\stopnarrower \blank[EpigraphDistance] \indenting[next]}] \long\def\epigraph#1#2#3{% \startEpigraphText #1 \par \stopEpigraphText \ifx\hfuzz#3\hfuzz \rightaligned{\ssa\sl --- #2} \else \rightaligned{\ssa\sl --- #2, \tf\ssa\symbol[leftquote]#3\symbol[rightquote]} \fi } \starttext Testing epigraph \epigraph{% para \quotation{My birthday-present!} he whispered to himself, as he had often done in the endless dark days. \quotation{That's what we\unknown} }{J.R.R. Tolkien}{The Hobbit} That was an epigraph. \stoptext

2 1

Font Maguntia not rendering correctly...
by Willi Egger 09 Jun '21

09 Jun '21

Hi all, i have a text from an old publication typeset in Fraktur. — The font I would like to use here is Unifraktur Maguntia. This Font worked before quite well, however now I get wrong renderings. Please refer to the attached MWE. Context version: 2021.06.01 17:04 LMTX What is oing wrong here? Kind regards Willi

3 2

PDF forms not creating proper children
by Leonard Janis Robert König 09 Jun '21

09 Jun '21

Hello, I currently try to build a document with AcroForms, e.g., the examples from [1] also linked in the Wiki. The most recent example is in the form/ subdirectory and dated mid 2020. However, when building the file(s) with my ConTeXt distribution the file created is broken, in comparison to the PDF checked in into the Git. The more obvious difference is that the PDF in the repository can be filled in the Firefox 89 beta, while the other PDF doesn't even display the fields as forms. Running pdfinfo also yields: > Syntax Error: Form field child is not a dictionary I suspect that due to some breakage the forms children aren't set in the AcroForms dictionary anymore? My ConTeXt versions are: # ArchLinux Repos context --version mtx-context | ConTeXt Process Management 1.04 mtx-context | mtx-context | main context file: /usr/share/texmf- dist/tex/context/base/mkiv/context.mkiv mtx-context | current version: 2021.03.05 19:11 mtx-context | main context file: /usr/share/texmf- dist/tex/context/base/mkxl/context.mkxl mtx-context | current version: 2021.03.05 19:11 # Debian Buster Repos mtx-context | ConTeXt Process Management 1.02 mtx-context | mtx-context | main context file: /usr/share/texmf/tex/context/base/mkiv/context.mkiv mtx-context | current version: 2018.04.04 00:51 # Docker Image: Current mtx-context | ConTeXt Process Management 1.03 mtx-context | mtx-context | main context file: /context/tex/texmf- context/tex/context/base/mkiv/context.mkiv mtx-context | current version: 2020.01.30 14:13 mtx-context | main context file: /context/tex/texmf- context/tex/context/base/mkiv/context.mkxl mtx-context | current version: 2020.01.30 14:13 # Docker Image: Beta mtx-context | ConTeXt Process Management 1.03 mtx-context | mtx-context | main context file: /context/tex/texmf- context/tex/context/base/mkiv/context.mkiv mtx-context | current version: 2020.01.30 14:13 mtx-context | main context file: /context/tex/texmf- context/tex/context/base/mkiv/context.mkxl mtx-context | current version: 2020.01.30 14:13 # Docker Image: LMTX mtx-context | ConTeXt Process Management 1.04 mtx-context | mtx-context | main context file: /context/tex/texmf- context/tex/context/base/mkiv/context.mkiv mtx-context | current version: 2021.04.14 22:58 mtx-context | main context file: /context/tex/texmf- context/tex/context/base/mkxl/context.mkxl mtx-context | current version: 2021.04.14 22:58 # Docker Image: Beta from 2019-09 mtx-context | ConTeXt Process Management 1.03 mtx-context | mtx-context | main context file: /context/tex/texmf- context/tex/context/base/mkiv/context.mkiv mtx-context | current version: 2019.09.10 20:03 mtx-context | main context file: /context/tex/texmf- context/tex/context/base/mkiv/context.mkxl mtx-context | current version: 2019.09.10 20:03 The ConTeXt version of the checked in file on GitHub is (according to pdfinfo): > Creator: LuaTeX 1.10 7138 + ConTeXt MkIV 2019.08.20 17:20 > Producer: macOS Version 10.14.6 (Build 18G5033) Quartz PDFContext I also remebered another time were I toyed around with it and indeed that PDF also had broken form field children: > Creator: LuaTeX 1.12 7306 + ConTeXt MkIV 2020.03.10 14:44 > Producer: LuaTeX-1.12 A broken version that I built locally using the :current IslandOfTeX [2] can be found at [3]. The version difference between the broken and not-broken PDFs hints that the breakage occured between 2019.08.20 and 2019.09.10-beta, while the Debian breackage might be due to a backport(?) Alternatively, this could just be a mistake on my side? Cheers Leo König [1]: https://github.com/fiee/ConTeXt [2]: https://gitlab.com/islandoftex/images/context/ [3]: https://mega.nz/file/ThcghTqb#JqRCSsd59bBbMgTMr3ahaPdU4LPlZd0cpA3ZFBJLJ3A

4 12

LMTX problem with doifmode in \setuplayout
by Rik Kabel 08 Jun '21

08 Jun '21

Hello all, Odd issue with LMTX (works fine with --luatex). \doifelsemode causes a failure if it is used to set a simple value such as on, off, yes, no, ..., but works with a dimension. Perhaps better illustrated by an example: \definemode [aaa] [keep] \setuplayout [ marking=\doifmodeelse{aaa}{off}{on}, % marking=on, grid=\doifmodeelse{aaa}{yes}{no}, % grid=yes, bottom=\doifmodeelse{aaa}{10mm}{20mm}, ] \starttext This is only a test. \stoptext The problem occurs whether or not the mode is enabled. In the example, it fails with the mode test in the either the marking= or grid= keys, but has no problem with it in the bottom= key. The error message says: The control sequence marked <to be read again> should not appear between \csname and \endcsname. pointing to the enabled \doifmodeelse line. The same issue occurs with the following, using \doifmode: \definemode [aaa] [keep] \setuplayout [ marking=on, marking=\doifmode{aaa}{off}, ] \starttext This is only a test. \stoptext This did work without complaint a few months ago. -- Rik Kabel

3 6

keep text strictly inside box
by Hans van der Meer 08 Jun '21

08 Jun '21

As can be seen in this screenshot, the text overshoots on the right side the boundary of the fairly narrow \framedtext. My intent is to not hyphenate the names, even if that means a ragged edge on the right side. But at the same time the text must absolutely stay within the confines of the framedtext. I played with \raggedright, \tolerance, \looseness, \emergencystretch but no combination of them seems sufficient fto force all gtext inside. Is there another solution to force the text inside the box or do I have to insert linebreaks manually (a nuisance to be avoided, of course.) dr. Hans van der Meer

3 3

frameoffset in textbackground?
by Hans van der Meer 08 Jun '21

08 Jun '21

In the following MWE for a textbackground the parameter backgroundoffset has the expected effect. But for positive values only. A backgroundoffset=-10pt has no visible effect. But the real problem is the parameter frameoffset. No effect at all, whatever frameoffset= is set to. dr. Hans van der Meer

2 1