April 2020 - ntg-context - NTG Mailing Lists

Right-to-left pagecolumns and markings in ConTeXt
by Joey McCollum 29 Apr '20

29 Apr '20

I am trying to typeset a two-column text in a right-to-left language using ConTeXt. Because I want footnotes in this text to be set ragged-bottom, I am using pagecolumns instead of standard columns. I also want to track the first and last textual division markers on each page using markings. Combining all of these ingredients seems to create a problem in ConTeXt: when the pagecolumns are set in right-to-left order, the top and bottom marks fetched are always the top and bottom marks in the left column, while the true top mark should be the top mark of the right column. I've observed this behavior using text in an actual right-to-left language, but the following minimal (non-)working example with an English sample text demonstrates the same behavior: ``` %Create a new markset to use: \definemarking[TestMark] %Setup header to reflect top and bottom marks: \setupheadertexts[top: {\fetchmark[TestMark][top]}][bottom: {\fetchmark[TestMark][bottom]}][top: {\fetchmark[TestMark][top]}][bottom: {\fetchmark[TestMark][bottom]}] %even left, even right, odd left, odd right %Setup the columns layout: \definepagecolumns [example] [ n=2, %number of columns direction=reverse, %does not work if ending comma is removed! ] \starttext \startpagecolumns[example] \marking[TestMark]{1}(1) \input knuth\par \marking[TestMark]{2}(2) \input knuth\par \marking[TestMark]{3}(3) \input knuth\par \marking[TestMark]{4}(4) \input knuth\par \marking[TestMark]{5}(5) \input knuth\par \marking[TestMark]{6}(6) \input knuth\par \marking[TestMark]{7}(7) \input knuth\par \marking[TestMark]{8}(8) \input knuth\par \stoppagecolumns \stoptext ``` I typeset this using ConTeXt version 2020.03.10, as released with TeXLive 2020. As my comment in the MWE points out, there also seems to be a bug in the key-value parsing of \definepagecolumns that causes it not to parse the direction=reverse input unless it is followed by a comma. Is there some way I can improve my syntax to avoid this issue, or is it just a bug that needs to be fixed? I've tried fetching other marks (e.g., first and last), but the only marks that ConTeXt seems to find are the ones for the left column. Thank you, Joey

2 4

Typesetting Markdown - Part 8
by Thangalin 28 Apr '20

28 Apr '20

Hi folks! Second last part in the series. The project, product, component, and environment relations were a bit finicky. Those willing to point out improvements that can be made---Hans, Taco, Wolfgang, Aditya, and others---would be very much appreciated. Specifically, it seems that some project environment settings can be overruled inside products, but other project settings cannot. It's not really clear to me whether this is due to dependency order or that definitions cannot be redefined across environments. Thoughts? https://dave.autonoma.ca/blog/2020/04/28/typesetting-markdown-part-8/ Stay safe!

5 13

\setupinteraction [ ...] do not work
by Fabrice Couvreur 28 Apr '20

28 Apr '20

Hi, It seemed to me that it was necessary to proceed in this way to create a link in a pdf file; it does not work. Thank you. Fabrice \setupinteraction[state=start,color=darkred] \starttext \goto{vecteur directeur}[https://www.youtube.com/watch?v=6VdSz-0QT4Y] \stoptext

2 1

LMTX rollback?
by Rik Kabel 27 Apr '20

27 Apr '20

My update script reports: 2020-04-27T12:40:53 ConTeXt updated from 2020.04.19T19:43 to 2020.04.26T19:59 2020-04-27T12:40:53 LuaMetaTeX updated from 2.05.02 to 2.05.01 2020-04-27T12:40:53 LuaMetaTeX functionality updated from 20200413 to 20200402 Is this an intentional rollback or a website issue? -- Rik

1 0

Using \startuseMPgraphic inside a textext() in METAPOST?
by Gerben Wierda 27 Apr '20

27 Apr '20

I am in METAPOST which is inside TeX and I’d like (in METAPOST, not in TeX) to let a string follow a path. Adapting from the MetaFun manual around page 250, I tried this: \starttext \startMPpage[instance=doublefun] pic p; p := textext("\startuseMPgraphic[instance=doublefun]{followtokens}path RotPath ; RotPath := fullcircle ;\stopuseMPgraphic\followtokens{I am creating a line of text here}"); draw pic; \stopMPpage \stoptext But that does not work. Is there a way I can from within my METAPOST code get string (TeX code) to follow a path I have? In my case, I cannot escape outside the \startMPpage- -\stopMPpage environment, as the actual path is constructed there and the text following the path should be on that page. G

2 1

simple way to change value of 'x' in \tfx?
by jbf 27 Apr '20

27 Apr '20

I have the following setup currently, involving: \definefontfamily[mainface][rm][texgyrepagella][tf=file:texgyrepagella-regular.otf] \setupbodyfont[mainface,11pt] \setupbodyfontenvironment[default] [em=italic] \setupdelimitedtext[blockquote][style=\tfx, before={\blank\setupinterlinespace[line=2.4ex]},after={\blank}] My publisher thinks the default scaling of 'x' (0.8) is too small for the blockquoted text, and asked if I could simply indicate 10pt for 'x'. I tried by using the line: \definebodyfontenvironment[default] [x=10pt] but that did nothing to change things. I tried a few other possibilities as well, e.g. just adding the x=10pt after the em=italic declaration, but that didn't work either. I'm sure there is a simple way to achieve what I want. Any thoughts? Julian

2 1

\framed[align=] with a 'maxwidth' instead of a width? Or get the size of a \frame returned to lua without typesetting it in a document?
by Gerben Wierda 26 Apr '20

26 Apr '20

Setup: TeX file Contains ctxlua{} call lua function reads XML file and calls context() statements, it puts out a \startMPpage METAPOST statements Of which some are textext(\framed[align, so vbox, width defined by lua]{}, the whole TeX string for textext is created in lua and passed whole on to METAPOST as METAPOST is pretty limited on string manipulation compared to lua) \stopMPpage I am running into the following conundrum: If I do give it a width (say a maximum) it will also be that wide when it is mostly empty if there is little text, this interferes with optimal placing If I don’t give a width to the \framed command it may grow far too wide if there is a lot of text instead of wrapping around. The typeset text is rather unpredictable. It may be three very short single-word lines with \\ at the end but it may also be a large paragraph. So, I would like to find a way to let the \framed become a box as narrow as possible, but not growing wider than a pre-set width. This is way too difficult for my meagre skills. What I could do (but I think is clumsy) is pass two \framed commands to METAPOST, have both by typeset with textext() and if the one without a width becomes too wide, take the one with the width set. But I would like a neater option. E.g. call TeX (no need for METAPOST in between) from lua to typeset the \framed without actually typesetting it to the document and getting the dimensions back to lua so I can decide there what to do. Or a \specialframed command where I don’t give it a ‘width', but a ‘maxwidth’ Is one of these doable? If not, I’ll have to implement the ‘clumsy’ route. Preferably (the what is now thought of as ideal solution) I would be able in lua to give a piece of TeX code to TeX and ask it to produce a box and give me the dimensions of that box. That alone would give me an option, e.g. - trial typeset - if box surface (w*h) is more than what is available, shrink font size and lineheight so the total will be within set limits, then typeset it again with the smaller font and max width available - else, if trial box width is more than the width available, typeset again with fixed width - else use the way the trial box was typeset But I’m still puzzling. G PS. Why is the bbox of a textext(\framed[]{}) slightly larger than the frame?

1 0

register entries in footnotes
by Thomas A. Schmitz 26 Apr '20

26 Apr '20

A similar question has been asked three years ago, and Rik Kabel posted a partial solution, adapted here: \defineprocessor [Footnote] [right={ n}] \define[1] \fnindex{\index[Footnote->]{#1}} \starttext This is a sentence with a term in the index: cat\index{cat}. This sentence also has an indexed term, dog\index{dog}, as well as a footnote.\footnote{The footnote also contains an indexed term, mouse.\fnindex{mouse}} \page \placeregister [index] \stoptext And about a year ago, Hans made a different suggestion to my own question; however, I didn't realize then that it was not what I was looking for: \starttext Test1: \index{cat}cat. \footnote{Test2: \index{mouse}mouse.} \footnote[note:cat]{Test3: \index[cat]{cat in \in{note}[note:cat]}cat.} \page \placeregister [index] \stoptext This creates two index entries for cat, which is not the intended result. My questions are: 1. I would like to have the footnote number appear in the index, so it says something like "cat 1 n. 1." Is that possible? 2. Is it necessary to define an extra command for index entries in footnotes? Is context clever enough to know that it's processing a footnote, so one could write a condition such as (pseudocode): \define [1] \myindex {\infinfootnoteelse {\index[Footnote]->{#1}} {\index{#1}}} Thanks a lot, have a good Sunday! Thomas

2 1

Re: [NTG-context] Basic font question (Optima, but no bold, no italics). Standalone ConTeXt does not work. TeX Live 2019 works
by Benjamin Buchmuller 26 Apr '20

26 Apr '20

Hi Hraban, Thanks for the hint (and the proper fontfamily setup), I have forced reload ten times now from various accounts; the issue persists unfortunately. This might confirm Taco‘s notion that there is some bug lmtx. I’m running macOS Catalina too, but as it worked with the standalone of Dec/Jan three days ago, I suspect it is not because of a new feature in the OS, but rather an old feature in mtxrun or elsewhere having disappeared. > On 24. Apr 2020, at 11:35, Henning Hraban Ramm <texml(a)fiee.net> wrote: > > > >> Am 24.04.2020 um 11:13 schrieb Benjamin Buchmuller <benjamin.buchmuller(a)gmail.com>: >> >> Although this specific issue might have been solved meanwhile by the helpful suggestions of Wolfang, Thomas and others, I would like to point out that I have recently also encountered an odd behaviour of the mtxrun fontloader in the standalone, in which not all typefaces are available/properly identified. >> >> (I’m sorry to hijack this thread if this is unrelated.) >> >> Basically, when I load “Helvetica Neue”, I get instead of “regular/normal” always “light/italic”. >> >> I have switched to heros for the meantime, but potentially there is some bug behind this. (My documents with the version from Dec/Jan worked fine.) >> >> I recite my example from Wed: >> >> ConTeXt ver: 2020.01.30 14:13 MKIV beta fmt: 2020.4.22 (LuaTeX 1.11.1) >> >> and I’m running now into troubles with system font indexing: >> >> mtxrun --script fonts --reload >> >> locates fonts in macOS directories appropriately (*.afm fonts placed in .../tex/texmf-fonts are not found, but at the moment I don’t care too much), however it resolves only to a single font variant for *.ttc instead of all variants: >> >> mtxrun --script fonts --list --all --pattern=helvetica >> >> helvetica helvetica helveticalightoblique /System/Library/Fonts/Helvetica.ttc 6 >> helveticalightoblique helvetica helveticalightoblique /System/Library/Fonts/Helvetica.ttc 6 >> helveticaneuedeskinterfaceheavy helveticaneuedeskinterface helveticaneuedeskinterfaceheavy /System/Library/Fonts/HelveticaNeueDeskInterface.ttc 10 >> helveticaneuethinitalic helveticaneue helveticaneuethinitalic /System/Library/Fonts/HelveticaNeue.ttc 14 > > If the subfont is not listed, ConTeXt/mtxrun couldn’t find it and won’t find it regardless of your setup. > > Try > mtxrun --script fonts --reload --force > maybe even twice or thrice and check again. > (Unfortunately this is not reliable.) > > I get the full list: > > identifier familyname fontname filename subfont instances > > helvetica helvetica helvetica /System/Library/Fonts/Helvetica.ttc 1 > helveticaneueblack helveticaneue helveticaneuecondensedblack /System/Library/Fonts/HelveticaNeue.ttc 10 > helveticaneuedeskinterfaceextrabold helveticaneuedeskinterface helveticaneuedeskinterfaceheavy /System/Library/Fonts/HelveticaNeueDeskInterface.ttc 10 > helveticaneuedeskinterfaceextralight helveticaneuedeskinterface helveticaneuedeskinterfacethin /System/Library/Fonts/HelveticaNeueDeskInterface.ttc 8 > helveticaneuedeskinterfacemedium helveticaneuedeskinterface helveticaneuedeskinterfacemediump4 /System/Library/Fonts/HelveticaNeueDeskInterface.ttc 5 > helveticaneuedeskinterfacemediumitalic helveticaneuedeskinterface helveticaneuedeskinterfacemediumitalicp4 /System/Library/Fonts/HelveticaNeueDeskInterface.ttc 6 > helveticaneuedeskinterfacenormal helveticaneuedeskinterface helveticaneuedeskinterfaceregular /System/Library/Fonts/HelveticaNeueDeskInterface.ttc 1 > helveticaneueextralight helveticaneue helveticaneuethin /System/Library/Fonts/HelveticaNeue.ttc 13 > helveticaneuenormal helveticaneue helveticaneue /System/Library/Fonts/HelveticaNeue.ttc 1 > helveticaneueregular helveticaneue helveticaneue /System/Library/Fonts/HelveticaNeue.ttc 1 > helveticanormal helvetica helvetica /System/Library/Fonts/Helvetica.ttc 1 > helveticaregular helvetica helvetica /System/Library/Fonts/Helvetica.ttc 1 > > > >> As a consequence, when I run >> >> \definefontfamily[mainface][ss][Helvetica Neue] >> \setupbodyfont[mainface, 12pt, sans] >> >> I get the light oblique variant instead of the regular one. I have tried to fix this via a >> >> \definefontfamily[mainface][ss][Helvetica Neue][tf=style:normal] > > tf=helveticaregular > should work. > > > Best, Hraban > >

2 2

Using \doifelse to change the itemization symbol in specific cases
by Kevin Kenan 25 Apr '20

25 Apr '20

I’m trying to create a conditional that changes the symbol used for certain item numbers. Here’s my code that doesn’t work. \define[1]\StepsCommand{\doifelse{#1}{2}{k}{#1}\ignorespaces} \defineitemgroup[Steps] \setupitemgroup[Steps][each][n,packed] \setupitemgroup[Steps][each][left=\StepsCommand] \starttext \startSteps \item A \item B % the item number should be replaced with 'k' \item C \stopSteps \stoptext The second item “B” should have the letter ‘k’ instead of the number “2.” Is this possible? Thanks, -kk

2 4