- ntg-context - NTG Mailing Lists

check installed program using Lua
by Pablo Rodriguez 02 Nov '21

02 Nov '21

Dear list, is there a way if a program is installed on the computer using Lua. I have the following sample: \starttext \startluacode filename = tex.jobname .. ".pdf" os.exec("dir " .. filename) \stopluacode \stoptext Is there a way to wrap os.exec() in the sample so that it only runs if "dir" is available? I know that os.name would be an option here, but not in my real world document. Many thanks for your help, Pablo -- http://www.ousia.tk

3 4

new upload
by Hans Hagen 02 Nov '21

02 Nov '21

Hi, I uploaded a new lmtx. There is a new low level manual (lowlevel-localboxes) that describes the extended local boxes interface as well as a layer on top. New is that we have classes which makes this (engine) mechanism more useable. It definitely is some low level mechanism but you can for instance use if for high performance efficient maginal notes and line numbering in controlled situations. (There is some room fro optimization in the engien but I deal with that later.) 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 -----------------------------------------------------------------

3 4

Cannot access Lua function
by kauśika cittūr 01 Nov '21

01 Nov '21

Dear list, In one of my documents I have this lua block – \startluacode local metadata = lpdf.getmetadata(); ⋮ \stopluacode to modify some metadata fields. The file compiles with the latest LMTX (2021.10.28) without any errors but no PDF is produced. In the log, the following line is present : backend > blocked > function 'getmetadata' is not accessible I recall that this was working just fine some days back. Please advise. Thanks, kauśika

3 7

SyncTeX setup
by Henning Hraban Ramm 01 Nov '21

01 Nov '21

Hi, has anyone a working SyncTeX setup with ConTeXt LMTX for anything else than TeXshop? With LaTeX, it works for me with Skim (+ Atom/Textadept/TeXworks) and TeXstudio. And in TeXshop we got our own parser. But can we expect others to adapt to our “superior” SyncTeX format? (I don’t understand the format yet and can’t tell what’s the difference.) Hraban

2 1

\externafigure and \at do not work together
by a badin 01 Nov '21

01 Nov '21

Hello, perhaps this is how it is supposed to work, but: \externalfigure[clock.jpg][clock] \starttext \at[clock] refers to a clock \stoptext produces "?? refers to a clock", instead of "1 refers a clock" --- the wiki of \at says, that labels of figures can be refered to. Is this a bug, or can only some labels be refered to? I was able to fix it using: \reference[clock]{\externalfigure[clock.jpg]} \starttext \at[clock] refers to a clock \stoptext and \externalfigure[clock][] \starttext \at \stoptext produces an interesting artwork... so I thought maybe this could be a bug. LMTX 2021.10.21 Best, Andrej

4 4

Bloody PDF boxes and cr*p marks!
by Henning Hraban Ramm 01 Nov '21

01 Nov '21

Comrades! There must have been a time when PDF boxes (i.e. CropBox, TrimBox, BleedBox) worked as they should, but it must have been a short time. The wiki page https://wiki.contextgarden.net/PDF_Boxes is mostly a documentation of bugs and workarounds, but even that is not entirely correct (any more?). I gave up on that subject and still fix the boxes with Acrobat Pro whenever I need them to be correct. But that’s no solution. (And it’s probably also my fault, since I was one of those who wanted this implemented and then didn’t check enough if it does what it should.) Ok, calm down, try again... * MediaBox is supposed to be the biggest, i.e. print sheet. * CropBox is the viewable area – usually same as MediaBox or TrimBox makes sense. * BleedBox must be TrimBox plus bleed. * ArtBox is neglectable for whole layouts (same as TrimBox might make sense), just interesting in \externalfigure. According to PDF Standard, MediaBox is the only required one; but CropBox is the most important one, since PDF viewers rely on it. With \setuppaper[A4] all boxes can be the same (or undefined). Status quo. With \setuppaper[A4][A4,oversized] I’d expect ConTeXt to set... * MediaBox and CropBox to the bigger format * TrimBox to the page format (A4) * BleedBox also to the page format, while no bleed is defined Status quo: no additional boxes defined. Beware, \showframe draws a green box at the paper size (supposed TrimBox), this is misleading in Acrobat Reader & Pro, where the TrimBox is shown in green (if this option is activated)! Bleed is to be defined in \setuplayout: \showframe \setuppapersize[A4][A4,oversized] \setuplayout[ location=middle, bleedoffset=-5mm, marking=on, ] %\setupinteractionscreen[width=max,height=max] \starttext X \stoptext Status quo: * works only with \setupinteractionscreen * bleedoffset must be negative * BleedBox is set, but not TrimBox (TB > BB is a nesting error) Beware, \setupbleeding refers only to stretching of images (\externalfigure)! This gives more or less correct results: \setuplayout[ location=middle, trimoffset=-7.5mm, bleedoffset=3mm, marking=on, ] * still works only with \setupinteractionscreen * negative trimoffset defines the TrimBox (actual page size), value is subtracted from sheet size. * positive bleedoffset (internally added to trimoffset) * to define trimoffset, I must know the difference between paper size and sheet size ("oversized" adds 15 mm). I.e. ConTeXt “thinks” from the sheet size instead from the (IMO more natural) paper size. For me, the more natural setup would be: \setuplayout[ trimoffset=0mm, % from the paper size outward bleedoffset=3mm, % same, independent of trimoffset ] % ... and no dependency on interactionscreen! So far, I ignored cropoffset. The parameter in itself works as I expected it to: cropoffset=3mm decreases the visible area by 3mm on each side. But it also moves the other boxes by that value! IMO that makes no sense, CropBox can be completely independent of the other boxes! It would be logical to use negative values here, since the size decreases, in opposite to my suggested handling of the other parameters. Yes, that would mean a backward incompatible change. But how many of us rely on the strange current behavior? Please let us fix this! In some cases it makes sense to have different values on e.g. inner and outer edges; I feat that would add a plethora of additional parameters (leftbleedoffset, innerbleedoffset...). Maybe give a list like in CSS? bleedoffset={3mm,3mm,3mm,0mm} – top, right, bottom, left order implied; not very “ConTeXt style”. Hraban

4 9

PDF text stream with delimitedtext
by Adam Reviczky 30 Oct '21

30 Oct '21

Hi, I am trying to use quotation or blockquote in a document, but when selecting and copying the text in poppler/pdf.js I get double marks. Looking at the minimal example below: \nopdfcompression \setuppagenumbering[location=] \starttext \startquotation Hello world! \stopquotation \setupquotation[left=«,right=»] \startquotation Hello world! \stopquotation \setupquotation[left=‘,right=’] \startquotation Hello world! \stopquotation \setupdelimitedtext[blockquote][left=‘,right=’] \startblockquote Hello world! \stopblockquote “Hello world!”\\ «Hello world!»\\ ‘Hello world!’\\ \startquote Hello world! \stopquote\\ \quote{Hello world!}\\ \stoptext The text stream seems to have an additional object for the quotation and blockquote lines. Rendering the PDF in itself is fine but when copying the text I get mixed results (poppler/evince and pdf.js give double marks but mupdf does not: “Hello world!””,«Hello world!»»,‘Hello world!’’,‘Hello world!’’). Is this just an issue with poppler/pdf.js when trying to extract the text (as the PDF rendering seems all fine)? Adam

2 1

getting number from context.figurewidth
by Pablo Rodriguez 28 Oct '21

28 Oct '21

Dear list, I have the following sample: \starttext \startluacode document.sample = "xml-mkiv.pdf" context.getfiguredimensions({ document.sample }) context(context.figureheight) context.par() context(tex.dimen.paperheight) \stopluacode \stoptext Is there a way that I can get a number from context.figureheight, such as I get it from tex.dimen.paperheight? Many thanks for your help, Pablo -- http://www.ousia.tk

3 4

XML-processing: Error with hyphenatedurl
by denis.maier＠unibe.ch 27 Oct '21

27 Oct '21

Hi, I've receiving errors that seem to be related to hyphenatedurl. The problem is I cannot really reproduce it in a MWE. The example below works, but if I include the link in my real document, I'am getting this error message (copied from -error.log): %%%%%%%%%%%%%%%%%%%%% return { ["errortype"]=1, ["filename"]="", ["lastcontext"]="\n<to be read again> \n \n \\xdef<file://xdef> \n<macro> \\expanded<file://expanded> \n #1->\\xdef<file://xdef> \n \\m_syst_helpers_expanded<file://m_syst_helpers_expanded> {\\noexpand #1}\\m_syst_helpers_expanded \n<macro> \\22>:xml:ext-link<file://22%3e:xml:ext-link> \n #1->\\begingroup<file://begingroup> \\expandUx<file://expandUx> \\expanded<file://expanded> \n {\\goto {\\hyphenatedurl {\\xmlflush {#1}}}[url(\\xmlflush<file://xmlflush> {#1})]}\\endgroup \n<lua output> \n<macro> \\22>:xml:p<file://22%3e:xml:p> \n #1->\\begingroup<file://begingroup> \\xmlfilter<file://xmlfilter> {#1}{.[@xml:lang]/command(xml:language)}\\xmlflush {#1}\n \\endgroup<file://endgroup> \\par<file://par> \n<lua output> \n ...\n\n<initial> \n \n\n", ["lastluaerror"]="", ["lasttexerror"]="Missing number, treated as zero", ["lasttexhelp"]="A number should have been here; I inserted '0'. (If you can't figure out why I\nneeded to see a number, look up 'weird error' in the index to The TeXbook.)", ["linenumber"]=9, ["offset"]=10, ["skiplinenumber"]=5, } %%%%%%%%%%%%%%%%%%%% Two things seemed to make it work: 1. Remove the link. 2. Comment out \expandUx in this setup: %%%%%%%%%%%%%%%%%%%% \startxmlsetups xml:ext-link \begingroup \expandUx % commenting this out makes it work \expanded{\goto{\hyphenatedurl{\xmlflush{#1}}}[url(\xmlflush{#1})]} \endgroup \stopxmlsetups %%%%%%%%%%%%%%%%%%%% Best, Denis %%%%%%%%%%%%%%%%%%%%%%%%%%%% \setupinteraction[state=start] \startxmlsetups xml:test \xmlsetsetup{#1}{*}{-} \xmlsetsetup{#1}{article|ext-link}{xml:*} \stopxmlsetups \xmlregistersetup{xml:test} \startxmlsetups xml:article \starttext \xmlflush{#1} \stoptext \stopxmlsetups \startxmlsetups xml:ext-link \begingroup \expandUx \expanded{\goto{\hyphenatedurl{\xmlflush{#1}}}[url(\xmlflush{#1})]} \endgroup \stopxmlsetups \startbuffer[test] <?xml version="1.0" encoding="utf-8" ?> <article> asdf <ext-link ext-link-type="uri" xlink:href="asdf">asdf</ext-link> </article> \stopbuffer \xmlprocessbuffer{test}{test}{} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

3 5

Line breaks, \wordright and hyphenation
by kauśika cittūr 26 Oct '21

26 Oct '21

Dear list, In short, my question is: why does \wordright cause a line-break when the line preceding it breaks with hyphenation and does not otherwise? How to avoid it (if at all possible)? I have this delimitedtext instance called 'amnata' defined as below: \definemeasure[amnatamargin][3cm] \definedelimitedtext [amnata] \setupdelimitedtext [amnata] [ left=, right=, leftmargin={\measure{amnatamargin}}, rightmargin={\measure{amnatamargin}}, ] I am using this to quote paragraph(s) of other authors. Since, I also want to indicate where the quote is from, I use this as follows: \startamnata ⋮ \wordright{author} \stopamnata so that the author's name appears at the right-edge of the block. When the penultimate line inside the block breaks without hyphenation, then argument of \wordright is typeset as expected (i.e. in the same line). On the other hand, when the penultimate line inside this block breaks with hyphenation, the argument of \wordright is pushed to the next line even when there is enough space for it on the same line. Here is a sample illustrating this: \startamnata A quote from another author : Suppose that thereisalongwordhere. \wordright{– author} \stopamnata \startamnata A quote from another author : Here, the text does not cause hyphenation in the first line. \wordright{– author} \stopamnata [I have attached the output here as an image] How do I ensure that this does not happen, if that is possible at all? I suppose this is expected behaviour but I am not able to understand why. Thanks, kauśika

1 3