- ntg-context - NTG Mailing Lists

Problem with dataset not being populated
by Bruce Horrocks 14 Nov '23

14 Nov '23

I produce a small magazine where the source files are organised into a product and component structure. For each issue, I have a main file with \startproduct <issue no> and then, inside that, several \component[filename] statements, one per article. Each article starts in a standard way, including adding some information to a dataset. At the end of the issue, the dataset information is retrieved to be printed and this is where strange things start to happen. Sometimes the whole issue will compile without problems and the dataset information is retrieved perfectly. I can make edits to both the product file and/or any of the component files and it all recompiles to produce a PDF each time. But at othertimes, something "breaks" and the dataset can't be retrieved or isn't created. Everything else compiles fine but there is simply no data being retrieved from the dataset so nothing to print. If I go back to the product file and comment out most of the components then it will (99% of the time) 'unbreak' itself and the dataset gets created. Once it has started working again I can then uncomment the rest of the components and it will recompile with all the dataset data present ... until the next time. Has anyone seen anything like this? I'm using Lua to retrieve the dataset info and, when it goes wrong, the dataset returned by Lua is a nil value. But the .tua file (not a .tuc) has the full dataset structure included. I can't produce an MWE as I have no idea what to remove and include - it doesn't seem to matter which components are left or removed in order to fix the problem just as long as several ar removed. In other words, it's not one component file causing the problem. And sometimes commenting out just two components will fix it, othertimes I have to do most of them. It could be a caching issue - when it goes wrong, reprocessing the product file causes much less to scroll past on the TeXShop log display. And when it is fixed, the first successful run has a much longer log, suggesting that more is being processed. So, unless someone else has seen the above and has the solution, the first question is: does LMTX cache the result of previous runs to save reprocessing if nothing has changed? If so, how can I force a full reprocess or remove those cache files? I'm using TeXShop to edit and compile, but running "context product_filename.tex" from the command line makes no difference. Version is 2023.09.26 18:19. — Bruce Horrocks Hampshire, UK

1 0

\doifemptyelse issue?
by Bruce Horrocks 14 Nov '23

14 Nov '23

The following MWE gives "false false" instead of "false true". I'm sure this used to work (18 months or more ago!) as I found it when re-processing an old document. ---start--- \setdataset [DS] [a=aaaa] \setdataset [DS] [a=] \starttext :\datasetvariable{DS}{1}{a}: \crlf :\datasetvariable{DS}{2}{a}: \crlf \doifemptyelse{\datasetvariable{DS}{1}{a}}{true}{false} \doifemptyelse{\datasetvariable{DS}{2}{a}}{true}{false} \stoptext ---end--- ConTeXt ver: 2023.09.26 18:19 LMTX There is a work-around using \doiftext instead as that seems to work okay with recalled dataset values but I thought I'd raise it. — Bruce Horrocks Hampshire, UK

3 3

named destinations in /GoToR
by Pablo Rodriguez 14 Nov '23

14 Nov '23

Dear list, imagine I have a PDF document (named "whatever.pdf") such as the one generated from this source: \setupinteraction[state=start, focus=standard] \starttext \dorecurse{5} {\chapter[\recurselevel]{Chapter} \section[sec-\recurselevel]{Section}} \stoptext This source only samples PDF documents not generated with ConTeXt. This means that no .tuc file may be available and named destinations may contain only digits. Then I need to access some named destinations from a document such as: \setupinteraction[state=start, focus=standard] \starttext \dorecurse{5} {\goto{Chapter \recurselevel}[whatever.pdf::\recurselevel], \goto{section \recurselevel}[whatever.pdf::sec-\recurselevel]\par} \stoptext Sorry, but after reading again lpdf-ano.mkxl I’m not sure whether I can get named destinations (such as "/D (4)" or "/D (sec-3)") and how I could get them. With the previous code, the only destination reads in the PDF source: 5 0 obj << /D [ 0 /Fit ] /F (whatever.pdf) /S /GoToR >> endobj Many thanks for your help, Pablo

2 2

hard clipping in ConTeXt?
by Pablo Rodriguez 13 Nov '23

13 Nov '23

Dear list, I have the following sample: \setuplayout[page] \starttext \clip[hoffset=5mm, voffset=5mm, width=15mm, height=15mm] {\externalfigure[xml-mkiv.pdf]} \stopnotmode \stoptext Using PDF.js (from current Firefox), xpdf, Evince, mupdf-gl, the viewer is able to find "dealing", altough it cannot be displayed. Okular and Sumatra (with Wine in Linux) are not able to find "dealing". Is there any way to have a hard clipping in ConTeXt? I mean, that only the contents of the clipped section is added to the resulting PDF document? Many thanks for your help, Pablo

2 2

placing \externalfigure
by Hans van der Meer 12 Nov '23

12 Nov '23

In placing \externalfigure[file] I observe the figure is placed with its bottom on the baseline, thus: [externalfigure] --------------------------- baseline I looked into the option list of setupexternalfigure but could not find one whereby the location with respect to the baseline is either top, bottom or center. Such an option might not exist and then I must program around it. But I prefer to use something that is already there and fits my needs. yours sincerely dr. Hans van der Meer

3 2

How can I use \naturalnumbers
by Ursula Hermann 12 Nov '23

12 Nov '23

Dear List, I would like to use the \integers. I would like to use the commands in Text and also in Formulars. How to write in the right way? I have: \starttext \naturalnumbers {N} stoptext Many thanka Uschi

2 1

XRechnung
by juh+ntg-context＠mailbox.org 10 Nov '23

10 Nov '23

Hi, does anyone produce the XML-based invoice called XRechnung with ConTeXt? I don't know if XRechnung is the european name or only used in Germany? juh

3 3

\xmlinclusion?
by Hans van der Meer 09 Nov '23

09 Nov '23

I cannot get \xmlinclusion working. Below an MWE with its output. What I am missing here or is \xmlinclusion not ok? \startxmlsetups demo \xmlsetsetup{#1}{ root|test }{demo:*} \stopxmlsetups \xmlregisterdocumentsetup{demo}{demo} \startxmlsetups demo:root \blank{\bf Start test}\crlf \string\xmlinclusion\{\#1\} ===> "\xmlinclusion{#1}"\crlf \xmlflush{#1}\crlf {\bf End test}\blank \stopxmlsetups \startxmlsetups demo:test \xmlflush{#1} \stopxmlsetups \starttext \currentdate-\currenttime\quad\ConTeXt-version=\contextversion\blank The following text is on page 52 of xml-mkiv.pdf:\crlf \type{\xmlinclusion{...}}\crlf returns the file where the node comes from\blank Follows the test of this claim: \xmlprocessfile{demo}{demo.xml}{}\blank \typefile{demo.xml} \stoptext <?xml version="1.0" encoding="UTF-8"?> <root> <test>node TEST</test> </root> met vriendelijke groet, dr. Hans van der Meer Burgemeester Rijnderslaan 244 ingang D 1185 MC Amstelveen tel. 020 6452701 / 06 53743629

1 0

how can I do this in CLD
by Henning Hraban Ramm 08 Nov '23

08 Nov '23

Hi, I’m generating some code in a loop and would like to use formatting like \quotation or \emph in a caption: context.startplacefigure({title = "Style " .. context.quotation(style)}) That doesn’t work. I guess context.quotation is executed immediately and doesn’t return anything? How can I do this? Hraban

1 1

Wiki problem - labels missing from the main diagram on the Page Layout page
by Bruce Horrocks 08 Nov '23

08 Nov '23

When I go to the page layout page on the Contextgarden wiki <https://wiki.contextgarden.net/Page_Layout> the "typesetting areas" diagram is missing its text labels. If I click on the diagram the Wiki takes me to the underlying layout.svg file and then I /do/ see the labels. This happens in both Safari and Firefox. Any suggestions as to what needs to be changed to get the diagram to display correctly? — Bruce Horrocks Hampshire, UK

5 6