September 2026 - ntg-context

Cron <context@cgl> /var/www/aanhet.net/context/bin/cron/context-mirror
by root＠ntg.nl 08 Sep '26

08 Sep '26

receiving incremental file list ./ ctan.lsr document-2.htm download-1.htm download-2.htm logo-ade.png logo-cts.png logo-pod.png rss.xml show-fil.pdf context/latest/ context/latest/cont-lmt.zip context/latest/cont-mpd.zip context/latest/cont-ppc.zip context/latest/cont-sci.zip context/latest/cont-tmf.zip context/latest/cont-tst.7z context/latest/cont-tst.tar.xz context/latest/cont-tst.zip general/manuals/ general/manuals/luametafun.pdf general/manuals/luametatex.pdf general/manuals/threed.pdf sent 249,498 bytes received 92,142,725 bytes 8,034,106.35 bytes/sec total size is 818,568,700 speedup is 8.86 Running archiver: New dir: /var/www/aanhet.net/context//htdocs/archives/context-2026-09-08.00 411886551 /var/www/aanhet.net/context//htdocs/archives/context-2026-09-08.00/latest 126741221 /var/www/aanhet.net/context//htdocs/archives/context-2026-09-08.00/current 538627772 /var/www/aanhet.net/context//htdocs/archives/context-2026-09-08.00 538627772 total

1 0

manual
by Hans Hagen 08 Sep '26

08 Sep '26

Hi, btw: The metafun threed.pdf manual has been added to the distribution. We need to come up with an addendum covering Keith's fabulous work (his presentation at the meeting was a highlight but is a bit large to put online so we need to figure that out). 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 -----------------------------------------------------------------

1 0

update
by Hans Hagen 08 Sep '26

08 Sep '26

Hi, I pushed an update (hopefully ok because bots were saturating the logs and such so we ran out of disk space). Apart from fixes for issues mentioned / encountered the last weeks we also added a goodie file for lm so that Jim can go forward. I don't know how many users process multiple documents at the same time, but as a side effect of improving performance of processing the test suite (I run 8 parallel processes for it). Here is some fun: context --parallel test-1.tex test-2.tex or with output: context --parallel test-*.tex --terminal a bit more distinct: context --parallel test-*.tex --terminal=ansi if you have more cores: context --parallel test-*.tex --terminal=ansi --runners=16 A tex process needs one core so a rule of thumb is the nofcores multiplied by 2 if hyperthreading is available. On linux one can get 100% saturation, on windows a bit less (the test suite went down from 800 sec for 2200 files to 425 and on linux (wsl) we get to some 260 sec. (A 2018 laptop with 4 xeon cores that hyperthread; I can't test on a faster machine.) Profiled binaries perform some 10% faster. 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 -----------------------------------------------------------------

1 0

regular online meet-up
by Henning Hraban Ramm 07 Sep '26

07 Sep '26

You’re invited to our regular online meet-up, this Wednesday, September 9th, 19:30 CEST (UTC+2) at https://lecture.senfcall.de/hen-rbr-rku-oke (same, but shorter: https://u.mtxrun.eu/ctxmtg) ConTeXt users of all levels are welcome! Looking forward to seeing you, Hraban (Same blurb as always:) [Howto] * No special software installation required; most modern browsers should work (WebRTC required). * Open the URL above, accept the privacy statement, * enter your name, * click "join" (or "start" if you’re the first), * click "with microphone", allow your browser to access it, check the audio. * Your microphone is muted if you join. Activate microphone and/or camera with the buttons at the bottom. * Minimize the default presentation with the "screen" button, bottom right. * If you’d like to share your screen or upload a file, you can make yourself the presenter: Click on your user name, change the setting, then you’ll see the "screen sharing" button beside the camera button; also there’s now "manage presentations" behind the "plus" button. Beware there is only one presenter at a time, so don’t kill someone else’s presentation. [Technical hints] * Sound is usually better if you use a headset (less noise for everyone). * Connection problems are mostly due to low bandwidth or high latency on your side, esp. with mobile connections. * Sometimes leaving and re-entering helps. * If audio/video doesn’t work for you, you can still use the text chat. * Screen sharing needs a lot of bandwidth. * BigBlueButton documentation applies: https://bigbluebutton.org/tutorials/ [Netiquette] * Please use a name that we recognize from here. Some feel uncomfortable with anonymous lurkers. * Mute your microphone while you’re not talking. * It’s nice to show your face at least when you join. * If there are connection problems, stop camera sharing. * The room is set to “everyone’s a moderator“, I trust you...

1 0

Yet Another itemize question
by Jim Diamond 07 Sep '26

07 Sep '26

Hi all, If I have \setupwhitespace[big] ... Labs: \startitemize[nowhite,a] \item black \item chocolate \item yellow \stopitemize then the "Labs:" line is immediately followed by the "black" item, as desired (in this case). But if the list is a sub-list, as in the example below, there is a blank line after "Labs:". I have tried a variety of different options, all to no avail. As a work-around, if I make the outer list "packed" and put a \blank after the last item in each sub-list, I get what I want. But is there a set of options which automagically packs the sub-lists (and the text above them) together but separates the sub-lists by blank lines? Thanks. Jim %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \startdocument \setupwhitespace[big] % Items follow preceding line with no extra vertical space, as desired. Labs: \startitemize[nowhite,a] \item black \item chocolate \item yellow \stopitemize \blank\hrule\blank \startitemize[n] % I could add ,packed % Items *do not* follow preceding line with no extra vertical space \item Labs: \startitemize[nowhite,a] \item black \item chocolate \item yellow \stopitemize % I could use \blank \item Corgis: \startitemize[nowhite,a] \item Pembroke \item Welsh \stopitemize \stopitemize \stopdocument %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

2 2

bug(!?) in \its (?)
by Jim Diamond 06 Sep '26

06 Sep '26

Hi all, In the wiki page for \its https://wiki.contextgarden.net/Command/its we see the example \startitemize [5, % open circles packed] % no blank lines between items (counteracts \setupwhitespace[big]) [width=6em, % width of area in which bullets/numbers/scales are placed distance=2em, % distance between last scale point and item items=4] % number of points in scale % First we print a range key \ran{yes \hss no} % \hss produces an infinitely stretchable (and squashable) space \its I think Lord Peter Wimsey is the bee's knees. \its I think Bunter is the caterpillar's waistcoat. \stopitemize When I run that with the context in TL 2025, I get the output as shown in the wiki page. However, both TL26 and standalone give me different output, in which "I" is overlaid on "think" (or vice versa). No doubt someone familiar with the code will immediately see the problem. Unfortunately, that isn't me. :-) Thanks. Jim

3 3

Missing font glyphs in PDF A
by Lukas/ConTeXt 06 Sep '26

06 Sep '26

Hello, ConTeXt normally generates a "normal" kind of PDF. But it doesn't necessarily involve all font glyphs. Is there an option, e.g.\setupbackend[??], to force ConTeXt to involve all font glyphs into PDF? - The PDF A/3b does this, but A/3b has some disadvantages, so I'd rather use "normal" PDF with all font glyphs involved. How can I do this? (To work with MKIV, if possible.) Thanks in advance. Best regards, Lukas

2 1

Regarding \setupenumerations
by Fabrice Couvreur 04 Sep '26

04 Sep '26

Hi, I would like to get numbering like this : Chapter 1 Section 1.1 Definition 1.1.1 Definition 1.1.2 Definition 1.1.3 Section 1.2 Definition 1.2.1 Definition 1.2.2 Definition 1.2.3 Thanks for your help. Fabrice \starttext \setuphead [section] [style=\bfc, sectionsegments=chapter:section, distance=1em, align={flushleft,broad}, ] \setuphead [subsection] [sectionsegments=section:subsection, distance=1em, style=\bfb, align={flushleft,broad}] \setuphead [subsubsection] [sectionsegments=section:subsubsection, style=\bfa, align={flushleft,broad}] \setupenumerations [before={\blank}, after={\blank}, alternative=serried, distance=0.5\emwidth, title=yes, titleleft=, titleright=, titlestyle=normal, prefix=yes, stretch=0, shrink=0, headcommand=\groupedcommand{}{}, numbercommand=\groupedcommand{}{}, titlecommand=\groupedcommand{(}{)}, width=fit] \defineenumeration [definition] [text=D�finition, style=italic, ] \startchapter [title=Chapter one] \startsection [title=Section one] \dorecurse{3}{ \startdefinition \stopdefinition } \stopsection \startsection [title=Section two] \dorecurse{3}{ \startdefinition \stopdefinition } \stopsection \stopchapter \stoptext

4 10

[LMTX] mtx-install.lua fresh bootstrap breaks on Windows: unzipdir writes zip directory entries as 0-byte files
by Jiri Bělka 04 Sep '26

04 Sep '26

Fix util-zip.lua's unzipdir() writing zip directory-marker entries as 0-byte files, which then blocks mkdir() for the real files that belong underneath them (reproducible on Windows; the same code path happens to no-op harmlessly on POSIX, where opening a path ending in "/" for writing simply fails). Upstream: ConTeXt LMTX, tex/context/base/mkiv/util-zip.lua, function unzipdir - bundled (minified, concatenated with other core modules) into bin/mtxrun.lua inside the small bootstrap archive (context-<platform>.zip) that ships from lmtx.pragma-ade.com. This patch targets that bundled bin/mtxrun.lua copy, applied with `git apply -p1` from the app's install directory before the first `mtx-install.lua --update` bootstrap run. How this was found ------------------- The jirib/scoop-bucket "context" package's own Windows CI (GitHub Actions, windows-latest) failed a fresh `scoop install context` with mtxrun unable to find tex/texmf-context/web2c/contextcnf.lua, and every mtxrun/context script lookup afterwards failing with "unknown script ...". A diagnostic step added to that CI listed the tree after bootstrap: tex/texmf-context had only ~10 entries, and every one of them - including "web2c" - was a 0-byte *file* rather than a directory (`Get-ChildItem` showed `-a----` file-mode entries where directories were expected). To rule out a scoop-specific packaging mistake, the exact same `mtx-install.lua --update` bootstrap was reproduced by hand, fresh, directly against the real upstream context-linux-64.zip / context-win64.zip / texmf-context.zip - on Linux it succeeded on the first try (5723 real files under tex/texmf-context, web2c a real directory); the identical code, same zip, same directory-marker entries, failed only on Windows. Tracing unzipdir()/unzipfile() in the bundled bin/mtxrun.lua explained the split: unzipfile() returns "" for a directory-marker entry (0 compressed bytes), which Lua treats as truthy, so unzipdir() proceeds to mkdirs()+savedata() for it. Writing a file to a path ending in "/" apparently fails silently on POSIX (harmless no-op there) but succeeds on Windows as a literal 0-byte file with that name - which then blocks mkdir() for every real file meant to live underneath it. This patch was applied to bin/mtxrun.lua and verified against a real, fresh `scoop install context` on a GitHub Actions windows-latest runner (jirib/scoop-bucket CI): the bootstrap now completes in ~2 minutes (previously it either corrupted the tree, or - with a detect-and-retry workaround instead of this patch - took ~48 minutes falling back to downloading ~6700 files individually), tex/texmf-context contains the correct 5723 files with web2c as a real directory, and `context --version` / `mtxrun --generate` both work. Reported upstream to ntg-context(a)ntg.nl. --- a/bin/mtxrun.lua +++ b/bin/mtxrun.lua @@ -16453,7 +16453,8 @@ for i=1,count do local l=list[i] local n=l.filename - if not validate or validate(n) then + if n:sub(-1) == "/" then + elseif not validate or validate(n) then local d=unzipfile(z,n) if d then local p=filejoin(path,n)

2 1

ConTeXt floats: grid option behavior and fixing bottom whitespace
by autumnus 04 Sep '26

04 Sep '26

This example(https://tex.stackexchange.com/questions/46327/defining-the-size-of-… no longer compiles as expected. Originally, the top and bottom of the graphic would sit flush against the text, but now there is an extra one or two lines of blank space at the bottom. Original example: https://i.sstatic.net/LzJzx.png It seems I can only get it to work in the example below, but what role does `grid` play here? Once I add `setuplayout[grid=yes]`, things actually get worse: the text overlaps. I don't understand—has the mechanism for floats changed, or am I not using the correct settings to avoid the extra blank space at the bottom? my example: ``` \useMPlibrary [dum] \setupfloats [spaceafter=,] \starttext\showgrid \input ward \startplacefigure [location={left,grid,none}] \externalfigure [dum] [height=8\lineheight] \stopplacefigure \input knuth \stoptext ``` original example: ``` \useMPlibrary [dum] \setupfloats [spacebefore=, spaceafter=] \starttext \input ward \startplacefigure [location={left,none}] \externalfigure [dum] [height=8\lineheight] \stopplacefigure \input knuth \stoptext ```

1 0