October 2026 - ntg-context

Post update binary error
by Kirk Lowery 16 Oct '26

16 Oct '26

I updated to the latest and greatest just now and got the following responses: irk@erebor:.../local/context/bin$ mtxrun --version lua error : function call: /usr/local/context/tex/texmf-linux-64/bin/mtxrun.lmt:1738: attempt to index a nil value (global 'sequencer') stack traceback: /usr/local/context/tex/texmf-linux-64/bin/mtxrun.lmt:1738: in main chunk kirk@erebor:.../local/context/bin$ context --version lua error : function call: /usr/local/context/tex/texmf-linux-64/bin/mtxrun.lmt:1738: attempt to index a nil value (global 'sequencer') stack traceback: /usr/local/context/tex/texmf-linux-64/bin/mtxrun.lmt:1738: in main chunk [C]: in global 'dofile' /usr/local/context/tex/texmf-linux-64/bin/context.lmt:38: in main chunk I'm assume this is a problem local to me. Suggestions on how to recover? TIA! Kirk

3 4

update
by Hans Hagen 16 Oct '26

16 Oct '26

Hi, I pushed an update but before one immediately picks up that one, best maybe wait till those who normally do some tests are okay with it (or just make a copy of the old installation). The reason is the following: There was this report on the 1 sp difference between platforms (cpu specific) and although i don't loose sleep overt his esp not if it is the backend, i looked into it anyway. As a consequence the serialization of floats has been adapted. I don't know if that makes a difference but it had the side effect that i also decided to update serialization in general. The native lua format routing is not that fast so we aleasys had our own (in lua) and that has been updated. It also removes an intermediate call plus comes with bit performance gain. That mostly affects the backend, which happens to be kind of slow (but flexible). This means that (on my machine) the test suite gains 5-10% runtime, and documents (manuals) some 5% too. But ... for this i had to adapt quite some lua files so errors can have crept in (those are then mostly typos) .. easy to fix once known. I still have some 500 lines (over a few dozen files) to adapt but want an intermediate test. The 2200 file test suite compiles ok (no visual checking done as this is mostly crash testing). The previous upload already had the new (more efficient) http client code as used in the installed and the current has the lua socket library removed and replaced by a lightweight http server feature (as used in e.g. context help system), so less code and less dependencies. One can still compile with an option because I still need to adapt some private - pet project - servlet code here. In addition to all this .. the usual fixes etc as discussed in this list, 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 6

Cron <context@cgl> /var/www/aanhet.net/context/bin/cron/context-mirror
by root＠ntg.nl 16 Oct '26

16 Oct '26

receiving incremental file list ./ logo-ade.png logo-cts.png logo-pod.png 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 sent 245,230 bytes received 71,617,811 bytes 6,844,099.14 bytes/sec total size is 920,819,570 speedup is 12.81 Running archiver: New dir: /var/www/aanhet.net/context//htdocs/archives/context-2026-10-16.18 526629236 /var/www/aanhet.net/context//htdocs/archives/context-2026-10-16.18/latest 126741221 /var/www/aanhet.net/context//htdocs/archives/context-2026-10-16.18/current 653370457 /var/www/aanhet.net/context//htdocs/archives/context-2026-10-16.18 653370457 total

1 0

Cron <context@cgl> /var/www/aanhet.net/context/bin/cron/context-mirror
by root＠ntg.nl 16 Oct '26

16 Oct '26

receiving incremental file list ./ logo-ade.png logo-cts.png logo-pod.png 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 documents/general/manuals/mkxl/ documents/general/manuals/mkxl/luametafun.pdf documents/general/manuals/mkxl/luametatex.pdf sent 277,664 bytes received 104,294,666 bytes 8,365,786.40 bytes/sec total size is 920,819,804 speedup is 8.81 Running archiver: New dir: /var/www/aanhet.net/context//htdocs/archives/context-2026-10-16.12 526630529 /var/www/aanhet.net/context//htdocs/archives/context-2026-10-16.12/latest 126741221 /var/www/aanhet.net/context//htdocs/archives/context-2026-10-16.12/current 653371750 /var/www/aanhet.net/context//htdocs/archives/context-2026-10-16.12 653371750 total

1 0

regular online meet-up
by Henning Hraban Ramm 14 Oct '26

14 Oct '26

You’re invited to our regular online meet-up, this upcoming Wednesday, October 14th, 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 1

no joy as yet on RGB issue
by jbf 13 Oct '26

13 Oct '26

I asked the printshop to run a diagnostic on the file prepress file which produced an error message in Acrobat Pro. Note the details below, which also include the version of ConTeXt I used and the version of Acrobat Pro used by prepress. Preflight Summary Report for: prd_Foundations-2_12-10-26.pdf Profile: Verify compliance with PDF/X-1a (Processed pages 1 to 260) Processed by Pre Press, Date: 13/10/2026 6:18 pm Results (Summary) Error ICC profile missing in PDF/X OutputIntent (RegistryName not www.color.org) ICC profile missing in PDF/X OutputIntent (no RegistryName entry) Invalid color space (520 matches on 260 pages) - 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33... Object uses RGB (1045 matches on 260 pages) - 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33... PDF/X entry is not “PDF/X-1a:2003” Trapped key not true or false Uses device independent color (60 matches on 1 page) - 5 Document information File name: "prd_Foundations-2_12-10-26.pdf" Path: "/Users/prepress/Desktop/2026 COVENTRY titles/ WORK IN PROGRESS/Foundations book 2/dont use" PDF version number: "1.3" File size (MB): 1.4 Title: "Foundations, Book Two" Creator: "LuaMetaTeX 2.11.10 20260922 + ConTeXt LMTX 2026.09.22 16:20" Producer: "LuaMetaTeX" Created: "12/10/2026 1:53 pm" Modified: "13/10/2026 6:09 pm" Trapping: "Unknown" PDF/X Version: "PDF/X-1a:2001" Number of plates: 4 Names of plates: "(Cyan) (Magenta) (Yellow) (Black) " Environment Preflight, 18.6.0 (276) Acrobat version: 26.002 Operating system: macOS 15.7.3 ----------- Compare this with the relevant information in the log file for this item. backend > profiles > setting format 'PDF/X-1a:2001' to 'PDF/X-1a:2001' backend > profiles > using color intent 'Coated FOGRA39 (ISO 12647-2:2004)' backend > profiles > using icc profiles 'Coated FOGRA39 (ISO 12647-2:2004)' backend > profiles > forcing pdf version 1.3, compression level 3, object compression disabled backend > profiles > permitted colorspaces: rgb 'yes', cmyk 'yes' colors > defining > supported models: gray 'true', rgb 'true', cmyk 'true', spot 'true' transparencies > support > transparency is supported viewerlayers > viewerlayers are supported backend > xmp > using file '/home/julian/Data/context/tex/texmf-context/tex/context/base/mkiv/lpdf-pdx.xml' backend > profiles > profile specification 'Coated FOGRA39 (ISO 12647-2:2004)' loaded from '/home/julian/Data/context/tex/texmf-context/colors/icc/context/colorprofiles.xml' backend > profiles > setting 'CoatedFOGRA39.icc' as default 'cmyk' color space backend > profiles > setting output intent to 'Coated FOGRA39 (ISO 12647-2:2004)' with id 'FOGRA39' for entry 1 ---------- Apart from a possible discrepancy between PDF/X-1a:2001 and 2003, which I think I can easily adjust. two other matters leave me confused. (1) the Preflight error message seems to suggest a missing profile, while the ConTeXt log clearly indicates a profile (2) I obviously need the log to show rgb 'false' instead of 'true' (or 'no' instead of 'yes'). As Hraban suggested, I tried \setupcolors[rgb=no] but that made no difference. I ran the pdf through the Preflighter.app, also as he suggested, and among other things it told me: [✓] Output intent: FOGRA39 [✓] Color space: contains print colors So it too is complaining. I tried a different output intent (SWOP2006_Coated3v2.icc) to see if that made any difference. it did not. BUT if I compile the same file with an early 2025 version of ConTeXt prepress checks all pass muster and there is no problem. Is anybody able to shed any light on this for me? And in particular, is there a way I can exclude the rgb problem which seems to be the major issue? Julian

1 0

Acrobat Pro issue
by jbf 13 Oct '26

13 Oct '26

I just need to check if anyone else has had the small issue I have recently had. Using the more or less latest 2026 ConTeXt version (I usually compile with an older 2025 version), the print shop recipient told me that when she went to open the pdf she received a message 'could not get nr of components'. Then making a second attempt to open it, the file did open. I recompiled the pdf in my 2025 version and she had no problems. Checks I have made on the integrity of the '2026' pdf (I don't have Acrobat at all, I am on Linux and use other, e.g. qpdf which I used to do the checking) show no issues with it. My suspicion is that the 'nr of components' is in reference to the ICC profile, so the issue is with how Acrobat Pro is interpreting things, not with the ConTeXt product itself. Having said that, I know next to nothing about how PDF programs like Acrobat work, so I am merely guessing. But I am wondering... is there something in how I set up the backend that might help overcome this issue if I wish to compile with the latest ConTeXt rather than the older one? Currently I simply have: \setupbackend [format=PDF/X-1a:2001, intent={Coated FOGRA39 (ISO 12647-2:2004)}] Julian

2 2

adding to my understanding of the Acrobat Pro issue
by jbf 13 Oct '26

13 Oct '26

I can sort of see where the problem lies, but am not so sure what to do to resolve it. I suspect it will be in how I set up the backend: Currently I have \setupbackend [format=PDF/X-1a:2001, intent={Coated FOGRA39 (ISO 12647-2:2004)}] If I compile my file with the older early 2025 version of ConTeXt (which I normally use without problems), the log file tells me: ... backend > profiles > setting format to 'PDF/X-1a:2001' backend > profiles > forcing pdf version 1.3, compression level 3, object compression disabled backend > profiles > permitted colorspaces: rgb 'no', cmyk 'yes' colors > defining > supported models: gray 'true', rgb 'false', cmyk 'true', spot 'true' transparencies > support > transparency is not supported viewerlayers > viewerlayers are not supported backend > xmp > using file '/home/julian/bin/context/tex/texmf-context/tex/context/base/mkiv/lpdf-pdx.xml' backend > profiles > profile specification 'Coated FOGRA39 (ISO 12647-2:2004)' loaded from '/home/julian/bin/context/tex/texmf-context/colors/icc/context/colorprofiles.xml' backend > profiles > omitting reference to profile for intent 'Coated FOGRA39 (ISO 12647-2:2004)' ... If I compile with the more recent 2026 version of ConTeXt, the log file tells me: ... backend > profiles > setting format 'PDF/X-1a:2001' to 'PDF/X-1a:2001' backend > profiles > using color intent 'Coated FOGRA39 (ISO 12647-2:2004)' backend > profiles > using icc profiles 'Coated FOGRA39 (ISO 12647-2:2004)' backend > profiles > forcing pdf version 1.3, compression level 3, object compression disabled backend > profiles > permitted colorspaces: rgb 'yes', cmyk 'yes' colors > defining > supported models: gray 'true', rgb 'true', cmyk 'true', spot 'true' transparencies > support > transparency is supported viewerlayers > viewerlayers are supported backend > xmp > using file '/home/julian/Data/context/tex/texmf-context/tex/context/base/mkiv/lpdf-pdx.xml' backend > profiles > profile specification 'Coated FOGRA39 (ISO 12647-2:2004)' loaded from '/home/julian/Data/context/tex/texmf-context/colors/icc/context/colorprofiles.xml' backend > profiles > setting 'CoatedFOGRA39.icc' as default 'cmyk' color space backend > profiles > setting output intent to 'Coated FOGRA39 (ISO 12647-2:2004)' with id 'FOGRA39' for entry 1 ... There are two obvious differences: the older version does not permit rbg while the later one does; the older version omits reference to the profile for intent, while the later version sets output to the FOGRA39 intent. It seems to me that permitting rgb is the problem. How do I prevent that while still using the same profile/intent (ie fogra39 etc.) And I am wondering why it is that the more recent version of ConTeXt treats my \setupbackend differently to how the the earlier one does, by allowing rgb. Perhaps I now need to specifically stop it from doing this. Julian

2 1

New columnsets / subcolumnsets: alignment of Greek–French parallel text
by Jean-Pierre Delange 12 Oct '26

12 Oct '26

Dear all, I am preparing a scholarly bilingual edition of Aristotle's /Nicomachean Ethics/(Greek and French) using ConTeXt LMTX (2026-10-01). I have been experimenting with the new columnsets/subcolumnsets mechanism and would appreciate advice on three related issues. My layout uses a 180 × 260 mm page, with two asymmetrical columns (58 mm for Greek, 86 mm for French, separated by 6 mm). The French translation is substantially longer because it includes Greek terminology and explanatory material. 1. Synchronization I currently use |\flushsubcolumnsets|once per chapter. Flushing at every Bekker reference produces excessive whitespace. Is there a way to synchronize corresponding passages more gently, without forcing paragraph breaks or large vertical gaps? Would |\flushsubcolumnsets[spread]|be appropriate? 2. Bekker references in running headers Bekker references are the standard scholarly reference system for locating passages in Aristotle's works, just as Stephanus references are used for Plato. In a bilingual scholarly edition, these references must remain inline, sometimes in the middle of a sentence, while the running headers should indicate the range of passages appearing on each page. My current marking mechanism sometimes produces reversed ranges (e.g. |1096b25–1096b20|), apparently because the two subcolumns contribute marks in processing order rather than textual order. Is there a recommended way to obtain the correct logical Bekker range in the running header? (To Jim : Aristotle may have taught us that everything aims at some good, but persuading Greek and French to aim at the same bottom of the page is proving rather difficult!) 3. Scholarly footnotes I eventually need two independent note streams: textual/philological notes and translation notes. What would be the most reliable approach with unequal subcolumns? I have followed earlier discussions by Hans Hagen and Pablo Rodríguez concerning the new columnsets and parallel-text synchronization. I would be particularly interested in any subsequent improvements or recommended practices. I have attached a ZIP archive containing a standalone |.tex|MWE. It reproduces the relevant layout without requiring my project files. The sample text was reconstructed from a proof PDF and is intended to demonstrate layout behaviour, not to serve as an authoritative transcription. Many thanks for your advice. Best//JP

1 1

Cannot insert annotation(highlight, etc) of annotated pdf through \externalfigure.
by scerac＠mail.ustc.edu.cn 11 Oct '26

11 Oct '26

Prepare a tmp.tex. `tmp.tex` ``` \starttext tmp \stoptext ``` after that, I annotated it with Firefox browser. Use \externalfigure, `test.tex` ``` \starttext \externalfigure[tmp.pdf][width=\textwidth,frame=on,interaction=yes] \stoptext ``` And no annotation was rendered in test.pdf, which left me confounded.

2 4