- ntg-context - Mailing Lists

upload / metafun
by Hans Hagen 16 Dec '20

16 Dec '20

Hi, I uploaded a new lmtx. This time some work has been done at the metafun end. At the tex end we're working on some overload protection mechanism which involves tagging many macro so that will take a while (most tagging is done but checking left over cases take time). So, I wondered if we can do something similar in metapost: flag macros and variables in order to issue a warnings when a dangerous overload happens. The current lmtx has this as an experiment: primitives, permanent, immutable and mutable (so far); you can look at the *.mpxl files to get an idea. If you say for instance path fullcircle; there will be a warning metapost > overloading immutable 'fullcircle' or when \enabledirectives[metapost.overloadmode=error] has been given you get an error. The implementation is rather simple and non intrusive and differs from the one in tex. There it is tightly integrated for all kind of reasons (millions of macro definitions per run so efficiency matters). Also, in metapost variables (tags) can have complex name constructions and we only test the top level. When a overload happens (overloadmode internal <> 0) a lua callback takes care of it. A side effect is that we can query all hash entries and check for consistency which is what i've done and am doing now. Of course that can introduce issues ... just that you know. 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 -----------------------------------------------------------------

4 4

Issue with math alignments
by Jairo A. del Rio 15 Dec '20

15 Dec '20

Hi, list! The following \starttext \startformula \startmathalignment \NC 0 ; \NC 0,3 ; \NC 0,33 ; \NC 0,333 ; \NC 0,3333 ; \NC \cdots \NR \NC 1 ; \NC 0,4 ; \NC 0,34 ; \NC 0,334 ; \NC 0,3334 ; \NC \cdots \NR \stopmathalignment \stopformula \stoptext stops working both in MkIV and LMTX. The latter includes the following: The file ended when scanning an argument. (text) {0,33 ; \NC 0,333 ; \NC 0,3333 ; \NC \cdots \NR \NC 1 ; \NC 0,4 ; \NC 0,34 ; \NC 0,334 ; \NC 0,3334 ; \NC \cdots \NR \stopmathalignment \stopformula \stoptext It seems ConTeXt is introducing a brace before 0,33. What should I do? Thank you in advance. Jairo :)

1 1

thinrules width
by Rik Kabel 15 Dec '20

15 Dec '20

Hello list, and developers in particular, I note that thinrules (and its setup, setupthinrules, and relatives thinrule and hairline) does not have a width setting, and always sets a rule (or rules) the full width of the text area (less any text set on the same line. No problem, but I do see (non-authoratative) references in the mailing list to such a parameter, and more problematically, I note that the default ConTeXt template for Pandoc includes the line: \setupthinrules[width=15em] % width of horizontal rules Before reporting an issue to the Pandoc team, I just want to confirm that this is, in fact, both the case (as reading the source and my testing indicate) and the intent (which I cannot attest). I would further suggest that Pandoc should be using setupblackrule with appropriate height, depth, rulethickness, and width. I suspect that Pandoc typography is not sensitive to control of the actual placement relative to the baseline, as ConTeXt supports, but a pointer to a clear explanation of the interaction of height, depth, and rulethickness would be welcome. -- Rik

1 0

distance parameter of \setupinitial
by Henning Hraban Ramm 14 Dec '20

14 Dec '20

Hi, I’d like an initial in front of the text block. It works more or less, but I find the behaviour of the distance parameter confusing: If I set it to -1em (like in the MWE below), the text lines get indented a bit on the right side. With somewhat bigger values, the indent disappears, but the initial is moved to the left. But the initial should nearly touch the first characters, and the text lines should be full width. The behavior is the same with (latest) MkIV and LMTX. Is there a misunderstanding of the parameters on my side, or is this a bug? \setupinitial[ distance=-1em, n=2, hoffset=1em, ] \showframe \starttext \placeinitial{L}orem ipsum. \input lorem \stoptext Hraban

1 0

two arguments for \xmlraw?
by Pablo Rodriguez 13 Dec '20

13 Dec '20

Dear list, this is a basic question about \xmlraw. Altough both xml-mkiv.pdf and i-context.pdf only grant them the node argument, I think it works with two arguments: \xmlraw { NODE } { PATH } I used in some documents and I couldn’t make it work with one argument (even when path reads "."). Am I missing something or does \xmlraw have two arguments? Many thanks for your help, Pablo -- http://www.ousia.tk

1 0

Re: [NTG-context] LMTX and addplot table
by Hans Hagen 12 Dec '20

12 Dec '20

On 12/12/2020 12:19 PM, Jean-Philippe Rey wrote: > > >> Le 12 déc. 2020 à 11:58, Hans Hagen <j.hagen(a)xs4all.nl >> <mailto:j.hagen@xs4all.nl>> a écrit : >> >> On 12/12/2020 10:36 AM, Jean-Philippe Rey wrote: >>> Dear list, >>> I am unable to draw a plot using an external file with LMTX. Here is >>> a minimal example >>> ===================== >>> \usemodule[pgfplots] >>> \starttext >>> \starttikzpicture >>> \startaxis >>> \addplot table [x=t, y=E] {mfe-addplot.dat}; >>> \stopaxis >>> \stoptikzpicture >>> \stoptext >>> ===================== >>> The ConTeXt run fails with an "Undefined control sequence" error (see >>> attached log file). I don't know if this is an LMTX or PGF issue. The >>> example works fine with MkIV. I tried with the tikz version available >>> on ConTeXt garden (pgf 3.0.1, pgfplots 1.16) and the last one from >>> CTAN (pgf 3.1.7a, pgfplots 1.17) with the same result. >> >> looks like \PackageError is undefined which is something PGF >> >> but maybe you should >> >> ! Package pgfplots Warning: running in backwards compatibility mode >> (unsuitable tick labels; missing features). Consider writing >> \pgfplotsset{compat=1.17} into your preamble. >> >> use that setting in order to not end up at that command > > Thanks Hans, but it doesn't seem to change anything. I am not really a tikz user but can run it here wipe t-tikz.* on your machine in order to use m-tikz.mkiv in the distribution because it looks like it does a bad catcode table change ... that went unnoticed because a chardef'd constant gets typeset (and when that happens before \starttext you won't see it but it actually does introduce some funny char); in lmtx we don't use chardef for that so that can give an error then when loading anyway, you're not suffering from that i think but better use the right module loader anyway then you can define the \PackageError macro (I did it with some trial and error but the next seems to work: \let\PackageError\writestatus so I'll add something like that to m-tikz.mkiv after all that I get this: PGFPlots: reading {mfe-addplot.dat} pgfplots > Sorry, could not retrieve column 't' from table '\\pgfplotstableread@filename@@table@name '. Please check spelling (or introduce name aliases). pgfplots > Sorry, could not retrieve column 'E' from table '\\pgfplotstableread@filename@@table@name '. Please check spelling (or introduce name aliases). pgfplots > Sorry, the requested column number '' in table 'mfe-addplot.dat' does not exist!? Please verify you used the correct index 0 <= i < N. pgfplots > Sorry, the requested column number '' in table 'mfe-addplot.dat' does not exist!? Please verify you used the correct index 0 <= i < N. NOTE: coordinate (,) has been dropped because of a coordinate filter. (see also unbounded coords=jump). pgfplots > Sorry, the requested column number '' in table 'mfe-addplot.dat' does not exist!? Please verify you used the correct index 0 <= i < N. pgfplots > Sorry, the requested column number '' in table 'mfe-addplot.dat' does not exist!? Please verify you used the correct index 0 <= i < N. NOTE: coordinate (,) has been dropped because of a coordinate filter. (see also unbounded coords=jump). pgfplots > Sorry, the requested column number '' in table 'mfe-addplot.dat' does not exist!? Please verify you used the correct index 0 <= i < N. pgfplots > Sorry, the requested column number '' in table 'mfe-addplot.dat' does not exist!? Please verify you used the correct index 0 <= i < N. NOTE: coordinate (,) has been dropped because of a coordinate filter. (see also unbounded coords=jump). pgfplots > Sorry, the requested column number '' in table 'mfe-addplot.dat' does not exist!? Please verify you used the correct index 0 <= i < N. pgfplots > Sorry, the requested column number '' in table 'mfe-addplot.dat' does not exist!? Please verify you used the correct index 0 <= i < N. NOTE: coordinate (,) has been dropped because of a coordinate filter. (see also unbounded coords=jump). ! Package pgfplots Warning: the current plot has no coordinates (or all have been filtered away) [warning /pgfplots/warning/plot without coordinates] ! Package pgfplots Warning: You have an axis with empty range (in direction y). Replacing it with a default range and clearing all plots. [warning /pgfplots/warning/empty range cleared axis] which still looks bad but at least you don't crash .. i'll look into that later today (as i'm splitting mkiv/lmtx code i use a slightly different file reader so maybe that's it) 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

Inkscape 1 without gui
by Martin Hasoň 12 Dec '20

12 Dec '20

Hi all, Inkscape version detection doesn't work for Inkscape 1 and non-gui environments (eg docker). The output of the `inkscape --version` command is > Unable to init server: Could not connect: Connection refused > Inkscape 1.0 (4035a4fb49, 2020-05-01) > Pango version: 1.44.7 This is standard commands behaviour without `--batch-process` and `--without-gui` options https://wiki.inkscape.org/wiki/index.php/Using_the_Command_Line Because the version of inkscape is on the second line, ConTeXt use options for 0.92 version and Inkscape 1 ends with an error message: > Unable to init server: Could not connect: Connection refused Unknown option --export-pdf=/app/pages/m_k_i_v_01.svg.pdf Best regards Martin Hasoň

2 1

Update luatex for linuxmusl-64 platform in standalone distribution
by Martin Hasoň 12 Dec '20

12 Dec '20

Hi all, would it be possible to update luatex for the linuxmusl-64 platform, please? The current version ( http://minimals.contextgarden.net/setup/linuxmusl-64/bin/) ends with error: > This is LuaTeX, Version 1.09.2 (TeX Live 2019/dev) (INITEX) > system commands enabled. > (/usr/share/tex/texmf-context/tex/context/base/mkiv/cont-en.mkiv (/usr/share/tex/texmf-context/tex/context/base/mkiv/context.mkiv (/usr/share/tex/texmf-context/tex/context/base/mkiv/syst-ini.mkiv) !!!!: Your luatex binary is too old, you need at least version 1.10!)) > 121 words of node memory still in use: > 11 glue_spec nodes > avail lists: 2:2,5:1 > No pages of output. In Dockerfile for alpine linux https://github.com/minidocks/context, I fixed this bug by downloading the latest version from https://github.com/TeX-Live/luatex. Best regards Martin Hasoň

3 2

LMTX musl broken download link
by Martin Hasoň 12 Dec '20

12 Dec '20

Hi Hans, download link http://lmtx.pragma-ade.nl/install-lmtx/context-linuxmusl.zip on http://www.pragma-ade.com/install.htm (64 bit Linux Musl) page doesn't exist. Would it be possible to fix the link? Best regards Martin Hasoň

1 0

LMTX and addplot table
by Jean-Philippe Rey 12 Dec '20

12 Dec '20

Dear list, I am unable to draw a plot using an external file with LMTX. Here is a minimal example ===================== \usemodule[pgfplots] \starttext \starttikzpicture \startaxis \addplot table [x=t, y=E] {mfe-addplot.dat}; \stopaxis \stoptikzpicture \stoptext ===================== The ConTeXt run fails with an "Undefined control sequence" error (see attached log file). I don't know if this is an LMTX or PGF issue. The example works fine with MkIV. I tried with the tikz version available on ConTeXt garden (pgf 3.0.1, pgfplots 1.16) and the last one from CTAN (pgf 3.1.7a, pgfplots 1.17) with the same result. The same example works fine if I provide the data inline. This looks like a problem while reading the data file. Thanks for your insight. Best regards, -- Jean-Philippe Rey jean-philippe.rey(a)centralesupelec.fr 91192 Gif-sur-Yvette Cedex - France Empreinte PGP : 807A 5B2C 69E4 D4B5 783A 428A 1B5E E83E 261B BF51

2 1