- ntg-context - Mailing Lists

Re: [NTG-context] Polar pgfplots seems to be broken (Aditya Mahajan)
by Gavin 21 Dec '20

21 Dec '20

> Date: Mon, 21 Dec 2020 00:51:06 -0500 (EST) > From: Aditya Mahajan <adityam(a)umich.edu> > > On Sun, 20 Dec 2020, Gavin wrote: > >> Hello, >> >> I’m drawing solar system diagrams using the pgfplots module, but the polar library seem to be broken. It looks like it is calling something with a \begin{axis}... command. Here is the error message for a short document, included below. >> >>> tex error > tex error on line 18 in file /Users/Gavin/Documents/Computer/ConTeXt/Polar Plot Experiments/Untitled.tex: ! Undefined control sequence >>> >>> \pgfplots@environment@polaraxis@ [#1]->\begin >>> {axis}[#1,data cs=polar,axis type=polar] >>> l.18 \startpolaraxis[] >>> >>> 16 \startsection[title={Polar Plot}] >>> 17 \starttikzpicture % tikz code >>> 18 >> \startpolaraxis[] >>> 19 \addplot+[mark=none,domain=0:720,samples=600]{sin(4*x)}; >>> 20 \stoppolaraxis >>> 21 \stoptikzpicture >>> 22 \stopsection >>> >>> mtx-context | fatal error: return code: 256 >> >> I’d love a fix or an idea for a work around. > > This is a bug in pgfplots macro pgfplotsdefineaxistype, which uses an explicit \begin{axis} ... \end{axis} instead of the more portable \axis ... \endaxis. Attached is a quick fix, but it has been quite a while since I looked at pgfplots code, so it may be better to post a bug-report on pgfplots mailing list for a proper fix. > > Aditya > > URL: <http://mailman.ntg.nl/pipermail/ntg-context/attachments/20201221/d7278aca/a…> Aditya, Thanks for the quick fix! I will post a bug report on the pgfplots mailing list, including your helpful diagnosis of the problem. Gavin

1 0

Off-topic: Struggles with LPEG grammar
by Mojca Miklavec 21 Dec '20

21 Dec '20

Hi, I'm sorry for being slightly off-topic here, but this list might still be the best place to resolve lpeg-related questions :) 0.) Disclaimer: the challenge that triggered this curiosity came from Advent of Code 2020. In case you are taking part and you wan't to avoid spoilers, please stop reading here! (You have been warned.) https://adventofcode.com/2020/day/19 1.) My question: I don't understand why I cannot get ^1 to work "as advertised". Isn't this supposed to mean "one or more occurences of the pattern"? If I change "lpeg.P('b')" into "lpeg.P('b')^1" in the example below, the strings that match the initial grammar no longer match the modified grammar. (I would naively imagine that the secord pattern would get more rather than less matches.) 2.) Background: Most definitely the task on that page is supposed to be solved in a different way, but many people use Advent of Code as an opportunity to learn a new programming language, and when I read the task description, I wanted to figure out if I could solve it using the cute little lpeg. My initial attempt worked correctly (at least to solve the first puzzle), but then I realized that I cannot easily change the pattern from "matches a letter b" into "matches any number of b-s", and I fail to figure out why. Any hints would be greatly appreciated. Below is a not-so-minimal example. I can certainly try to reduce it further, but I would first like to ask whether I'm doing something obviously wrong by trying to replace r5 = lpeg.P('b') by r5 = lpeg.P('b')^1 in order to allow more than one occurrences of the letter b? My only explanation would be that perhaps "^1" is so greedy that the rest of the pattern doesn't get found. But I don't want to believe that explanation. local lpeg = require "lpeg" --[[ 0: 4 1 5 1: 2 3 | 3 2 2: 4 4 | 5 5 3: 4 5 | 5 4 4: "a" 5: "b" ]]-- local parser = lpeg.P{ "r0"; r0 = lpeg.V"r4" * lpeg.V"r1" * lpeg.V"r5", r1 = lpeg.V"r2" * lpeg.V"r3" + lpeg.V"r3" * lpeg.V"r2", r2 = lpeg.V"r4" * lpeg.V"r4" + lpeg.V"r5" * lpeg.V"r5", r3 = lpeg.V"r4" * lpeg.V"r5" + lpeg.V"r5" * lpeg.V"r4", r4 = lpeg.P('a'), r5 = lpeg.P('b'), } * -1 local parser1 = lpeg.P{ "r0"; r0 = lpeg.V"r4" * lpeg.V"r1" * lpeg.V"r5", r1 = lpeg.V"r2" * lpeg.V"r3" + lpeg.V"r3" * lpeg.V"r2", r2 = lpeg.V"r4" * lpeg.V"r4" + lpeg.V"r5" * lpeg.V"r5", r3 = lpeg.V"r4" * lpeg.V"r5" + lpeg.V"r5" * lpeg.V"r4", r4 = lpeg.P('a'), r5 = lpeg.P('b')^1, -- modified part that doesn't seem to work } * -1 strings = { "ababbb", "bababa", "abbbab", "aaabbb", "aaaabbb", }; local total = 0 local total1 = 0 for _, s in ipairs(strings) do if lpeg.match(parser, s) then total = total + 1 end if lpeg.match(parser1, s) then total = total + 1 end end print('total:', total, total1) In this example, total=2, total1=0. What I don't understand is why total1 is zero. Thank you, Mojca

5 10

Problem with \Caps
by Joaquín Ataz López 21 Dec '20

21 Dec '20

Good day to everyone. I have a problem that may be very simple to solve, but for which I have not found an answer in the wiki. Until relatively recently I was using the January 2020 version of ConTeXt Standalone, recently I have updated it to the December 18th version, and when I compile a file that I used to compile well, an error is generated because the \Caps macro is not recognized. I can't get the desired effect with \sc, it has to be with \Caps. I have tried to compile it with LMTX but there too is considered that \Caps is an undefined control sequence. -- Joaquín Ataz López Derecho Civil Universidad de Murcia

2 1

Problem with shading in Metafun (LMTX)
by Jairo A. del Rio 21 Dec '20

21 Dec '20

While shading works with MkIV, with LMTX the following breaks: \startMPpage comment("two shades with named colors"); fill fullcircle scaled 5cm withshademethod "circular" withshadecolors ((1,0,0),(0,0,1,0)) ; fill fullcircle scaled 5cm shifted (6cm,0) withshademethod "circular" withcolor (1,0,0,0) shadedinto "blue" ; \stopMPpage The log says: metafun > log > metafun > log > error: Unknown relation will be considered false metafun > log > metapost > log > metapost > log > <to be read again> metapost > log > : metapost > log > <for(100.21985)> if(EXPR0)>temp_any_u: metapost > log > temp_any_u:=(EXPR0);fi ENDFOR metapost > log > max->begingroup.save.temp_any_u;if.pair(EXPR2):pair.temp_any_u.elseif.string(EXPR2): string.temp_any_u.fi ;temp_any_u=(EXPR2)for.i=(TEXT3):if.i>temp_any_u:temp_any_u:=i;fi.endfor metapost > log > .temp_any_u.endgroup metapost > log > mfun_max_radius->...er(EXPR2)-xpart.center(EXPR2))++(ypart.center(EXPR2)-ypart.lrcorner(EXPR2)),(xpart.urcorner(EXPR2)-xpart.center(EXPR2))++(ypart.urcorner(EXPR2)-ypart.center(EXPR2))) metapost > log > endgroup metapost > log > withshademethod->...cimal.center(EXPR0)withprescript"sh_center_b="&ddecimal.center(EXPR0)withprescript"sh_radius_a="&decimal0withprescript"sh_radius_b="&decimal.mfun_max_radius((EXPR0)) metapost > log > fi metapost > log > <to be read again> metapost > log > withprescript metapost > log > withshadecolors->...hprescript"sh_color_a_"&decimal.mfun_shade_step&"="&colordecimals(EXPR0)withprescript"sh_color_b_"&decimal.mfun_shade_step&"="&colordecimals(EXPR1)else:withprescript metapost > log > "sh_color=in... metapost > log > <*> comment("two shades with named colors"); fill fullcircle scaled 5cm withshademethod "circular" withshadecolors ((1,0,0),(0,0,1,0)) metafun > log > metafun > log > Oh dear. I can't decide if the expression above is positive, negative, or zero. So this comparison test won't be 'true'. metafun > log > metapost > log > ; fill fullcircle scaled 5cm shifted (6cm,0) withshademethod "... metapost > log > Hans told me not to use the B tag, so I'm asking first in case I've missed something. Thank you in advance. Best regards, Jairo

2 2

Polar pgfplots seems to be broken
by Gavin 21 Dec '20

21 Dec '20

Hello, I’m drawing solar system diagrams using the pgfplots module, but the polar library seem to be broken. It looks like it is calling something with a \begin{axis}... command. Here is the error message for a short document, included below. > tex error > tex error on line 18 in file /Users/Gavin/Documents/Computer/ConTeXt/Polar Plot Experiments/Untitled.tex: ! Undefined control sequence > > \pgfplots@environment@polaraxis@ [#1]->\begin > {axis}[#1,data cs=polar,axis type=polar] > l.18 \startpolaraxis[] > > 16 \startsection[title={Polar Plot}] > 17 \starttikzpicture % tikz code > 18 >> \startpolaraxis[] > 19 \addplot+[mark=none,domain=0:720,samples=600]{sin(4*x)}; > 20 \stoppolaraxis > 21 \stoptikzpicture > 22 \stopsection > > mtx-context | fatal error: return code: 256 I’d love a fix or an idea for a work around. I’m using ConTeXt Suite (ver: 2020.12.15 10:10 MKIV fmt: 2020.12.16), but I’ve tried it with TeX Live 2020 and the MkIV included with LMTX. Both have the same issue. Thanks! Gavin \usemodule[tikz] % TikZ module for diagrams \usemodule[pgfplots] % PGFPlots module for graphs \pgfplotsset{compat=1.17} % Required to get new any features that could cause compatibility issues. \usepgfplotslibrary[polar] % ConTEXt \starttext \startsection[title={Rectangular Plot}] \starttikzpicture% tikz code \startaxis[] \addplot+[mark=none,domain=0:720,samples=600]{sin(4*x)}; \stopaxis \stoptikzpicture \stopsection \startsection[title={Polar Plot}] \starttikzpicture % tikz code \startpolaraxis[] \addplot+[mark=none,domain=0:720,samples=600]{sin(4*x)}; \stoppolaraxis \stoptikzpicture \stopsection \stoptext

2 1

Update for old computers
by Alain Delmotte 20 Dec '20

20 Dec '20

2 1

\dodoubleemptywithset not working in LMTX
by Christoph Reller 20 Dec '20

20 Dec '20

Hi, The following MNWE uses \dodoubleemptywithset. In ConTeXt MKIV it works as expected, In ConTeXt LMTX an error results. \starttext \unexpanded\def\doMyDefs[#1][#2]{\getparameters[MD#1][#2]} \def\MyDefs{\dodoubleemptywithset\doMyDefs} \MyDefs[A,B][key=val,k=v] \MyDefs[C] \typeTEX{\MDAkey} is \MDAkey. \typeTEX{\MDBkey} is \MDBkey. \typeTEX{\MDAk} is \MDAk. \typeTEX{\MDBk} is \MDBk. \stoptext Possibly, \dodoubleemptywithset is not used anymore in LMTX, in which case it should be removed. Otherwise this could be a bug. Or maybe I use it wrongly? Cheers, Christoph

2 2

Using ConTeXt to create dictionnaries
by Alain Delmotte 20 Dec '20

20 Dec '20

2 2

lmtx-20201218 | \usetypescriptfile | typescripts > unknown library
by Richard Mahoney 20 Dec '20

20 Dec '20

In the recent versions of LMTX I've been having trouble with finding my own typescriptfile -- type-sourceserifpro.tex (attached) This file lives in the working directory of all of the other project files, and is also here: ~/lmtx-20201218/tex/texmf-fonts/tex/context/user/type- sourceserifpro.tex This file is found and executes correctly with: lmtx-20200920 but fails with: lmtx-20201218: fonts > typescripts > unknown library 'sourceserifpro' (Latin Modern is substituted for Adobe's Source Serif Pro.) My setups haven't changed and are: \usetypescriptfile[sourceserifpro] \setupbodyfont[sourceserifpro,11pt] following what's found here: https://wiki.contextgarden.net/Use_the_fonts_you_want Does anyone have any thoughts on this? Best, Richard -- T +6433121699 M +64210640216 E rmahoney(a)indica-et-buddhica.org IM https://t.me/rmahoney W https://indica-et-buddhica.org/ Indica et Buddhica Littledene Bay Road Oxford NZ

2 1

Indic fonts with MkIV and
by Robert Zydenbos 20 Dec '20

20 Dec '20

Dear all, Recently I read about the use of Harfbuzz-enhanced LuaTeX with LaTeX for writing Arabic. Following that example, I wrote a minimal test text that worked excellently for Kannada script as well: ----- \documentclass{article} \usepackage{fontspec} \setmainfont{Kedage}[RawFeature={mode=harf;script=knda;}] \begin{document} ಇದು ಹೇಗಿದೆ? ನಾನು ಹೀಗೆ ತುಂಬ ಬರೆಯಬೇಕೆಂದು ಯೋಚಿಸುತ್ತಿದ್ದೇನೆ. \end{document} ----- I want to do this in ConTeXt too! Can anyone tell me how to translate “\setmainfont{Kedage}[RawFeature={mode=harf;script=knda;}]” into ConTeXt? This could be the key to using MkIV with Indic scripts. Robert

5 9