- ntg-context - NTG Mailing Lists

Why is the fonts database not updated?
by Gerben Wierda 07 Jul '23

07 Jul '23

I moved the fonts form my personal library to the system-wide library but context keeps on keeping the font setup to look for the personal library After I have run mtxrun --script font --reload --force and I compile my test file I see: fonts > defining > source file '/Users/gerben/Library/Fonts/Ophian-Regular.otf' is not found fonts > defining > using source file '/Library/Fonts/Ophian-Regular.otf' due to cache mismatch fonts > defining > source file '/Users/gerben/Library/Fonts/Cabin-Regular.ttf' is not found fonts > defining > using source file '/Library/Fonts/Cabin-Regular.ttf' due to cache mismatch Is there a better way to force a real reload of the font cache? Gerben Wierda (LinkedIn <https://www.linkedin.com/in/gerbenwierda>, Mastodon <https://newsie.social/@gctwnl>) R&A IT Strategy <https://ea.rna.nl/> (main site) Book: Chess and the Art of Enterprise Architecture <https://ea.rna.nl/the-book/> Book: Mastering ArchiMate <https://ea.rna.nl/the-book-edition-iii/>

3 2

lpdf.registerfontmethod
by Max Chernoff 07 Jul '23

07 Jul '23

Hi Hans, In lpdf-emb.lmt, there's a commented-out definition of lpdf.registerfontmethod that I'd like to use. I've uncommented this in my installation and it's worked pretty well for me, so can you please uncomment/enable this by default in the distribution? If you want further background, this is how I'm using it: lpdf.registerfontmethod("rawpdf", function(filename, details) return details.properties.indexdata[1], 0.001, function(char) return char.code, char.width / t3_to_sp end, function() end, function() end end) local function make_glyph(codepoint, unicode, width, height, code) local spec <const> = { width = width * t3_to_sp, height = height * t3_to_sp, depth = 0, unicode = { utf8.codepoint(unicode or "", 1, -1, true) }, code = width .. " 0 d0 " .. code, } local tfmdata <const> = fonts.hashes.identifiers[font.current()] tfmdata.characters[codepoint] = spec fonts.dropins.swapone( "rawpdf", tfmdata, { code = spec }, codepoint ) fonts.constructors.addcharacters( font.current(), { characters = { [codepoint] = spec } } ) end I specifically need something that writes out a raw PDF stream. I'm using this to write a 4000+ page PDF file to be included by pdfTeX, so I can't have any XObject indirection or similar, otherwise the number of objects skyrockets and pdfTeX (+qpdf) gets really slow. If you're curious, the actual code where I'm using this is here: https://github.com/gucci-on-fleek/unnamed-emoji/blob/master/source/svg-to-p… https://github.com/gucci-on-fleek/unnamed-emoji/blob/master/source/unnamed-… I know that what I'm doing in the .cld file is wildly unsupported, but I'm the only one who will ever need to run that. I am planning on distributing the .lua file though, so I'd like to get this change in the distribution. I'm also open to using a different interface instead if you don't want to publicly-expose lpdf.registerfontmethod. Thanks, -- Max

2 1

\showmakeup adds unwanted space despite accurate readings; \qquad mandatory before \TeX\
by Carlos 07 Jul '23

07 Jul '23

\showmakeup on \TeX\ shows a tSP:3.282 THK:-1.721 H__E X HK:-1.291 SP:3.282 I assume that SP stands for space. Correct me if I'm wrong After reading some of the manuals that mention \showmakeup with and without [spaces] and so forth, I couldn't find any more info related to this. If anyone can provide me more information as to what does HK means there? Anyhow If I were to have, say, a control _word_ such as \TeX\, the sequence of TeX kerning is thrown off. The only remedy would be to have a a \qquad preceding the control sequence e.g., {\qquad \TeX\ significantly} to sort of ameliorate this side-effect. But then again. Spacing is off. e.g., The separation of any of these four components would have hurt {\qquad\TeX\ significantly}. If I had not participated fully in But it's interesting that \showmakeup, kerning and spaces would display what's to be expected. And yet, the aformentioned set of tSP:3.282 THK:-1.721 H__E X HK:-1.291 SP:3.282 is accurate nevertheless. And a very nice implementation at that, but one I fear can't be included on a final document either. With that being said, the introduction of a \qquad, in addition to \showmakeup with, say, redundancy aside, whatever is preceding the \TeX\ alongside the file would also introduce a space where no space was ever there before. Because of all of this, and unfortunately, \showmakeup is not quite helpful here either. Although it does so displays the amount of spaces and so forth, any addition of a \qquad also adds a very subtle unwanted space. And since we're looking here, I pressume, solely for accuracy and perfection, we're left here with neither: zero, zip. It evaporates. -- Dear Emily: I'm still confused as to what groups articles should be posted to. How about an example? -- Still Confused Dear Still: Ok. Let's say you want to report that Gretzky has been traded from the Oilers to the Kings. Now right away you might think rec.sport.hockey would be enough. WRONG. Many more people might be interested. This is a big trade! Since it's a NEWS article, it belongs in the news.* hierarchy as well. If you are a news admin, or there is one on your machine, try news.admin. If not, use news.misc. The Oilers are probably interested in geology, so try sci.physics. He is a big star, so post to sci.astro, and sci.space because they are also interested in stars. Next, his name is Polish sounding. So post to soc.culture.polish. But that group doesn't exist, so cross-post to news.groups suggesting it should be created. With this many groups of interest, your article will be quite bizarre, so post to talk.bizarre as well. (And post to comp.std.mumps, since they hardly get any articles there, and a "comp" group will propagate your article further.) You may also find it is more fun to post the article once in each group. If you list all the newsgroups in the same article, some newsreaders will only show the article to the reader once! Don't tolerate this. -- Emily Postnews Answers Your Questions on Netiquette

4 17

Title pages makeup extra blank pages
by Alan Braslau 06 Jul '23

06 Jul '23

Hello, Trying to use: \startdocument \startpagefigure[Coverpages.pdf][page=1]\stoppagefigure \startpagefigure[Coverpages.pdf][page=2]\stoppagefigure % before \startmakeup [titlepage] front \stopmakeup \startmakeup [titlepage] back \stopmakeup \startfrontmatter ... \stopfrontmatter ... \stopdocument inserts a "blank" page (with header/footer) between the two Coverpages and a second "blank" page after the backside title page. Not including the two \startpagefigure lines eliminates the blank page following the back side title page (and the other one between the two Coverpages, of course). This is not a MWE as it occurs in a complicated project but not in mwe snippets. I suspect that this artifact is somehow related to the page={...} settings later in the document. How can I turn OFF all automatic page generation settings for the very beginning of my document? Any ideas? (Title page makeup has always been somewhat of a mystery to me.) Alan

3 7

Re: floatcombinations seem to not work correctly after lmtx update
by Gerion Entrup 06 Jul '23

06 Jul '23

Am Donnerstag, 22. Juni 2023, 15:21:12 CEST schrieb Hans Hagen via ntg-context: > On 6/19/2023 1:53 PM, Gerion Entrup via ntg-context wrote: > > Hi, > > > > I recently updated my Context (lmtx). Unfortunately, I'm not aware of > > the previous version anymore. The current one is 2023.06.04 18:58. > > > > After the update one of my documents broke at floatcombinations. > > The relevant code is this one: > the fact that you don't see a) coudl point to some mismatsh in > start/stop or { } I checked that again but don't think that this is the reason: - This place is not the only one in the document where floatcombinations failed (there where 2 more, in fact all places with floatcombinations). - It happened directly after the update of lmtx without changing the document. Before that, it looked fine. I experienced around a little bit more. When I increase or decrease the pictures in the document they shift around but the captions do not fit to the pictures. Attached are the renderings for 1 picture (1.png), 2 pictures (2.png) up to 5 pictures. This is the respective source code in the document for 3 pictures. I simplified the title additionally so that there is no error. "rendering.of.example.code.png" is the direct rendering of this specific code: ``` \startplacefigure[title={foo}, location=top, reference={fig:04-sstg-mapping}] \startfloatcombination[nx=3] \startplacesubfigure[title=Test] \externalfigure[dummy][width=0.2\paperwidth] \stopplacesubfigure \startplacesubfigure[title=Test] \externalfigure[dummy][width=0.2\paperwidth] \stopplacesubfigure \startplacesubfigure[title=Test] \externalfigure[dummy][width=0.2\paperwidth] \stopplacesubfigure \stopfloatcombination \stopplacefigure ``` Gerion

1 0

MPcode inserts page
by Henning Hraban Ramm 06 Jul '23

06 Jul '23

Hi, if I use \startMPcode to declare variables etc. that I need in several MP graphics, it always inserts an empty page. How can I avoid that? (Actually, I load an external mp file, and for random value reasons I must load it only once. But the MWE already shows the problem.) """ \startMPcode numeric something; something := 1; \stopMPcode \starttext \chapter{Something} Anything. \stoptext """ Hraban

2 2

SVG converter bugs
by Max Chernoff 06 Jul '23

06 Jul '23

Hi Hans, I've been using the SVG->MP converter and I've found a few bugs. Most of the bugs are with "weird" SVG input, but I have no control over the input, and the metadata on most of the files shows that it was exported from Adobe Illustrator which is (unfortunately) quite popular. I've tested all of the below with the latest ConTeXt. If you give Metapost an out-of-order gradient, it produces an invalid PDF. Both pages in this example: \startbuffer[gradient] <svg viewBox="0 0 200 100" xmlns="http://www.w3.org/2000/svg"> <defs> <linearGradient id="gradient" x1="0%" y1="0%" x2="0" y2="100%"> <stop offset="0%" style="stop-color:green;" /> <stop offset="75%" style="stop-color:green;" /> <stop offset="25%" style="stop-color:blue;" /> <stop offset="100%" style="stop-color:blue;" /> </linearGradient> </defs> <rect x="0" y="0" width="100%" height="100%" fill="url(#gradient)" /> </svg> \stopbuffer \starttext \startTEXpage \includesvgbuffer[gradient] \stopTEXpage \startMPpage fill unitsquare xyscaled (200,100) withshademethod "linear" withshadecenteronefraction (0,1) withshadecentertwofraction (0,0) withshadestep ( withshadefraction 0 withshadecolors (green, green) ) withshadestep ( withshadefraction 0.75 withshadecolors (blue, green) ) withshadestep ( withshadefraction 0.25 withshadecolors (green, blue) ) withshadestep ( withshadefraction 1 withshadecolors (green, blue) ) ; \stopMPpage \stoptext write this to the PDF /Bounds [ 0 0.75 0.25 ] which is invalid and renders incorrectly in every PDF reader that I've tested. This is what the SVG 1.1 spec has to say: Each gradient offset value is required to be equal to or greater than the previous gradient stop's offset value. If a given gradient stop's offset value is not equal to or greater than all previous offset values, then the offset value is adjusted to be equal to the largest of all previous offset values. The following SVG renders incorrectly, but replacing the 0% stop with the commented one fixes it: <svg viewBox="0 0 200 100" xmlns="http://www.w3.org/2000/svg"> <defs> <linearGradient id="gradient" x1="0%" y1="0%" x2="0" y2="100%"> <stop offset="0%" style="stop-color:#FFFFFF;" />  <stop offset="100%" style="stop-color:blue;" /> </linearGradient> </defs> <rect x="0" y="0" width="100%" height="100%" fill="url(#gradient)" /> </svg> Using a gradient as the stroke colour outputs an invalid colour to MetaPost. This example: \startbuffer[gradient] <svg viewBox="0 0 200 100" xmlns="http://www.w3.org/2000/svg"> <defs> <linearGradient id="gradient" x1="0%" y1="0%" x2="0" y2="100%"> <stop offset="0%" style="stop-color:green;" /> <stop offset="100%" style="stop-color:blue;" /> </linearGradient> </defs> <rect x="0" y="0" width="100%" height="100%" stroke="url(#gradient)" /> </svg> \stopbuffer \startluacode print(metapost.svgtomp{data = buffers.getcontent("gradient")}) \stopluacode gives this output: [...] withcolor "url(#gradient)" [...] "stop-opacity" seems to be ignored: \startbuffer[gradient] <svg viewBox="0 0 200 100" xmlns="http://www.w3.org/2000/svg"> <defs> <linearGradient id="gradient" x1="0%" y1="0%" x2="0" y2="100%"> <stop offset="0%" style="stop-color:red;" stop-opacity="1" /> <stop offset="100%" style="stop-color:red;" stop-opacity="0" /> </linearGradient> </defs> <rect x="0" y="0" width="100%" height="100%" fill="url(#gradient)" /> </svg> \stopbuffer \startTEXpage \includesvgbuffer[gradient] \stopTEXpage Finally, "display:none;" doesn't work: \startbuffer[gradient] <svg viewBox="0 0 200 100" xmlns="http://www.w3.org/2000/svg"> <rect x="0" y="0" width="100%" height="100%" fill="black" style="display:none;" /> </svg> \stopbuffer \startTEXpage \includesvgbuffer[gradient] \stopTEXpage Thanks, -- Max

1 0

definedocument, begin & closing
by Henning Hraban Ramm 06 Jul '23

06 Jul '23

Hi, if I use \startdocument, the setups "document:start" and "document:stop" get inserted at the beginning and end. But if I define my own document, it doesn’t work. What’s wrong? """ \definedocument[presentation] \startsetups[presentation:start] {\bfd\documentvariable{title}} \stopsetups \startsetups[presentation:stop] Thanks for all the fish. \stopsetups \startpresentation[title=Something] anything \stoppresentation """ Hraban

1 0

Arch linux package for context lmtx
by Aditya Mahajan 05 Jul '23

05 Jul '23

Hi, I maintained a AUR (Arch User Repository) package for context lmtx which was called luametatex. Arch recently introduced an official extra package for luametatex (the binary) called ... well ... luametatex, which means that my package has a name conflict with the official package and hence was deleted. Arch also now provides an official texlive-context package for context. I assume that it syncs with texlive. At present, texlive-context includes ConTeXt 2023.03.10. So, I think that the AUR package is still useful. It is essentially just a wrapper around the official install scripts and provides the latest release of context lmtx (and user modules). So, I am renaming my old `luametatex` package and am re-releasing it as `context-lmtx-bin`. To use it: git clone https://aur.archlinux.org/context-lmtx-bin.git cd context-lmtx-bin makepkg sudo pacman -U context-lmtx-bin-<release-date>.pkg.tar.zst This will install context lmtx in /opt/context-lmtx. This is in parallel with any system installed context. I also keep the old .tar.zst files around. That makes it easy to revert back to a previous version in case there is something that doesn't work as expected in the latest release. Note that by default the following third user modules are installed: 'filter' 'visualcounter' 'vim' 'tikz' 'typescripts' This is obviously a biased list corresponding to what packages I use. To install other packages, simply open the PKGBUILD file and change the _module variable defined as: _modules=('filter' 'visualcounter' 'vim' 'tikz' 'typescripts') You can change it to _modules=('--all') to install all modules. When using the package for the first time, run source /opt/context-lmtx/setuptex mtxrun --generate For any future runs, simply use source /opt/context-lmtx/setuptex once in your shell. This adds /opt/context-lmtx/texmf-linux-{32,64}/bin to the PATH variable, so the binaries provided by context-lmtx are found before the system binaries. The setuptex file provides a `resettex` function which removes the above path from the PATH variable. The setuptex file also sets a variable CONTEXDIST to 'lmtx'. This can be used to set the value of PS1 variable (or similar) of your shell prompt to add a visual reminder of which context installation you are using. Let me know if there are any issues with this install. Aditya

2 1

MP setup for a progress bar
by Henning Hraban Ramm 05 Jul '23

05 Jul '23

Thanks to Taco, Robbert and Jules for their work! I used the downtime to setup a presentation style with a progress bar in the footer. But I get the same (first page) graphics on all pages, and the footer background appears under the text area as well as under the margins. * I thought \useMPgraphics would be enough to re-calculate on every page? * How can I set a background to the footer area, but page width? * I tried \overlaywidth as well as OverlayWidth in the MP graphic – how do I get the page size? """ \setuppapersize[SW] \startuseMPgraphic{Blocks} fill unitsquare xysized (800 * RealPageNumber/NOfPages, 50) withcolor blue; \stopuseMPgraphic \defineoverlay[Blocks][\useMPgraphic{Blocks}] \setupbackgrounds[footer][background={Blocks}] \starttext \dorecurse{10}{\recurselevel\page} \stoptext """ Hraban

3 3