@GonzaloMedina @JosephWright \/

@cfr is my memory playing tricks on me or did you change your avatar?

@SeanAllred At some point but a long time ago. Not at all recently. Why?

@SeanAllred They are just Celtic knots from an online programme. I was thinking of changing it recently but I haven't actually got around to it, so unless you are reading my mind, that is probably not relevant.

@egreg I tried quite hard to get updmap to do nasty things just by asking it to list maps but, although it indeed listed available maps, it completely failed to do anything nasty as far as I can tell.

@cfr Great! I just run updmap on my system, following your advice :)

@GonzaloMedina You can test the instructions in the final section next, then!

@cfr Just looks significantly different to me somehow – maybe I am reading your mind :) (And I do recognize the general Celtic pattern :) I spent a lot of time in Ireland gawking at such things.)

Morning @DavidCarlisle

@JosephWright mornin

@JosephWright was thinking about putting some notes on compatibility in the ltluatex wiki (not sure what or how yet, was just looking how github wiki works)

@DavidCarlisle It's just a wiki thing: you tell it to make a new page then write it in Markdown

@JosephWright yes is there a direct way to make (from home wiki page) a link to a specific new page eg luatexbase compatability or do you just make that as self standing page and then fix up the link by url afterwards.

@DavidCarlisle Not explored it that far!

@JosephWright I'll have a play....

In most wikies I think you have to create the new page, and then go back to edit the old that link to the new.

@DavidCarlisle Add the link with [link text](page name). When clicking that link, you go directly to the 'create new page'-form.

@TorbjørnT. ah so that's like trac etc, didn't see it in the help text:-) thanks.

I see various people have found the new chatroom

@JosephWright github.com/josephwright/ltluatex/wiki

@DavidCarlisle Yes, it does - but again that’s only for French (and Tibetan, that Élie contributed as well).

@DavidCarlisle If you’re only including polyglossia I don’t think the code paths hit luatexbase at all.

@ArthurReutenauer OK I may try to make that test file do something in French, (I don't suppose you have any example document spell checking french with every other line written backwards do you?:-)

tex.stackexchange.com/posts/255742/revisions :)

@egreg I had to be quick before you made a competing answer with % added!

@DavidCarlisle ’fraid not ;-) But if you want some help testing it you may as well contact Élie Roux; he wrote that part of the code, and contributed to luatexbase when it was initiated.

@DavidCarlisle Oh, wait, Korean!

@DavidCarlisle The Korean gloss file was written by Kim Dohyun (also the author of luatexko) a few weeks ago and definitely makes interesting use of luatexbase. But I haven’t looked into it at all. It’s on master on GitHub (and also in the latest upload to CTAN if you use that).

@ArthurReutenauer probably I'll ping him to make a separate test of luatexko I can probably squeeze some French into my texbook quote example, but Korean might be harder

@egreg On the fontspec question: this is all @WillRobertson's fault :-)

@JosephWright Which one?

@egreg I ran our 'deprecation cycle' to remove some old names: the problem is they tend to hang around

@JosephWright So the problem is having installed the last snapshot? I can't yet see it on the mirror I use (and on the TeX Live site either). A hasty update of fontspec is needed.

@egreg Yes

@egreg I've sent Will a pull request on GitHub

@egreg we've had a few of these over the time I've been working on expl3: I do my best to get people to fix their code in advance but it's not easy, and we don't want to 'bake in' issues into expl3 (still flexible)

@JosephWright In the meantime you could suggest a workaround (I'll remove my comment)

@egreg I have (see my comment)

@JosephWright Tell Karl to delay adding the last snapshot till fontspec is updated?

@egreg I can ask

@JosephWright Or anybody updating won't be able to use fontspec: it's really a no go.

@egreg Will's already travelling around Germany so not sure he's set up to do a quick fontspec update

@DavidCarlisle: I am already disturbing the peace in the newly created chat room. :)

@PauloCereda I was going to mention that ducks were not invited there!

@egreg It does happen that we have some occasions where a breakage is in the wild for a day or so, but I will see what I can do

Australia: 378 - for 2 (translation: quite a lot)

@DavidCarlisle Eeek

@DavidCarlisle Duck segregation. :)

@egreg Message sent

@StefanKottwitz I notice an increase in supporters. sudo is back, Genmutant is back, clemens is more active on LC. That is very very nice :-)

@DavidCarlisle Indeed!

@Johannes_B That's great!

@ArthurReutenauer example failed once I made it use french:( (but it works now:-)

@cfr Posted

@DavidCarlisle Nice to hear, about the example failing.

← 1 message moved from TUG2015

@DavidCarlisle I've discovered how to move messages between rooms :-)

@JosephWright but could you move it after the 1 hour later banner:-)

@DavidCarlisle Reload the room: those are dynamic

→ 103852 messages moved to some room

@JosephWright oh yes

Anyone know if there are special rules for \fontsize in beamer templates? Starting to hack a template together, but apparently the lineskip part keeps getting ignored

@ArthurReutenauer was a "known issue" in that @JosephWright and I intentionally decided not to copy the priority feature where you can insist your callback happens in a specific order, you mainly find out how much something is used by seeing what breaks if it is not there....

@DavidCarlisle Yes, of course.

@ArthurReutenauer here the priority was checked just to see if the callback was registered (ie was non zero) the new code has a boolean test specifically for that so I just made that emulation, you'll get a type error if you use the priority as a number with that emulation but if as here it's just used in an if .... it works well enough.

@PauloCereda Another duck based answer: tex.stackexchange.com/a/255776/4427

@egreg Yay! :)

@PauloCereda Mother duck with her chicks.

@egreg Very nice answer. :)

@DavidCarlisle How's Australia going?

@DavidCarlisle Oh, 468/5, not bad.

I reached a point in my life, where i am amused and not upset about the following snippet found in a template: \usepackage[english]{babel} % Required to compile in Windows

@Johannes_B There was also the myth that \usepackage[latin1]{inputenc} was sufficient for getting accented and strange characters. I usually told to try “Straße”.

@egreg I am just wondering, if he actually tried the templates himself. There is \addfontfeature but fontspec is never loaded.

@Johannes_B Interesting!

@JosephWright yes I thought added_to seemed a bit strained or .callback_registered ...

@Johannes_B It's called "peer review", you are not supposed to test your own code:-)

@JosephWright just printed out boarding pass and stuff, so I guess I am going....

@DavidCarlisle when i sent the changes to the maintainer, he asked a few questions. One of them: What does package setspace do? The package is in the template for years.

@DavidCarlisle Peer review fail ;-)

@DavidCarlisle Also good

@DavidCarlisle :-)

@DavidCarlisle It's not the callback that's registered, its the function (description) registered for the callback

@JosephWright One on't cross beams gone owt askew on treddle.

@DavidCarlisle I overlooked this half an hour ago: \usepackage{blindtext} % Allows font customization

@JosephWright yes perhaps, although in other languages one would refer to the function as the "callback (function)". in_callback is OK for me

@JosephWright gregorio not in texlive :( testing that will have to wait...

@DavidCarlisle Oh, I see what you mean

@DavidCarlisle i have been a bad boy, i was looking at a modified file in which i changed loading fontspec with blindtext.

@Johannes_B good plan as @JosephWright just changed expl3 and broke fontspec.

@DavidCarlisle I know, he breaks stuff all the time :-)

@DavidCarlisle chickenize is amazing :-)

@JosephWright makes supporting luatex in the latex kernel all worthwhile....

@DavidCarlisle Exactly

@JosephWright about that tibetan font, doesn't fontspec have a way of saying use this or this or this, depending what's found? the tibetan ldf file suggested \newfontfamily\tibetanfont{Tibetan Machine Uni} but I didn't have that so I just picked the font my browser was using...

@DavidCarlisle No. You could implement it.

@egreg \scrollmode and see if you end up with \nullfont? or some more subtle lua query?

@DavidCarlisle Maybe HarfBuzz has a way to query the OS for a suitable font and from this one could make a \newfont declaration

@egreg or I could do what I did and use a standard Microsoft Windows platform font and assume everyone will have that?

@DavidCarlisle Probably I would not have it, but you'll not bother, of course.

566/8 declared

@egreg hey! we get to have a go now then

@DavidCarlisle England might go for declaring around 700

@egreg meanwhile seems that Ulrike has a font query here but needs updating a bit looking at comments

@egreg 4 for 1

@DavidCarlisle Good start!

@DavidCarlisle The main code still works. Only the name of the luc-file is store somewhere else. But one can use a hard coded path: myfonts=dofile("D:/texlive/2015/texmf-var/luatex-cache/generic/names/luaotfload‌​-names.luc")

@DavidCarlisle: After some digging I found that this here works now: myfonts=dofile(config.luaotfload.paths.index_path_luc). I will update the answer.

Can somebody please run this code?

\documentclass{article}
\usepackage{subcaption}
\usepackage{graphicx}

\begin{document}

\begin{figure}
\captionsetup[subfigure]{labelformat=parens}
\centering
\begin{tabular}{@{}p{1cm}p{1cm}@{}}
\begin{subfigure}[c]{1cm}
\centering
\includegraphics[width=1.0cm]{example-image-a}
\end{subfigure} &
\begin{subfigure}[c]{1cm}
\includegraphics[width=1.0cm,height=4cm]{example-image-b}
\end{subfigure} \\
\captionof{subfigure}{}
\label{sfig:testa} &
\captionof{subfigure}{}
\label{sfig:testb}
\end{tabular}

Whithout the \addtocounter{figure}{-1} the figure number comes out wrong. Is this a known bug or am I missing something?

@GonzaloMedina I think it's expected. Why not using \subcaptionbox?

@GonzaloMedina The subfigure environment increments the figure counter, and \subcaption decrements it in order to have the correct \@currentlabel. You have no \subcaption command, so…

@GonzaloMedina But otherwise subfigure is just minipage:

\documentclass{article}
\usepackage{subcaption}
\usepackage{graphicx}

\begin{document}

\begin{figure}
\captionsetup[subfigure]{labelformat=parens}
\centering
\begin{tabular}{@{}p{1cm}p{1cm}@{}}
\begin{minipage}[c]{1cm}
\centering
\includegraphics[width=1.0cm]{example-image-a}
\end{minipage} &
\begin{minipage}[c]{1cm}
\includegraphics[width=1.0cm,height=4cm]{example-image-b}
\end{minipage} \\
\begin{subfigure}{1cm}
\subcaption{}\label{sfig:testa}
\end{subfigure} &
\begin{subfigure}{1cm}
\subcaption{}\label{sfig:testb}

Meanwhile, Australia 568/8 declared – England 78/4

@egreg Yes, thank you.

My ducks have arrived! :)

@PauloCereda Cool

@JosephWright From UK. :)

@PauloCereda We are gaining Brazilians in return: my summer student brought in a friend today

@JosephWright Oh no! :)

@PauloCereda She stayed in our office all afternoon :-)

On the train: tidying up tags for siunitx :-)

@UlrikeFischer ooh thanks,

@egreg well we felt bad after beating them so convincingly in the first test, gave them a head start this time.

@DavidCarlisle :-)

@DavidCarlisle Quite a big head start

@JosephWright Lazy Brazilians. :)

@PauloCereda Well she is only over for a visit

@JosephWright Nah. :)

@UlrikeFischer, @egreg, @JosephWright vvv

\directlua{
myfonts=dofile(config.luaotfload.paths.index_path_luc)
function testfontname(name)
for i,v in ipairs(myfonts.mappings) do
if v.plainname==name then
 return name
end
end
return nil
end
%
tex.print("\string\\newfontfamily\string\\tibetanfont{" .. (
testfontname("Tibetan Machine Uni") or
"Microsoft Himalaya" % or any other suitable font
) .. "}")
}

@DavidCarlisle Working on mathstyle stuff for possible LuaTeX primitive changes: need to adjust l3build again :-)

@JosephWright just wondering if I ought to start putting things in a case...

@DavidCarlisle When do you fly?

13:20 take off

@DavidCarlisle Loads of time

@JosephWright quite so:-)

@DavidCarlisle Getting lots of practice with Git: fiddling with breqn is now done I hope

@DavidCarlisle github.com/wspr/breqn/commit/…

@JosephWright breqn is a topic in my upcoming book, don't break it ;-)

@StefanKottwitz Not planning to

@StefanKottwitz When will review copies be available?

@JosephWright September, but for personal comments I'll gladly send you something

@StefanKottwitz Cool

@StefanKottwitz I read about it just an hour ago. Still sitting over it. The template stuff is taking up quite some time. Sorry.

@Johannes_B No problem! Was just sharing my thoughts with you as one of the first I would ask

@Johannes_B templates :-) can be a plague but can be an entry

@StefanKottwitz I hope you will find @barbarabeeton to maybe join the upcoming discussion.

@StefanKottwitz btw, i would be interested in that chapter as well ;-) just to see you thoughts on the matter.

@Johannes_B I'm happy to see Barbara again.

@Johannes_B Which one? The templates chapter?

@StefanKottwitz Yes, that one. :-)

Hi, I have a font implementation question. I'm helping to adapt a package to XeTeX that with LaTeX uses TIPA. Since many unicode fonts include IPA glyphs, and TIPA doesn't work well with XeTeX, I will provide a command to set the IPA font using fontspec, (roughly \newfontfamily\ipafont ... ) but I won't set a default IPA font. So can I just \let\ipafont\normalfont for the XeTeX version? This should cause the ipafont to be the same as whatever the chosen \setmainfont ... is by the user.

@Johannes_B Did you like the sound of the math chapter?

@StefanKottwitz Actually, not done reading yet. But so far, i like it. Ich kann nicht raus aus meiner Haut, i spotted some typos and wrote them down in a text file :-)

@StefanKottwitz I guess there will be some professional lectoring though :-)

@AlanMunn like this?

@Johannes_B :22852660 Each single typo is great to know! There are editors, but I rather rely on you as editors are not committed to TeX

@StefanKottwitz I guess I'll have to be ready for a review on my website (and yes @NicolaTalbot I know I've still got to finish reading your one!)

@JosephWright As your website is famous, yes :-)

@DavidCarlisle Well sort of. I'm totally aware of that fact. But this is for a package which has a command lets call it \ipafont for the IPA it uses; this will be TIPA for LaTeX and I want this to be settable by the user but otherwise default to whatever is the main font with XeTeX So am I right in assuming that I just need to \let\ipafont\normalfont and this will do the trick?

@StefanKottwitz I'm planning at least one post a day at the meeting

@JosephWright Nice!

@egreg Found this damn \Hrule -> en.wikibooks.org/wiki/LaTeX/Title_Creation

@Johannes_B Note also the other pearl

\documentclass[pdftex,12pt,a4paper]{report}
\usepackage[pdftex]{graphicx}

@AlanMunn well yes as far as the various ipa things are covered by unicode, it's been a long time I looked at tipa but I think it had extra commands and markup probably some of it is assuming th etipa encoding

@egreg Oh for goodness sake

@JosephWright You probably can look at the sources to see when the option has been deprecated.

@egreg Maybe i can tackle that in a wekk or two.

Going home now. Save trip to all going to Darmstadt.

@egreg Nothing in the public SVN

@JosephWright It's possibly in pdftex.def and dvips.def

@egreg Indeed: I thought I'd start with the 2e SVN as dvips.def is extracted from graphics.dtx

@egreg Will try TL

@egreg @DavidCarlisle might know: is there a non-public SVN for 2e going back before the current one or was it all CVS?

@JosephWright I think there was svn for a while as I seem to recall complaining I couldn't connect at one point when i got active again and you or Rainer telling me the latex2e-pbulic setup had been that way for a couple of years:-) was RCS for a long time.

@egreg pdftex option? it was sort of deprecated from the sart.

@egreg Certainly 9 years: TL SVN only goes back that far. tug.org/svn/texlive/trunk/Master/texmf-dist/tex/latex/…

@JosephWright what are you looking for? (as in what feature, not why are you looking)

@DavidCarlisle Automatic selection of pdftex versus dvips

@JosephWright that was there in the initial design 1993

@JosephWright although it was more of a site choice then as dvips was not so dominent and there were more dvi driver options you could reasonably want as default.

@egreg An answer, perhaps?

@DavidCarlisle Really? Thanh's thesis is dated 2000 (pragma-ade.com/pdftex/thesis.pdf)

@JosephWright yes sure but the design of having a graphics.cfg that tested whatever it tested and set the default accordingly was there from the beginning

@JosephWright emtex or fptex for example....

@DavidCarlisle Question came from chat.stackexchange.com/transcript/message/22852815#22852815, specifically about pdftex as an option

@JosephWright Sebastian and I made that when I think he I and Than were about the only people on pdftex list, probably before his thesis was done...

1997-06-07 David Carlisle <latex-bugs@latex-project.org>

        * Makefile: pdftex.def

        * grfguide.tex: pdftex option

        * 00readme.txt: pdftex.tex driver file

        * color.dtx: pdftex option

        * graphics.dtx: pdftex option

        * pdftex.def: add this file for Han The Thanh's TeX variant.

@JosephWright @egreg ^^^^

@DavidCarlisle Ah

@JosephWright changes.txt in graphics directory

@JosephWright bed soon as I have a plane to catch tomorrow:-) I pushed some more emulation and tests on luatex as you probably saw.

@DavidCarlisle Yes, I don't care about TIPA in a sense: the assumption for the XeTeX package is that the IPA will all be handled with a unicode IPA font, either the main font or a separate one. The IPA input is user driven, not package driven.

@AlanMunn I don't care either as it's nearly midnight and I've got to drive to the airport in the morning:-)

@DavidCarlisle :) Thanks