Morning @DavidCarlisle

@AlanMunn yes

@JosephWright morning, did you/Bruno change the lookahead stuff and break unicode-math while Will is out?

@DavidCarlisle The recent bool change has highlighted a few issues: a shame that Will is not about

@JosephWright ah it's that, I assumed it was something Bruno mentioned about changing the \halign brace stuff, but didn't look in detail

@DavidCarlisle Haven't seen that one

@JosephWright would you be free for a quick hangouts trial before the call, I never did figure out why the mic failed last time, it works in skype and other things, and there doesn't seem to be an echo test service for hangouts

@DavidCarlisle Sure: let me know when

@DavidCarlisle Yes, the lack of a test call is a bit of a pain

@DavidCarlisle Firing up Chrome now ...

thanks

@DavidCarlisle 'Hangout' me when you want: I'll be about in my office most of the morning

@DavidCarlisle not your fault?

@PauloCereda strangely, it never is.

@DavidCarlisle :)

I get the message "Latexmk: Errors, so I did not complete making targets" but the .pdf looks fine. Is latexmk known to create false alerts, if there is just an informative warning? I could not find a single line with "Error" in the .log file.

@JonasStein Do you have a MWE for us to try?

That might help.

@PauloCereda you could suggest using auctex instead of latexmk

@DavidCarlisle :)

@DavidCarlisle ^^

@PauloCereda clearly that picture code is in error, no \put commands.

@DavidCarlisle It's emacs. :)

@PauloCereda well is there anything else?

@DavidCarlisle ^^ there's vim

@DavidCarlisle ooh Windows 10 + Linux subsystem

@PauloCereda ^^ standard microsoft-issue ubuntu vim (that's as far as I got, I'll probably have to kill the window and re-start bash)

@DavidCarlisle You could also leave the window there... forever. :)

@PauloCereda no need, I can just do:

@DavidCarlisle you could become a registered vim user

@PauloCereda We could set up a register of vim unusers.

@egreg oh no

@PauloCereda unfortunately I have no MWE yet.

I used auctex long time ago, but it was very painful to use.

emacs stuff is not well documented and ships with an very unsecure packagemanagement system

@PauloCereda @CarLaTeX Help! I need you opinion, is the left "friendly" duck bill better than the old one?

@samcarter The left one is better.

@samcarter I like both!

@UlrikeFischer Thanks for your opinion! I also prefer the left one, however something with the perspective looks odd - I just don't know what @PauloCereda Maybe I should do both, a friendly duck and grumpy duck?

@samcarter Perhaps lowering the right eye in the left duck and probably bringing them together a bit. :) I like both! And the right duck isn't grumpy, it's just surprised!

@PauloCereda The eyes are surprised but the beak looks grumpy. The other one smiles ...

@UlrikeFischer ducks aren't grumpy. :)

@PauloCereda then it shouldn't like it were ;-)

@UlrikeFischer ooh

@samcarter It's difficult to choose, I think I prefer the left one :):):)

@PauloCereda Thanks for the tip, moving the eye looks better

@samcarter Awesome! Can we have the option for both beaks? :)

@samcarter: any plans for coming to TUG 2018?

@samcarter Those are perfect! :):):)

@PauloCereda That's the plan :) I will have to think about some intelligent way to handle the multitude of choices in a reasonable way ...

@PauloCereda Unfortunately chances are not good :( The trip would be quite expensive and many things for the next year are very vague (as I have a t***** to finish), which is a shame as TUG '15 was a lot of fun and I'd love go again.

@CarLaTeX Thank you :)

@samcarter In that case, poke me as I have interesting news for you. :)

@PauloCereda poke

@samcarter oh. :) As it involves sensitive information, could you poke me through email? :)

@PauloCereda I found your sourceforge email in the arara documentation - is this one convenient for you or shall I search for another?

@samcarter It will redirect to my real one. :)

@PauloCereda email is on the way :)

@samcarter :)

@samcarter Did you poke me already?

@samcarter: nevermind, got it. :)

@PauloCereda Great!

@samcarter re-poke

@PauloCereda Got it - thank you! Interesting information, I will think about it and get back to you asap :)

@samcarter Marvelous!

The duck population is increasing!

@egreg ooh

Which texeditor works fine with synctex? I want to try an alternative for Texstudio which crashes too often here and has too many bugs. Texmaker stretches the pdf in the internal viewer. Found a bugreport from 2014. seems to be open since then.

Is there a good alternative for Linux?

@PauloCereda I do not get the link between Franzl and Latex

@JonasStein 'Most of them'

is it correct, that I see warnings like "LaTeX Warning: Citation 'EGGEOD1973a' on page 5 undefined on input line 145." when compiling with latexmk, or should they not appear, because latexmk compiles in an intelligent order?

It looks a bit strange to me, because after such warning I would expect TeX could not solve the Citation, but the final .pdf includes the citation.

@JonasStein doesn't latexmk just re-run latex so that warning goes? (you have to get that warning on the first run of latex) (first two, usually)

@JonasStein There hasn't to be one, actually. Posted for fun.

@DavidCarlisle Is it correct, that I would always get the warning, even if things are fine?

Just because it always warns on the first run and latexmk does not filter that in the first run?

@JonasStein it's just speculation until the issue is reproduced, usually through a MWE.

@PauloCereda no, the question applies for all tex documents with an bib, compiled by latexmk.

@JonasStein Not true, just ran latexmk here on a paper of mine and nothing suspicious arose.

@JonasStein If you get warnings in the last run issued by latexmk, then your file has problems. Nothing is resolved in the first run, hence the warnings.

@JonasStein certainly latex will make the warning, I don't use latexmk so I don't know if it hides those warnings

@DavidCarlisle Every time latexmk finds a warning, it displays the message regardless the mode (verbose or quiet).

ok thank you

@PauloCereda No ducks here, only seagull ^^^

@CarLaTeX ooh

@CarLaTeX one mad seagull :D

They have webbed feet, however :)

@CarLaTeX I'll check them in a few days.

@egreg ooh

@PauloCereda Last day in Verona, now holidays!

@egreg I'll tell him to say hallo to you!

Hi, I have the same problem with acro, as reported here: tex.stackexchange.com/questions/383018/… how can I downgrade the l3kernel?

@JonasStein There was a lazy boolean evaluation added, perhaps patching acro seems a better approach.

@JonasStein \ExplSyntaxOn\bool_new:c { l__acro_RES_long-plural_set_bool } \ExplSyntaxOff should do it for the present

@JosephWright That. :)

@PauloCereda It's not actually the change in how expressions are tested, it's a parallel tidy up Bruno did (which is correct but shows up)

@JosephWright thank you very much. Where should I paste it to?

@JosephWright Really? That's.... interesting. :)

before begin document?

@JonasStein After loading acro should work

I'll probably put in a pull request later today

great!

@PauloCereda Bruno didn't put it as a separate commit so I missed it in the release notes

@JosephWright which reminds of my PR... :)

@JosephWright Ah Bruno... Still working on String Theory? :)

@PauloCereda I think so

@JosephWright :)

is what my section in the preamble looks like, but I still have the same warnings

@JosephWright is there something different for lualatex?

\documentclass{article}

\usepackage{acro}
 \ExplSyntaxOn\bool_new:c { l__acro_RES_long-plural_set_bool } \ExplSyntaxOff
\DeclareAcronym{RES}{
    short = RES ,
    long = Renewable Energy Sources ,
}
\listfiles
\begin{document}
    \ac{RES}
\end{document}

@AlanMunn: forgot to tell you: remember when we were talking about punctuation symbols in a paper title? I tried this last week, let's wait for the outcome. :)

@JonasStein ^^^ Works for me

I will test it now

@AlanMunn: I even paraphrased Spencer's survival of the fittest. :)

@PauloCereda Colons and erudition. A sure fire winner!

@JosephWright your example works here fine. I am still looking for the difference to my code

@AlanMunn I will laugh a lot if it works. :)

@JosephWright could you try this? gist.github.com/anonymous/8e31753da27663d3ecb31c7101451f3b

I think it works only for one acronym and fails for many

@JonasStein The boolean variable Joseph created relates to RES only.

@JonasStein You would need booleans for all your acronyms. E.g. \bool_new:c {l__acro_PRES_long-plural_set_bool}

@UlrikeFischer Exactly. :)

@UlrikeFischer ah. Thank you.

@JonasStein But there must be a better way. Just looking how acro creates the bools.

great. I will have to leave for an hour and will be back later. cu.

@JonasStein You can try this. But it is only meant as a work-around, I simply added a test if the boolean is defined.

\documentclass[paper=a4,fontsize=12pt, twoside, BCOR=8mm]{scrbook}
\usepackage{acro}
\ExplSyntaxOn

 \cs_undefine:N\__acro_declare_long:nn
 \acro_declare_property:nnn {long} {long}
  {
    \acro_for_endings_do:n
      { \prop_put:cnn {l__acro_long_##1_form_prop} {#1} { \c_false_bool } }
    \prop_put:NnV \l__acro_long_indefinite_prop
      {#1}
      \l__acro_default_indefinite_tl
    \acro_for_endings_do:n
      {
       \bool_if_exist:cT {l__acro_#1_long-##1_set_bool}
        {\bool_if:cF {l__acro_#1_long-##1_set_bool}

@UlrikeFischer add a PR to Clemens!

@PauloCereda I added the code to the bitbucket issue tracker.

@UlrikeFischer Thanks!

@AlanMunn, @egreg: you should definitely try this with your students: twitter.com/AcademiaObscura/status/890982750865543168

@PauloCereda Definitely 2 ;-)

@egreg Exactly!

Something like the prisoner’s dilemma

@PauloCereda Cute! I guess the interesting question is what's the % threshold at which people tend to choose the larger value.

@egreg I was thinking the same thing. :)

@AlanMunn It would be a very interesting social experiment. :)

Also, I wonder if the size of the class makes a difference. In a class of 9 I suspect nobody would want to be "that person". But in a class of 100, you're betting on being the 11th.

@PauloCereda I assume that was what the experiment actually was.

@AlanMunn Or a sadist teacher. :)

I also like this one:

@AlanMunn: I have a game for XONE named Trivia Murder Party. It's a 4-player game in which the murderer asks questions and you need to answer them correctly in order to survive. There was one along the lines of:

@AlanMunn LOLOL

Oh there's the last part: if everybody gets the money, everybody dies!

@egreg @PauloCereda Substitute "6 points" with "withdraw all your money at once" and "2 points" with "withdraw later" and you'll get the explanation of bank runs (directly from my Game Theory course).

Is automating this answer something that LuaTeX could do?

@AlanMunn yes (if I understand the code:-)

@DavidCarlisle There's a new question about it here: tex.stackexchange.com/q/383756/2693 See in particular my comment to show what's needed.

@DavidCarlisle Wow! My first real luatex answer. This is very cool.

@AlanMunn Awesome!

@PauloCereda Except it doesn't actually work. It only appeared to work. :( Deleted for debugging.

@AlanMunn ouch.

@PauloCereda When I centre the glosses it becomes clear that it's not doing what it's supposed to.

@PauloCereda Maybe I don't understand lua regex. Shouldn't %p*?%s give me punctuation delimited by a space?

@AlanMunn Yes. Sample?

@PauloCereda Take a look at the deleted answer. The second substitution.

@AlanMunn Parenthesis escaping?

@PauloCereda I'm not quite sure? I tried to use [!,.;:?] in the expression but that gives me garbage in the output.

\documentclass{article}
\usepackage{expex}
\lingset{glwordalign=center}

\usepackage{luacode}
\begin{luacode*}

function replace_punc ( buff )
   if string.find ( buff , "\\gla" ) then
      buff = string.gsub ( buff , "(``)" ,
             "\\nogloss{".."%1".."} @ " )
      buff = string.gsub (buff, "([!,.;:]+)",
             "\\nogloss{".."%1".."} @ " )
   end
   return ( buff )
end

luatexbase.add_to_callback ( "process_input_buffer", replace_punc, "replace_punc" )
\end{luacode*}


\begin{document}

@AlanMunn: sorry for being dumb, what is the transformation here?

@PauloCereda Trying to turn "foo!" into \nogloss{"} @ foo @ \nogloss{!"} (I've used " instead of backticks for simplicity of markup in the comment.)

@AlanMunn Ah I see.

@PauloCereda The first substitution appears to work.

@PauloCereda Doh. Figured it out. Edit in progress.

@PauloCereda This is a very powerful technique. Kind of scary. :)

@AlanMunn Sorry, I got distracted by a problem here. Glad you made it work.

@AlanMunn they are patterns not regex (similar but different)

@DavidCarlisle Ok, I was treating them as regex, but I see now they are not.

@DavidCarlisle Does tabularx process the table content twice?

\documentclass{article}
\usepackage{expex}
\usepackage{tabularx}
\parindent=0pt
\begin{document}


\begin{tabularx}{\linewidth}{@{}XX@{}}
%\begin{tabular}{@{}p{.5\linewidth}p{.5\linewidth}}
\ex
I want these examples to be on the same line.
\xe
&
\ex
But they're not.
\xe
%\end{tabular}
\end{tabularx}
\end{document}

With tabular the numbering is correct.

@AlanMunn yes, but it doesn't make sense to use tabularx here.

@UlrikeFischer But expex should still get the numbering right if you did...

@samcarter Lovely new avatar!

@AlanMunn Then expex should use latex counters instead of tex count. latex counters are reset by tabularx.

@UlrikeFischer Yes, it's problem. The author of ExPex is a dyed-in-the-wool plain TeX user who doesn't really know any LaTeX at all. But all his users use LaTeX...

@UlrikeFischer For a while the glossing macros were a separate part from the example numbering macros, which meant you could use other packages for numbering, but the more recent versions of the package have lumped everything together.

@AlanMunn usually more

@AlanMunn tabularx preserves all latex defined counters and only increments them once but if the macros are using \newcount defined registers tabularx won't know about them. (oh @UlrikeFischer said that already, I wasn't here:-)

@AlanMunn if you know which counters you want to preserve you could add them to latex's reset list then tabularx (and also \include) will know about them and set them to appropriate values

Want to pick up a LaTeX book bargain? :-o amazon.de/gp/offer-listing/B01FGMWRD4?language=en_GB

@DavidCarlisle Yeah, I don't really use the package, and I also have a hacked version that allows you to use the glossing macros with the other linguistics example numbering packages if needed so if someone else comes up with the problem I'll delve into it then.

@StefanKottwitz LOL This must be the collector's edition.