What can cause latetx to produce an empty document even though there is [....] in the document? Never seen anything like this. I am trying to typeset long auto-generated code, and noticed that this one does not typeset at all. I get blank where I was supposed to get the equation typesetted.

@PauloCereda nice :)

Heiko update: -36 to be 100K

Hurray!

@HeikoOberdiek Just a good occasion for thanking you for your invaluable contribution to the TeX world!

@HeikoOberdiek Congratulations! :)

@cgnieder Maybe you should address them to Heiko. ;-)

@egreg sorry for the wrong ping

@cgnieder Never mind: I'm always happy if somebody congratulates. ;-)

@HeikoOberdiek Congratulations!!

@PauloCereda So a new bird? :)

@egreg :)

@egreg @cgnieder @papiro Thank you for the congratulations.

@HeikoOberdiek @PauloCereda should be here soon and bring cake.

@egreg I brought cake! :)

@HeikoOberdiek Congrats! :)

@cgnieder Yep. :)

But I'll resist to my urge of creating a project named duck. :P

@PauloCereda Thanks. Where is the cake? @egreg promised you would bring one?

@PauloCereda You could use one of the Latin names for a species of the family Anatidae.

@HeikoOberdiek: Congratulations (from OG to FR ;-))

@ChristianHupfer Two Baden citizens! Although OG used to be a free town, IIRC.

@egreg: I live in OG here only by force (my fiancée ;-)), but I do not regard myself as a Baden citizen... I am a Black Forestrian :D There is a huuuuuuuuuuuge difference! ;-)

@ChristianHupfer Did you bring your regular cuckoo clock?

@egreg: I live in such a clock :D

@egreg: The fact is, my family used to work in the clock industry for generations, but due to the decline in the last 20 years I chose another direction

@ChristianHupfer Thanks. Not FR, now I am living in Sasbach/Ortenaukreis (OG).

Small question. Why did anyone invent \@? What's the advantage of that versus a “forced space” \ to ensure a correct spacing after a point which does not end a sentence? E.g., Dr. Knuth, Dr.\ Knuth.

Or is it useful in any other occasion where \ is not?

@HeikoOberdiek: Near Achern? Well, almost neighbours :D

@ChristianHupfer Yes.

How about a TeX - Stammtisch Ortenau... You and me?

@HeikoOberdiek: :D

@ChristianHupfer There is a Stammtisch in Karlsruhe, usually the first Thursday of a month. Next month with 14th August is an exception.

It is very desirable that packages like tcolorbox and mdframed agree on a common set of option names. This way one could easily switch from using one over the other in not too complex uses of them. Would that be feasible?

@HeikoOberdiek: I was rather joking, since the region is too small for a Stammtisch. And most probably I am moving back to Black Forest during the next months, away from the Rhine valley, so Karlsruhe would not be an option

@Manuel have you seen this:

@cgnieder Ah… I had my mind wrong, due to this before/after different use. Thanks.

@ClaudioFiandrino @Mico @GonzaloMedina I think we are way too fast for deleting this, besides it's a wrong answer not a not-an-answer

tex.stackexchange.com/questions/21317/appendix-without-sections/…

@HeikoOberdiek No spammer today?

@JosephWright They didn't want to spoil the party.

@egreg :-)

@JosephWright I have seen neither the daily spam nor a suspicious account. @egreg :-)

@HeikoOberdiek Cool: I'm on holiday so my internet is on-and-off, so I was worried I might have missed it!

Have had to hunt for a connection today!

Hi, probably simple question. How do I get a subscript before the character rather than after? e.g. a_b puts the b at the bottom right, I wish to do something similar on the bottom left. Alternatively: any good notations for the \frac{f+2}{f}-log of a variable? :-)

@1010011010 tex.stackexchange.com/a/11566/37907

@1010011010 Concerning the \log part: no idea what notation you mean :-)

@HeikoOberdiek Congratulations!

Hi, my question (tex.stackexchange.com/questions/192637/grid-system-for-slides) is getting very low attention. Any idea to make it more attractive/tips to start working on it?

@Johannes_B Basically writing a regular log as _e{log} or something

@Johannes_B or _10{log} depending on what notation you're used to.

@Johannes_B In my case I want to take the log of some variable, so yeah _\frac{f+2}{f}{log} if you know what I mean :p

@s__C Put a bounty on it. People will come.

how much @1010011010?

@JosephWright One for you

@GonzaloMedina Thanks.

@percusse undeleted from my side :)

@egreg Nobody brought me a cake, thus I have baked one:

@HeikoOberdiek A new version of hyperref? ;-)

@egreg No a real cake :-)

@HeikoOberdiek I want a big slice!

@HeikoOberdiek I was hungry, so I ate the cake. :)

@HeikoOberdiek I could call enten. :)

@PauloCereda And you again forgot to vote! Never mind, I'm rep capped anyway. :P

@egreg Today it's not my fault. :) We had a special mass today in the morning, it was Saint Anne. :)

@PauloCereda Excuses. ;-) Did you sing?

@egreg Yep, I was in charge of the liturgy. :) The song booklets made in LaTeX, of course. :)

So Lorde is New Zealand? Ouch, I thought she was British. :(

@PauloCereda noooooooo!

@tohecz Das Cake ist gone. :)

@PauloCereda Now I have to go and bake something ...

@tohecz Bread! :)

but only once yum upgrade will have finished, I'm not gonna work with an old crap :p

@PauloCereda very likely it's gonna be a cake with cottage cheese, strawberries, cacao and chocolate :)

@tohecz ooh me wanteee. :)

@Paulo: Could we not use a boolean methode like: rule is in yaml-format ---> use v3, else ---> use v4

@cis Oh hello! Nice to see you in here! :)

@cis: There's a problem: I cannot guarantee a smooth transition. But I have a backup plan: I will write an updated version with more syntatic sugar for YAML rules. :)

The real struggle is to use a data serialization format as an expression language. That's what bit me back. :(

I've never imagined how far the use would be. At first, it would simply be a string expansion. :)

But I won't give up on the tool. Never. :)

@PauloCereda I shall send you some when it's finished, right? But it'll take like 3 hours in total

@tohecz :)

@Paulo: Ok, another idea: arara: foo

@Paulo: Ok, another idea: arara: foo ---> Attention, no foo.XYZ was found, are you using the old syntax?! Try arara3: foo ------> arara3: foo ---> Running Foo ...................

@cis It looks good, but the real struggle is that a "compatibility mode" is very expensive. :)

I really prefer to keep the "engines" apart.

But I'll release an updated version, with rules following the YAML thing. :)

Good evening everybody

Finally @cis is here

@Johannes_B hi! :)

@PauloCereda Any breaking news concerning old/new arara?

@Johannes_B The current trend is: arara won't be broken, it will be forked into another tool; it will receive later updates, with bugfixes and improvements.

@PauloCereda So i have to learn how to draw a nightingale? :-)

@Johannes_B o.O

DO WE HAVE A INDUSTRIAL SPY?!

@PauloCereda ?

@Johannes_B This is the codename. :P

@Johannes_B Hope it is not martingale then you have to learn a lot to draw it

@percusse ooh.

@Johannes: but you can use arara which will be marked as maintenance release. :)

@Johannes_B I've been spotted. :)

Sometimes I talk too much. :)

@PauloCereda :-D

@PauloCereda Why can't we f%$%^&^%$#& interview you then?

@percusse Might be a bit too complicated for a two year old.

BÄMMM

@Johannes_B Actually, Brent and I are thinking of a potential name in case of a fork. Personally, I want to troll people by saying complicated tupian words. :)

@percusse Soon. :)

@PauloCereda Not my field of expertise.

@PauloCereda sandmaennchen.de/freunde/beitraege/piggeldy_und_frederick.html

@PauloCereda Help! A piano is being damaged beyond repair! ;-)

radiosvizzeraclassica.ch/en/songinfo/sid/…

@PauloCereda wait for the first document with % nightingale: arara: { pdflatex: { shell: on } } ;)

Hello everybody!

I see there are many expl3 experts here. How do I prevent ignoring spaces in \tl_set:Nn \l_ls_str_tl { #1 }?

@HenriMenke Why would you?

@egreg The input is a multi word string where I want to replace several stuff.

@HenriMenke I don't understand... spaces in #1 are kept if it contains them.

@HenriMenke Spaces are ignored in the definition, so it would be the same as doing \tl_set:Nn\l_ls_str_tl{#1}. But when you call the function doing that, spaces are not ignored, unless you're still under \ExplSyntaxOn

@egreg Ah! Now I see the problem. I'm doing the output inside of \ExplSyntaxOn...\ExplSyntaxOff. Can I do something like \obeyspaces inside it?

@HenriMenke Don't. Separate the definition part from the usage part.

@egreg I revise my statement. The problem is with \tl_map_inline:Nn. It ignores all space tokens in <token list>.

If you need spaces under \ExplSyntaxOn, use ~

@HenriMenke Oh, that's another story!

\seq_set_split:Nnn \l_ls_str_seq { } { #1 }\seq_map_inline:Nn \l_ls_str_seq {...}

When you split a token list at “nothing”, spaces are preserved

@egreg Seems like I was really naive when I started using expl3 a few days ago with the thought in mind that it would be easier.

@HenriMenke Correction: they don't :(

@HenriMenke Probably it's better if you explain your problem

@egreg You were saying that I should separate the definition from the rest. How would I do that?

@HenriMenke I was thinking to something different. What's the precise problem?

@egreg It's not a real problem. I was just playing around with expl3 a little and thought that I might write a dumb letterspacing macro which just inserts \, after every token of the input.

\documentclass{article}
\usepackage{xparse}
\ExplSyntaxOn
\tl_new:N \l_ls_str_tl
\tl_new:N \l_ls_out_tl
\NewDocumentCommand\textls{ m }
 {
   \tl_set:Nn \l_ls_str_tl { #1 }
   \tl_map_inline:Nn \l_ls_str_tl {
     \tl_if_blank:nTF { ##1 } {
       \tl_set:No \l_ls_out_tl { \l_ls_out_tl ~ }
     }{
       \tl_set:No \l_ls_out_tl { \l_ls_out_tl \, ##1 }
     }
   }
   \l_ls_out_tl
 }
\ExplSyntaxOff
\begin{document}
\textls{More words}
\end{document}

@egreg That is my current code.

@HenriMenke You have first to split at word boundaries (in a sequence), then do the mapping in each word. But with sequences it's easier.

@HenriMenke Here's an example

\documentclass{article}
\usepackage{xparse}
\ExplSyntaxOn
\NewDocumentCommand{\ptextls}{m}
 {
  \henri_ptextls:n { #1 }
 }

\seq_new:N \l_henri_sentenceinput_seq
\seq_new:N \l_henri_sentenceoutput_seq
\seq_new:N \l_henri_wordinput_seq

\cs_new_protected:Npn \henri_ptextls:n #1
 {
  \seq_set_split:Nnn \l_henri_sentenceinput_seq { ~ } { #1 }
  \seq_clear:N \l_henri_sentenceoutput_seq
  \seq_map_inline:Nn \l_henri_sentenceinput_seq
   {
    \seq_set_split:Nnn \l_henri_wordinput_seq { } { ##1 }
    \seq_put_right:Nx \l_henri_sentenceoutput_seq

@egreg Wow! You're right. It is much easier with seq, especially the output via \seq_use:Nn \l_henri_sentenceoutput_seq { ~ }.

@egreg I'm going to upvote a random post of yours!

@egreg I don't quite understand why you split off \henri_ptextls. What is the advantage besides from \henri_ptextls being etex-protected?

@HenriMenke Frank Mittelbach recommends doing it. Enough said. ;-)

@egreg I totally accept that! :-)

@HenriMenke The idea is that a user level macro is defined in terms of a programmer's level function. Of course it's not so strict, so the replacement text in \NewDocumentCommand can contain more than a single function call. Basically it should "normalize" the arguments, disposing of optional arguments and so on; it's not so strict, however. But "complicated" code should go inside its function.

@HenriMenke it's doing assignments so it's not expandable anyway.

hi, quick question, is there a command like \noexpand, only that it completely stops the expansion of \edef (or \xappto in my case)?

@canaaerus You can put your tokens in a \toks register and put it in \edef.

\toks0={\bf not expanded}
\edef\something{\the\toks0}
\tt\meaning\something
\bye

unfortunately then I'm back to unbalanced braces

@canaaerus \unexpanded{<tokens>}

ah nice, that works, egreg :-)

@AlanMunn Now you're an expert in Korean, just like @DavidCarlisle is a TikZ expert.

@egreg Yes, reading the documentation for the kotex package was a blast.

@AlanMunn Now you should try \tracingall=1 ;-)

@egreg The name of the package is also quite amusing since it's the same as a brand of tampon, at least in N. America. Somehow I don't think the package authors probably knew that...

@egreg In the package code there is \ifx가가% ...

@AlanMunn Interesting conditional.

@AlanMunn They only want predictable code.

@Paulo ^ Almost ready for the oven ;)

@AlanMunn It's quite nice, actually: if a Unicode engine is used the conditional is true, otherwise it's false, because the UTF-8 representation of 가 is <EA><B0><80>

@egreg I see. So it's ifxetex and ifluatex all in one unreadable ifx :)

@AlanMunn \ifxetex OR \ifluatex

@egreg Yes, that wasn't a logical and... :)

@AlanMunn This is not expandable, but does the same using only ASCII:

\begingroup
\catcode`\^=7 \catcode30=12 \catcode`\!=12 % for safety
\edef\next{\@gobble^^^^0021}
\expandafter\endgroup
\ifx\next\@empty % Start of code for Unicode engines

@egreg How does that work?

@AlanMunn In a Unicode engine, ^^^^0021 is a single token; in an 8-bit one it's six

So in an 8-bit engine, \next is ^0021, while in a Unicode engine it's empty.

@egreg Cute.

@AlanMunn After discovering it, I found out that Heiko already had a similar trick. ;-)

@egreg Which should not surprise us...

@AlanMunn :)

Hi guys

@1010011010 Hello!

@AlanMunn :-) Suuuuurely you all know how to typeset math up to and including the details, am I right?

@1010011010 Well not all of us. I'm a linguist, so I don't know much about the details of math typesetting at all, although I know a good amount about LaTeX more generally. Now if you want to know about trees...

@AlanMunn Linguists. Ahhh. Love that. Won't cut it though if I want to know how to subscript a character before another character. I somehow thing $_x\log$ is the wrong way to go about it, buuuuuut maybe there's no alternative. :-((

@AlanMunn Trees? So you're into botany? Or are you a linguist? I'm confused.

@1010011010 Almost ${}_x\log$

@percusse Ha. And that's the best practice way to go about it, too?

@1010011010 Not really. But I never put things on the left side of things.

@1010011010 I think the ideal is to use mathtools package and write $\prescript{}{x}{\log}$

The empty slot is for the upper left entry.

Aand, mathtools loads/enhances/fixes amsmath so you don't need to load amsmath separately

@percusse I got myself into typesetting a nasty equation anyway. :-( e.g. i.imgur.com/hGh3yx8.png?2

@AlanMunn Ah. Botanist indeed. wink wink

@1010011010 Is f number?

I mean is (f+2)/f is regular number?

With math I can't even find a safe word to ask if it is just a fraction that involves numbers.

regular fraction, normal number all taken...

@cgnieder LOL

@egreg Oh no! :)

Thanks @percusse \prescript worked for me.

@PauloCereda Ready for the cake? :)