@baxx -- this is most easily done in several steps: \numberwithin{section}{chapter} will set up the first part which will also result in a two-part section number, then \newtheorem{thm}{Theorem}[section] will add the theorem number to the precomposed section number. (based on amsthm and amsmath, the latter for \numberwithin.)

Greetings from the seaside! ^^^

@CarLaTeX Hi to prof. van Duck! I hope you and he will enjoy the sunshine!

@egreg Of course! Yesterday was cloudy but today it's sunny!!!

@egreg Prof. van Duck says hallo back to you! :):):)

@CarLaTeX ooh

@PauloCereda Deratting in progress... he's doing it directly! ^^^

@CarLaTeX ooh catcodes

@PauloCereda LOL <3

@UlrikeFischer My mother has just told me that a letter from Mönchengladbach has arrived to my home today. I'm looking forward to seeing it (I'm on holiday till July, 31st). <3 <3 <3

@CarLaTeX Surprise, surprise ;-) @PauloCereda. The italian post system is faster than the brazilian, we posted the letter on friday evening.

@UlrikeFischer I will create a postal service based on duck delivery. :)

The messenger quacks when the letter is arriving. :)

@PauloCereda You all are known by my mother as "the friends of the duck" hahaha

@CarLaTeX awww <3

Amici dello papero

@PauloCereda Yes, she calls you "gli amici dell'anatroccolo"

@CarLaTeX that's so sweet

And prof. van Duck's video is "il video dell'anatroccolo che balla" :):):) (dancing duck video)

@CarLaTeX ooh mum knows a lot of things about ducks

@PauloCereda She's a prof. van Duck's fan :):):)

@CarLaTeX ooh

@CarLaTeX She told you that the letter if for the professore?

@UlrikeFischer No, unfortunately she can't read English...

@CarLaTeX It is on the address "Professore van Duck c/o Carla ..."

@UlrikeFischer Hahaha I'll ask her tonight!

@UlrikeFischer I was wondering what the postman could have thought LOL

@CarLaTeX Our postman always smiles when he delivers things to me, "is it another duck?"

@PauloCereda ROFL! He knows you well, then!

@CarLaTeX :)

@UlrikeFischer My mother has just emailed me the letter is addressed to Professore Paulinho van Duck c/o Carla... LOL

Did the latest L3 update breake MDT template again?

@Johannes_B MDT?

@DavidCarlisle we miss you

@Johannes_B ???

@ChristianHupfer Hallo!

bbc.co.uk/news/technology-40716304

@JosephWright We all be dead by 2020. :)

@PauloCereda You are a flash bot? I knew it!

@StefanKottwitz I am a duck. :)

@PauloCereda That's the output.

@PauloCereda bbc.co.uk/news/technology-40714395

@JosephWright David's gold TikZ badge is safe!

hello latexers, I have a quick question that i feel would belong better in the chat than on the "forum"

about microtype package. It is praised a lot on the web, and I couldn't find a single person commenting something bad about it or something we should take care about it. Many advocate its use for the writing of thesis

so far so good

but many universities have special guidelines about margins, like MIT which claims not to have less than 1 inch margins. So if I set a margin of 1 inch in the preambule and if I use microtype..... then I am toasted because microtype does protrusion right? So that in reality the real margin will be lesser than 1 inch despite me setting it in the preambule. am i right or am i missing something?

@no_choice99 Well, if your faculty secretary is an old fashioned person that judges theses using a ruler, you're indeed toasted. But protrusion is done in order to avoid optical effects that usually make commas, period or hyphens to look a tad inside the type block. Its effect is measured in fractions of a point, which is less than 1/72 of an inch. In metric units, probably under 0.2mm (a point is about 0.3mm).

@no_choice99 But probably your regulations require double spacing and similar aberrations, so there's no point in trying to improve the typesetting. :-(

@PauloCereda it's Ok I came back stopped @egreg getting a tick:-)

thank you egreg, ok!

@DavidCarlisle Now that was some tick stealing! :)

@AlanMunn sorry (for you, not for the other answer)

@DavidCarlisle what did you mean with that "are you sure"? here

@DavidCarlisle :) Yeah, I fixed the first error and you fixed the second. I suspected that my having run the document first with pdflatex was why it compiled with lualatex for me.

@Skillmon I meant \edef\contentsnameinlanguage{{\cyr\CYRZ\cyrm\cyrii\cyrs\cyrt}‌​} produces

! Undefined control sequence.
\GenericError  ...
                                                    #4  \errhelp \@err@     ...
l.7 ...uage{{\cyr\CYRZ\cyrm\cyrii\cyrs\cyrt}‌​
                                                  }

@DavidCarlisle It doesn't in the MWE posted by OP (I tested before posting)

@Skillmon what did you use? luatex?

@DavidCarlisle pdflatex (which I always use except if that's absolutely impossible)

\documentclass{article}
\usepackage[T2A,T1]{fontenc}
\usepackage[utf8]{inputenc}
\usepackage[english,ukrainian]{babel}
\begin{document}

\edef\contentsnameinlanguage{{\cyr\CYRZ\cyrm\cyrii\cyrs\cyrt}‌​}


\end{document}

@Skillmon ^^ ?

@Skillmon even if it works (which it doesn't in the above) , using general text in an \edef only ever works by luck, that's why there's \protected@edef

@DavidCarlisle you didn't use the code I provided in the comments

@DavidCarlisle you made a \def into a \edef

You might either compare \def\contentsnameinlanguage{{\cyr\CYRZ\cyrm\cyrii\cyrs\cyrt}} with \contentsname in \ifx or compare \edef\contentsnameinlanguage{{\cyrЗміст}} with \edef\contentsnameedef{\contentsname}.

@DavidCarlisle One uses \def the other \edef. That I personally like neither solution is not important here.

@DavidCarlisle and you're welcome to provide a better solution over there. As I said, I don't like mine as it's only a dirty workaround. I should ask some questions in the comments there...

@Skillmon you used \edef\contentsnameedef{\contentsname} so in that form you are in the "luck" category, it manages somehow not to make an error but the actual expansion is unusable nonsense, in general passing LICR text into an edef will give errors as in teh form I used

@DavidCarlisle ok, I'll try to remember that in the future.

@DavidCarlisle (Back to the skull question) Is there some principled reason why luatex doesn't trigger mktextfm since it will use the files if they exist.

@Skillmon workarounds is all you can do (the package is wrong as *name commands should just be text and that has defined it to be a group and a font command:-)

@DavidCarlisle When I read that last reply I was a bit offended (until I read it till the end...) :)

@AlanMunn i suspect Hans only loaded tfm at all as a sop to compatibility and just never thought anyone would want to use metafont generated fonts:-)

@DavidCarlisle I see. It's bit of a weird situation in the end.

@Skillmon :-) (no I'd only be that rude to @egreg, and that doesn't really count)

@DavidCarlisle Whats with @PauloCereda? And sometimes @CarLaTeX?

@Skillmon You mean the pineapple pizza thesis completion committee? I think David and I are founding members.

@AlanMunn it would probably be a trivial addition to luaotfload, although possibly not. with pdftex if the tfm is missing it's going to be a fatal error so you may as well try running some script to make one. But with luatex if there isn't a tfm you can move on and look for otf or ttf etc, so it's not necessarily so clear when you should try to make a tfm (but doing it before giving up with a missing font error couldn't hurt)

@AlanMunn Yes, I mean I'm a bit on both sites her. Hawaiian pizza can be delicious, but often it's annoying and there are so much better ones. I eat it, but never would pay for it.

@Skillmon never rude to @PauloCereda just remind him to finish his thesis, as a kindness.

@DavidCarlisle True.

@DavidCarlisle you should remind me more often that procrastination is bad...

@Skillmon I would do, but I have something to attend to first....

@Skillmon The fact that you admit that Hawaiian pizza can be delicious already puts you on a side. (The wrong one according to some...) :)

@AlanMunn not really. It depends so much on my mood. And the sauce used along with it. In general I eat it, if it's free of charge and I'm hungry :)

But I think I'd eat a lot under those circumstances which I don't really like.

@Skillmon The mere fact that you would eat it under any circumstances clearly shows you lack principles and moral virtue, not to mention good taste. (I'm channelling the other side here.)

@AlanMunn so you're on @CarLaTeX purist side?

@Skillmon Not at all, I'm just giving you a sense of how pure you must be to meet their expectations. :)

@Skillmon no @AlanMunn is a trained linguist, he knows how to emulate an Italian.

@DavidCarlisle Certo!

@DavidCarlisle pretty well (too bad we couldn't see his hands while he typed that)

@Skillmon Google was not involved.

@AlanMunn that's why I said I feel like being on both sides.

@Skillmon But that's my point. With extremism you can't be on both sides. :)

@AlanMunn But I can turn full Murican and bomb both sides, no?

@Skillmon Well nobody said that Americans are bastions of principles, moral virtue or good taste. But I only live there.

My chocolate is empty X(

anybody got chocolate? I have to write a report till Friday and need the sugar for my brain.

@DavidCarlisle An example of good way to use pineapple ^^^

@CarLaTeX all it needs is a pizza base at the bottom, and it would be perfect

@CarLaTeX What a waste of good pineapple and whatever alcohol is inside. :)

@Skillmon Since @AlanMunn is an excelent linguist, he knows it is not correct to call pizza the pineapple thing!

@CarLaTeX @Skillmon Unfortunately for Carla I have quoted Italian dictionaries whose definitions must necessarily allow pineapple pizza to be called pizza, but nobody listens to reason around here.

@AlanMunn Don't worry it was non-alcoholic :)

@CarLaTeX I"m not sure whether this is a reason to worry...

@AlanMunn I replied to that message :)

@AlanMunn Of course not, this is the internet

@AlanMunn And based on a German dictionary cold pizza is a salad.

@Skillmon That's right. Blame the Germans. :)

@AlanMunn I'm one... The definition provided there was just a bit too broad :)

@CarLaTeX What @DavidCarlisle and I really need to do is to find some particular dish for which your, @egreg and @Moriambar 's local versions are different enough for none of you to agree on which one is authentic. :) Then you can complain about each other.

@Skillmon Kind of like the "Is muesli+milk soup?" question. :)

@AlanMunn yeah. It just said "Cold food served with marinade or dressing" and one could argue that the tomato sauce on pizza is a marinade or dressing... Exactly like the muesli+milk = soup question

@AlanMunn All Italian versions are authentic. That's a great quality of Italian cousine: every area has its own dishes!

@CarLaTeX You say that now, but I'm sure there are heated internal battles about the "right" way to cook some things. :)

@Skillmon But that is really an atrocious definition for sure.

@AlanMunn Of course there are, but it's a pleasure to taste everything to choose which is better!

@CarLaTeX :)

In other news: (from the latest TeXShop release notes):

Another sharing option is "Airdrop". I think of this as an option for graduate students relaxing in Starbucks. If such a student notices someone interesting drinking coffee, they can use Airdrop to share a selected portion of TeX source code, or a selected region of Preview output. I keep hoping to be invited to a Wedding due to this feature, but not yet.

@Skillmon -- oh, but the best chocolate doesn't have all that much sugar. (i'm enjoying some 70% callebaut at this very moment.) see -- there are "purists" on lots of different subjects.

@barbarabeeton I don't like the ones with lots of sugar, too. But still it has a little bit (was eating 50% chocolate).

Why doesn't \@nameuse{\detokenize{εσιιε}} throw an error? I am pretty sure I never defined anything with that name and I think it's unlikely that it already is defined.

@CarLaTeX @Skillmon @barbarabeeton I can satisfy you all at the same time: flowersbuydelivery.co.uk/fab-fusion-chocolate-pizza-c-55-p-3519

@DavidCarlisle seems to be too sweet for me.

@Skillmon \csname zzzz\endcsname defines \zzzz to be \relax

@DavidCarlisle how unfortunate :(

@DavidCarlisle -- nope. not real pizza. just mimics the shape. an imposter. good shortbread also comes in that shape, and nobody is going to call that pizza.

@Skillmon exactly: it's an over-sweet, non-chocolate non-pizza, chocolate pizza, so offends pizza and chocolate purists equally

@Skillmon it's responsible for dozens of bugs in macro packages (and if i recall correctly the odd cheque for bugs in tex-the-program)

@DavidCarlisle OMG

@DavidCarlisle who needs that sweet stuff anyway? trying to kill an elephant with diabetes?

@DavidCarlisle You have an unequalled ability!

@AlanMunn thank you!

@AlanMunn To find the world champions of junk food over the Internet?

@CarLaTeX No, it's a much more refined ability than that. It's the ability to find perfectly offensive foods for the local audience. This takes real skill.

@AlanMunn I agree :):):)

@DavidCarlisle I don't quite get the reasoning for that definition of \csname to be honest.

@Skillmon it's just the way it is.

@DavidCarlisle yes it is, but I don't get the reasoning why it is that way (not like TeX fell from the skies -- I guess Knuth did know what he was doing)

I'm way to good at procrastination, didn't write a single line for that report, just set up a more or less useless system for defining named variables in that project...

@Skillmon it's why latex's \@ifundefined{foo} actually defines \foo to be \relax. (which means that if someone else does the same test as \ifx\foo\@undefined` they get a different answer. so a package like ifluatex which just needs to test if \directlua is defined has to go through all sorts of contortions to avoid defining that command and give the right answer if someone else has already defined it...

@Skillmon it's for the same reason that tex doesn't hyphenate the first word of a paragraph

@DavidCarlisle what's with \ifcsname?

@Skillmon that's half an attempt to avoid the problem, but it's etex so not used in latex2e core

@DavidCarlisle I didn't know that this is part of etex. I guess I should read more :)

@DavidCarlisle Why is that only half an attempt? I thought it was a good way to check for existence of a macro.

@Skillmon partly because it didn't go back in time and change all the \expandafter\ifx\csname#1\endcsname\relax to \ifcsname#1\endcsname so the damage was already done (see \@ifundefined) and it doesn't address using the name so really you want the form you used at the start, so \csname foo\endcsname to generate an error.

@DavidCarlisle So it's "Thanks Donald"?

@Skillmon in particular \ifcsname#1\endcsname is only a reliable test if you know no package has already done \@ifundefined{#1} on the same argument, and basically you never know that, so you always have to test for undefined-or-relax anyway

@Skillmon yes

@DavidCarlisle so one has to use a \expandafter\ifx\csname foo\endcsname\relax and a \ifcsname foo\endcsname test? That's horrible. I begin to hate the heritage we got from the \csname-issue.

@Skillmon but if you do the first, then there is no need to do the second as the name will always be defined, unless you do the first in a group.

@DavidCarlisle so one can't check that expandable?

@Skillmon which is why \ifcsname isn't as useful as it would have been if \csname hadn't been \let\foo\relax ?

@Skillmon ? both \ifx and \ifcsname work by expansion. (\csname defining the name to be \relax is the only expandable assignment that is available in classic tex, and it's used for that purpose in some particularly sneaky macros...

@DavidCarlisle what was Knuth thinking back in the days? I mean I'm really not that good in programming, it's enough atm, but not that spectacular, and a guy like Knuth (writing the art of computer programming) hadn't the feeling that this is a bad idea?

@DavidCarlisle I meant e.g. building one macro which is fully expandable checking for the existence or \relax of a given name executing one true and one false path.

@Skillmon the tex macro language was never really tested during the design phase, typical plain tex documents just used some simple shorthand wrappers around the primitives, and it was frozen before there were enough users to stress test the design, it's a testament to Knuth that it works as well as it does when used to program things like expl3 or tikz with thousands of lines of code.

@Skillmon yes that's what \@ifundefined{foo}{yes}{no} does but it also (expandably) defines \foo to \relax if it wasn't defined.

@DavidCarlisle I agree that he did a fabulous job (and wouldn't be using and learning (La)TeX if I thought the other way), but that one particular design choice seems odd to me

@Skillmon "odd" is an excessively polite description.

@DavidCarlisle But it defines it to \relax if not grouped what I think is not good behaviour.

@DavidCarlisle I thought you were British, you should be good at and used to express stuff excessively polite.

@Skillmon it's flippin stupid, but having spend a lifetime writing code to work round the issue, it seems like arguing that it's an inconvenience that you can't breath underwater. it's true in some ways, but it's just a fact of life.

@Skillmon I may be British but my politeness only extends as far as offering menu choices.

@DavidCarlisle but the inability to breath underwater is an inconvenience. And while you might be used to circumvent that issue, I'm certainly not that used to it (which explains why I asked the initial question to this discussion).

@Skillmon yes but I've had latex.ltx in an emacs buffer almost continuously since 1987, you do spot the occasional thing in there during that time:-)

@DavidCarlisle I'm a bit envious about your and others' knowledge and experience. Why did I come that late to the earth?

@Skillmon Would you like to be as old as @DavidCarlisle?

@egreg nah just have his and your knowledge in TeX and stuff :)