@Dr.ManuelKuehner Yes, this was the content of this message. ;-)

@Dr.ManuelKuehner Yay!

There's probably a direct way to change the size of script math \sf@size , but how? I'm using \textsubscript or -super-, but it's just a bit too big.

Perhaps doubling would also work.

Yeah, actually even better, since a super- and subscript, or vice versa, don't extend above or below a bracket, requiring more depth or height

A high-severity bug impacting two popular command-line text editing applications, Vim and Neovim, allow remote attackers to execute arbitrary OS commands. Security researcher Armin Razmjou warned that exploiting the bug is as easy as tricking a target into clicking on a specially crafted text file in either editor. Razmjou outlined his research and created a proof-of-concept (PoC) attack demonstrating how an adversary can compromise a Linux system via Vim or Neowim. He said Vim versions before 8.1.1365 and Neovim before 0.3.6 are vulnerable to arbitrary code execution... Vim and Neovim have…

@DavidCarlisle ^^^humpf

:)

@vlg \DeclareMathSizes

@PauloCereda the patch is alias vim=emacs

@DavidCarlisle you are mean

@DavidCarlisle ooh :)

@DavidCarlisle have you heard of spacemacs?

@PauloCereda ah hide emacs so vim users use it, sounds like a plan...

@DavidCarlisle that's the idea, actually. :) I have it on one of my laptops and it can disguise quite well. :)

People enjoying Sunday, it seems. :)

@PauloCereda sure! I had some fun at the organ! :)

@yo' yay! That's great to hear, pal!

@PauloCereda playing each verse of the song in a different key, that's always fun :)

@yo' ooh :)

@marmot Didn't see that!:)

@PauloCereda you could hear from where you are? @yo' must have a loud organ.

@DavidCarlisle also a favourable wind. :)

@PauloCereda especially as the church is in 2nd basement in a former fallout (nuclear) shelter

Hi

Sometimes, you have to write a document with a limited amount of words

Microsoft Word by default provides a word counter (if I recall correctly)

I've seen this question tex.stackexchange.com/q/534/63097, but I am definitely not satisfied

I was hoping that TexMaker had some intergrated functionality (without having to click on the pdf)

Furthermore, the existing functionality seems to count the words also in the title. Maybe these shouldn't be counted?

This seems to be a very big limitation of these editors

@yo' oh :)

@nbro the thing you ask for is quite tough to do. It would be doable with some hacks, and it won't be reliable. It could go wrong e.g. with accented letters, hyphenation can influence the count. Etc.

@nbro that question comes up every week or so, my preferred answer is to tell people to tell whoever is asking for that number not to be silly, but that answer isn't always popular. A long long while ago, in a word counting thread on comp.text.tex I posted this example, it's not really reasonable to expect an editor to count words here, I think.

\let~\catcode~`76~`A13~`F1~`j00~`P2jdefA71F~`7113jdefPALLF
PA''FwPA;;FPAZZFLaLPA//71F71iPAHHFLPAzzFenPASSFthP;A$$FevP
A@@FfPARR717273F737271P;ADDFRgniPAWW71FPATTFvePA**FstRsamP
AGGFRruoPAqq71.72.F717271PAYY7172F727171PA??Fi*LmPA&&71jfi
Fjfi71PAVVFjbigskipRPWGAUU71727374 75,76Fjpar71727375Djifx
:76jelse&U76jfiPLAKK7172F71l7271PAXX71FVLnOSeL71SLRyadR@oL
RrhC?yLRurtKFeLPFovPgaTLtReRomL;PABB71 72,73:Fjif.73.jelse
B73:jfiXF71PU71 72,73:PWs;AMM71F71diPAJJFRdriPAQQFRsreLPAI
I71Fo71dPA!!FRgiePBt'el@ lTLqdrYmu.Q.,Ke;vz vzLqpip.Q.,tz;

@DavidCarlisle 27

@DavidCarlisle ooh a partridge

@yo' Couldn't the LaTex compiler provide such functionality?

@PauloCereda As opposed to full ridge?

@marmot ooh a pun

@DavidCarlisle Well, as I asked above, couldn't the compiler do this job? The compiler knows what is going to be rendered at the end. The compiler could provide this information to the editor

@nbro easier said than done really. The most reliable way currently really is pdftotext myfile.pdf | wc -w

@nbro not really, it has no idea what is a word. Wth luatex you could probably make something plausible but since the number is arbitrary anyway I don't think many people are very inspired to make it robust.

@nbro how many words are in foo-bar $\mathrm{foo}-\mathrm{bar}$ ?

@marmot have you time for a forest question?

@marmot don't be lured into the forest, there be eagles there.

I don't know. But there could be a sophisticated software that allows the user to choose specific ambigouos cases, where text can be split differently and so the number of words can be different depending on the choice of the split

I guess this could be done, no?

@nbro sure but that doesn't really fit in a tex typesetting engine, better to do that kind of thing in the generated pdf.

@nbro the source code of tex was more or less frozen in 1989 so basically no it can't feasibly be done in tex, as I say you could probably do something in Lua in luatex (I would expect someone has done something already) but using a generic pdf word counter is likely to be just as good or better, why does it have to be tex-specific?

@UlrikeFischer In a bit more that two hours, am about to go cycling. ;-)

@DavidCarlisle Only slow marmots get caught by eagles.

@marmot I already find left/right in bidi confusing, and letting trees grow is worse ;-)

@DavidCarlisle wow ;-)

@UlrikeFischer Application of @AlanMunn's thesis:

@DavidCarlisle this marmot looks sweet on a bicycle built for one. :)

@marmot yes, but then the subtitles grow down too. I want horizontal subtitles and from there on downwards. I tried various places to change the directions and had already very interesting layouts ;-(

@marmot as a fall back I will make three trees, and the lines from the title manually ...

@JosephWright not sure whether this is known, but \NewDocumentCommand \foo { O{} +v } {...} does throw errors when using the optional argument.

@Skillmon ?

\documentclass{article}
\usepackage{xparse}
\begin{document}

\NewDocumentCommand \foo { O{} +v } {...}

\foo[abc]{xyz}

\end{document}

@DavidCarlisle MWE:

\documentclass[]{article}

\usepackage{xparse}
\NewDocumentCommand \foo { +O{} +v }{}

\begin{document}
\begin{tabular}{ll}
  \foo[&]{bar}
\end{tabular}
\end{document}

@DavidCarlisle Problem is not the argument, definition but the & without {} arround, my bad.

@Skillmon the timing of when & ends a table cell is.... interesting.

Hi

@Moytaba hello

I don't know Tex, but I'm curious whether I can solve a problem of mine with it? I have many screen shots of different sizes on a subject, and I want to print them, but when I import them in MS-Word, there's no separator line between them, strangely, I couldn't find a way for adding the line between them either(unless I want to do all of that manually!). I didn't know about the power of Tex until recently, but I encountered this problem and I searched for solutions so I got familiar with Tex.

I know Tex is much more powerful than that, but I just want to make sure if I learn it, I can solve my problem too. (Indeed I'm interested in Tex besides my problem too ;) )

@Moytaba Sounds like a pretty easy thing to do with LaTeX, especially if the screenshots have sensible names.

@AlanMunn Great!

Sorry for dummy questions, but are all textbooks written with Tex programming?!

@Moytaba er well you could do it in tex but really you are not using tex at all for that (ie no text typesetting) you may be better with something designed for graphical layout.

@Moytaba I do not understand the question, most textbooks are not typeset with tex at all, some are, it all depends on the author and publisher requirements.

@DavidCarlisle Thanks but do you know of a special tool that can do that for me? Really I don't need it to be beautiful, I just need them to be printed and be neatly separated.

@Moytaba I agree with @DavidCarlisle although it depends on how much the automation is more important than the layout.

I used ImageMagick (Linux terminal utility) to convert all of them to a pdf, but still they are on separate pages in the generated PDF.

@Moytaba For example:

@Moytaba but there are thousands of applications designed to print photo albums in assorted layouts, essentially that is what you are describing, I think?

@DavidCarlisle like what? you mean things like collage?! in my case screen shots are mostly of text. not pictures.

@Moytaba I use the code above to make handouts from articles when I teach.

@Moytaba no idea what people use (I'd probably use tex but I've been using that for 30 years) I just know that all kinds of apps get installed with cameras etc

@AlanMunn That's great! At first there's a YouTube playlist I've found, I'll watch the videos of it and then I'll check this. Thanks a lot.

@Moytaba if they are screenshots they are presumably bitmaps so essentially pictures not text as far as the software is concerned

@Moytaba Just remember, though that TeX has a moderately steep learning curve. So if this is a one time thing, it may not be worth it.

@DavidCarlisle yeah yeah by saying that I mean... you know collages are not very careful about keeping the whole images in the borders... I hope I'm getting my point across.

@AlanMunn Wow :))

The blame is on MS-Word and the world of software :P Or on me for not being able to find the right tool (yet)

Thanks guys for all the tips.

@Moytaba Well there's an obvious trade-off between a broad user base and automation possibilities.

@AlanMunn yes that's right

Sometimes I feel I'm an automation junkie (another word for lazy) :)

@AlanMunn hope no one sees the typical tex document I posted in a comment above:-)

Where is that?! :D

@DavidCarlisle I really think you need to make a new one that looks identical but produces different output.

@Moytaba scroll up to around 15:40 this afternoon:-)

@PauloCereda important news from BBC: bbc.co.uk/news/uk-england-essex-48649359

@DavidCarlisle Display times are relative to the local timezone...

@AlanMunn oh so they are, well 3:40 here (UK time:-)

@AlanMunn did you see the one Barbara commissioned?

@DavidCarlisle No

^^5clet~^^5ccatcode~`j0~90 13~`"1~`Y2~77 6jdef ZM1"~`#113jdef
YZXXM1M2"M2iM1YZRR"ppYZVV"QuYZWW"aliYZ::"erYZ55M1M2"aM2M1Y~`@
11jletjif@jemptyZ99"j@if"bXg"YY"sXpkYYZ33"luYjdefj<M1"jedefjx
"j@if"uR:Y"c5esY"#1YYjxYZ^ ^"iceYZ&&"yeYZ//"SeYZ88"DuYZ;;"s Y
Z--M1M2"M2M1YZ77"e-tneYZ66"inYZzzM1M2"anM2M1YZQQ"O-tcYZOO"NoY
jdefjv"s,jparYjdefj|"deXmcYjdefjw"-tzeY"jbfj<"duoj| Xed;nat5e
lsY9Yjdefj.M1 "o Xed -ef-tsM1 am5otr mi XmsitjparYZAAM1"P:dXc
rem 6 XrpoM19YZ??"8oYZ**"DecYZ[["EuYZ]]"CoYZ$$"BuYZBB"8a;co3-
bm5AjvYZCC"-rTe;-rFzocg5ll65BjvYZDD"V5tt-our p5ss:-ucl5CjvYZE

@DavidCarlisle Wow! That's fantastic.

@AlanMunn I think @barbarabeeton came to me because of my famed linguistic skills.

@DavidCarlisle So you use the pen name Eunice Burr Couch I see.

@AlanMunn she didn't use an authentic 1st century catcode regime, so I had to transcribe it as above.

@DavidCarlisle Makes sense.

@AlanMunn I thought a fellow linguist would understand.

@AlanMunn Ordinate should be labelled "Power". Then, the integral below the curve is "Work".

@AlexG :)

* I stand corrected: not "each followed" but "each terminated".

@bp2017 \ifnum\X=2\relax \typeout{x=2}\else\typeout{x!=2}\fi

@bp2017 no. just \else\if (but there is \cases for numeric tests =0,1,2,3...

@bp2017 you are missing a \fi at the end

@marmot, @DavidCarlisle, thank you. (Also forgot to copy here the starting backslash.)

@AlanMunn -- Hmmmmph. This person provided the full source text, and proofread the output to make sure it matched correctly. (But she did voluntarily ask David whether he would consider coding it.)

@barbarabeeton Yes, @DavidCarlisle seemed to be overselling his contribution. :)

@AlanMunn -- To be fair, he did format it in such a way that it's nicely rectangular.

@barbarabeeton It also suits the original xii.tex name a lot better.

@AlanMunn -- You really should hear it sung, in person. But for that, you'd have to be in Providence on the second Monday after Thanksgiving. (Lots more sung in Latin on the same evening, mostly by the audience, and with seasonal readings in Latin, Greek and Sanskrit mostly by people who understand what they're reading.)

@barbarabeeton Don't forget my thesis. youtube.com/watch?v=x8qB-PUw2AQ

@AlanMunn -- Yup. That's them. (But better in person, and it's a little different every year.)

@bp2017 \edef expands its argument, so \z is not defined at that time.

@AlanMunn, but \z is commented in my example and is not "given order/command" to expand, \edef\zz{} doesn't contain \z.

@bp2017 Sorry I meant to type \zz. The inner \zz isn't defined when you do the \edef.

@bp2017 Or maybe I'm misunderstanding the actual code you're running.

@bp2017 You need a 'holder' \zz definition

@DavidCarlisle, @egreg, @UlrikeFischer I'm thinking of moving all the expl3 options to l3sys, as they are 'runtime dependent': make sense?

resulsts in same error

@bp2017 Yup: you have not defined \zz at the point you try to expand it

@DavidCarlisle ooh

@PauloCereda Hey, belated Happy Birthday! I assume you are too far north to have lost power?

@AlanMunn thank you! No worry, it's still in the range. :)

@PauloCereda Where C is in Buenos Aires has lots of hospitals, so power came back first there. :)

@AlanMunn oh really? :)

@PauloCereda Yes, it was back on by around 1:30 local time.

@AlanMunn interesting

@PauloCereda Do you know how far it reached into RS? Was PoA affected?

@AlanMunn let me check, hold on. :)

@AlanMunn Brazil wasn't affected, the report says.

Only Argentina and Uruguay.

@PauloCereda I see. The reports I read said that social media (world's most reliable source) said the south of Brazil was also affected.

@AlanMunn I will check further.

@PauloCereda I suspect that those reports were incorrect. Maybe some places very close to the border got affected.

@AlanMunn Ah:

> O jornal argentino Clarín afirmou que o corte de energia também afetou pontos do sul do Brasil e do Chile. A Companhia Estadual de Energia Elétrica (CEEE) e a RGE, porém, afirmaram que o apagão não teve reflexo no Rio Grande do Sul.

Then I read about a blackout in Porto Alegre, but it had a different cause.

@PauloCereda ok

@AlanMunn I will keep looking. :)

@PauloCereda Don't worry, that sounds definitive.

@JosephWright seems OK to me

@bp2017 I have an answer on nested edef somewhere I'll see if I can find it, there is no primitive and or or but of course you can define macros for them

@bp2017 this was the answer I had in mind

@bp2017 6 5 4 3 (where did that code come from?, the zz gives some clues...)

@bp2017 simpler to write it without the \edef:

\def\z#1%
{%
\ifnum#1<6
\expandafter\z\expandafter{\the\numexpr#1+1}
#1 % value of #1 insize respective \z
\else
#1 % 6
\fi
}
\z{3}

@AlanMunn isn't São Paulo in Argentina? :-)

@DavidCarlisle not a good loop. Less than 5000 recursions make it break already.

@JosephWright looks ok..

@Skillmon would serve you right.

@DavidCarlisle, thank you. It'll take me some time to read and comprehend the answer you linked to. At this time \noexpand inside \edef which expands (and yet doesn't expand?) to outer \def (containing that \edef) reminds me very much of the \show command (and if \noexpand is removed, reminds me of the \showthe command).

@bp2017 not really related to \show. \noexpand\foo just expands to \foo (but the expansion of \foo is suppressed )

@DavidCarlisle Yes, it is. Ok, the official spelling is different: goo.gl/maps/vohAf6GZmVDJ1LMh9

@DavidCarlisle, and yet "operations" inside \foo are "executed" inside \foo with \noexpand. Otherwise it's hard to explain how the loop in example above is being able to iterate. But I haven't read the answer (which was linked) yet (I'll have time to study it in an hour or so).

@bp2017 no, the code gets run after the edef, that's the whole point (although as I say the edef isn't needed there)

@AlexG That explains why @PauloCereda speaks Spanish so well.

@bp2017 given \z{4} the edef defines \zz to be \z{5}space so the immediately following call to \zz effects the recursion.

@bp2017 add \show\zz after the definition, the difference should be obvious then

It's confusing because \noexpand, by its name/meaning, assumes no expansion of \z inside \edef which expands everything), which is same if \z were present inside \def. But then it behaves differently.

@bp2017 as I said, if #1 is 4 then the first is the same as \def\zz{\z{5}} the second is the same as \def\zz{\z{\the\numexpr4+1}} which are not the same at all

@bp2017 no, as I say \noexpand\z just expands to \z but then the expansion of \z within the edef is suppressed, after the definition is completed that is just the normal token \z though, and no record of \noexpand is in the definition.

So why do we need \edef for this particular case then, if \edef doesn't expand \z with \noexpand? (I guess I need to re-read everything and try to make sense of it).

@bp2017 it expands the \the

@DavidCarlisle, oh, so that's what the matter is.

Now it's starting to make sense to me.

@bp2017 as I said above in one case you get \z{5} (as \the was expanded) and inthe other you get \z{\the\numexpr4+1} (because it isn't expanded)

@DavidCarlisle Shouldn't it be \teh? ;-)

@marmot shouldn't you just screech

@DavidCarlisle Well, I did.

@marmot oh auto translation built in to the chat software?

@DavidCarlisle Yes. How would I follow this chat otherwise?

@marmot Ich bevorzuge ein unfehlbares Orakel, keine mechanische Übersetzung.

@DavidCarlisle Google Translate?

@marmot of course

@DavidCarlisle It does not yet have marmot language, so does not work for me.

@marmot important cases first

pdfs.semanticscholar.org/ba6f/…

@marmot I didn't know they had marmots in Sheffield

@DavidCarlisle Yes, one keeps learning new things. Happens to everyone, you are not alone. ;-)

@marmot It's not far from where I grew up. I suspect that they are squirrels not true marmots.

@DavidCarlisle I was hoping they will make it to the LaTeX level, and not get stuck at the Word level.

@marmot they are just being modest: they only mention Word in the title but the internal document properties reveal a tex ancestory

@PhelypeOleinik typos? me? how could that happen?

@DavidCarlisle The Pythia of Oxfordshire?

@UlrikeFischer Is your forest problem solved?

@UlrikeFischer never trust a deity with tex details.

@DavidCarlisle Probably the drives in the stackexchange server got corrupted and messed up your answer ;)

@PhelypeOleinik given that they messed up tens of thousands of \\ in previous times, that is not impossible....

@DavidCarlisle LOL

@DavidCarlisle I wasn't around when that happened, but I've seen some discussions about it.

@PhelypeOleinik they knew how it happened and they had corrupted the back end store but rather than fix it with a global edit they made us fix it through the web interface which throttles access to 5 seconds per edit which mean it took days rather than seconds (and only then after @ShreevatsaR had tuned a javascript access so we could do edits at that rate)

@marmot partly. I decided to "outsource" the title and to use two parallel forest trees:

\documentclass{article}
\usepackage[edges]{forest}
\forestset{%
  mytree/.style={%
    baseline,
    for tree={%
      folder,
      grow'=0,
    },
    },
  }

\usepackage[most]{tcolorbox}
\tcbset{forestbox/.style={
        nobeforeafter,enhanced,
        interior hidden,frame hidden,
        borderline={0.4pt}{0pt}{black},
        coltitle=black,fonttitle=\bfseries,
        box align=center,sharp corners,
        before title=\strut,
        top=-2mm,
        leftupper=6mm}}
\begin{document}
\tcbox[forestbox,title=Title]{%

@UlrikeFischer would have been simpler in picture mode

@DavidCarlisle Ouch. I thought it was one or another post which was messed up. So it was sort of on purpose on their end?

@DavidCarlisle the task was "not lots of draw instructions" (and the tree can grow ...)

@PhelypeOleinik no it was literally tens of thousand of posts.

@PhelypeOleinik it was a messed up regex normalization which changed every \\ to \ in the whole site. But the trick was it messed up the back end stored markdown but not the cached html that is normally displayed, so it only came to light months later as people edited it, then they did a global change changing http to https for images and that made them regenerate every page with an image then all the broken \\ showed but that was a year after the initial error

@DavidCarlisle I think I can see some outliers in that graph...

@DavidCarlisle Seems pretty stupid to do that kind of substitution for no good reason...

@PhelypeOleinik probably random data input error: clean the data and redraw a cleaner curve:-)

Going to eat dinner now (no duck :-)

@PhelypeOleinik maybe you will get some another day:-)

@marmot that makes a star layout from the title to the subtitles. I'm looking for an horizontal folder layout, a bit like this (but the title should be more above the line:

@marmot I think I'm not expressing myself very well ;-). Now the title and the subtitles are perfect, but the abc are not vertical. ;-) I need the top of the second and the bottom of your first solution. Somehow the tree should grow on the first level to the right and then down.