Oh, I'm out of votes.

@DavidCarlisle Sorry, C is almost totally whitespace insensitive, isn't it? (Which is why pointers are such a pain, as some idiot reused * for pointer and multiplication, so if you don't do that perfectly you'll od the wrong one)

hello

@egreg what is the right way to write the characteristic of a field? \operatorname{char}?

any latex expert around? I am trying to find why this expression does not break automatically :

\documentclass{article}%
\begin{document}
\[
a_X+2 \xi  \theta ' \sin (\theta ) \psi ' \cos ^2(\phi )-\xi
   \left(\theta '\right)^2 \sin (\phi ) \cos (\phi )-2 \cos (\theta
   ) \xi ' \psi '+\frac{1}{2} \xi  \sin ^2(\theta ) \left(\psi
   '\right)^2 \sin (2 \phi )-2 \xi ' \phi '+2 r \theta ' \sin
   (\theta ) \psi ' \cos ^2(\phi )-r \left(\theta '\right)^2 \sin
   (\phi ) \cos (\phi )+\frac{1}{2} r \sin ^2(\theta ) \left(\psi

when I compile it, it generate this:

I do not see any \left( that is too long to prevent breaking the line?

The code was auto-generated by another program, FYI

@leo Sorry, too late! If you decide to name it “char” do it, but don't do \DeclareMathOperator\char{char}; name it \Char if you need the operator more than a couple of times.

@Nasser \[ is display math, that never breaks

@DavidCarlisle oh, I see ! thanks. so I should then just use $ ..... $ ? Ok, will try it :)

@DavidCarlisle it worked! You are amazing. Thanks again....

@PauloCereda, tragedy: weather.com/travel/…

@ForkrulAssail :-)

@ForkrulAssail Oops!

Exams, oh, boy!

I found a bug in latex.

I will a report now

@Nasser you sure? :-)

@DavidCarlisle, well, I do not know Latex. But from user point of view it looks like a bug to me. I changed font size from 10 pt to 12 pt, and now the same equation does not break correctly. I am using $.....$ for the same equation. With 12 pts, it overflows to the edge of the page losing many chars

I am making a report now

it has to do with a package I am including. I am trying to find out which package it is....making report...

still working on it, but here is a screen shot: same code, only different is one 10 pt and the other 12 pt, plus the problem package added

Ok, found the problem....

please try this code
\documentclass[12pt]{article}%
\usepackage[left=.7in,right=.7in,top=.9in,bottom=.9in]{geometry}
\begin{document}
$a_y =a_Y-2 \xi  \theta ' \sin (\theta ) \psi ' \sin (\phi ) \cos (\phi
   )-\xi  \left(\theta '\right)^2 \cos ^2(\phi )-\frac{1}{4} \xi
   \cos (2 \theta ) \left(\psi '\right)^2-2 \xi  \cos (\theta ) \psi
   ' \phi '+\frac{1}{2} \xi  \sin ^2(\theta ) \left(\psi '\right)^2
   \cos (2 \phi ) + \xi ''-\frac{3}{4} \xi  \left(\psi '\right)^2 - \xi
   \left(\phi '\right)^2-2 r \theta ' \sin (\theta ) \psi ' \sin

changing 12pt to 10 pts now Latex breaks the equation nice. With 12 pts, it overflows ! This is no good

Latex is supposed to figure where to break the equation so it does not overflow.

@Nasser without looking that's almost certainly not a bug. There are only a few allowable break points in math (no hyphenation, and no breaking of inner terms) so if you are typesetting long generated expressions you want to use \begin{raggedright}$.... (or center but don't try to justify th elines)

@DavidCarlisle, thanks, will try \begin{raggedright}$......$ now

@Nasser It will only do this within the constraints you have set, and if there is no stretchy space on the line and no breakpoint at the right edge, then texs line breaker has been given an infeasible optimisation problem so finds no solution

@Nasser Of course every usage of \left and \right in that code is wrong.

@DavidCarlisle it worked! Here is screen shot, using \raggedright:

\documentclass[12pt]{article}%
\usepackage[left=.7in,right=.7in,top=.9in,bottom=.9in]{geometry}
\begin{document}
\begin{raggedright}$a_y =a_Y-2 \xi  \theta ' \sin (\theta ) \psi ' \sin (\phi ) \cos (\phi
   )-\xi  \left(\theta '\right)^2 \cos ^2(\phi )-\frac{1}{4} \xi
   \cos (2 \theta ) \left(\psi '\right)^2-2 \xi  \cos (\theta ) \psi
   ' \phi '+\frac{1}{2} \xi  \sin ^2(\theta ) \left(\psi '\right)^2
   \cos (2 \phi ) + \xi ''-\frac{3}{4} \xi  \left(\psi '\right)^2 - \xi
   \left(\phi '\right)^2-2 r \theta ' \sin (\theta ) \psi ' \sin

@egreg thanks. But as I said, this is auto-generated code from Mathematica. It is long equation and I did not want to type it myself, so used Mathematica TeXFrom[] command

@egreg here is the command fyi:

@Nasser They're wrong.

All CAS systems support for Latex is not good. Maple does not do well either. But have no other option.

@Nasser Compare the output of $\xi\left(\phi'\right)^2$ with $\xi(\phi')^2$ and see what's good and what's bad.

\documentclass[12pt]{article}%
\usepackage[left=.7in,right=.7in,top=.9in,bottom=.9in]{geometry}
\usepackage{enumitem,lipsum}
\newenvironment{autobreakeq}
 {\begin{itemize}[label={},leftmargin=1em,itemindent=-1em]\raggedright
  \item\relax$}
 {$\end{itemize}}
\begin{document}
\lipsum*[2]
\begin{autobreakeq}
a_y =a_Y-2 \xi  \theta ' \sin (\theta ) \psi ' \sin (\phi ) \cos (\phi
   )-\xi  (\theta ')^2 \cos ^2(\phi )-\frac{1}{4} \xi
   \cos (2 \theta ) (\psi ')^2-2 \xi  \cos (\theta ) \psi
   ' \phi '+\frac{1}{2} \xi  \sin ^2(\theta ) (\psi ')^2

@Nasser don't sound so surprised that my suggestions work:-)

@Nasser You may also adjust the spacing before and after the equation block.

@Nasser Can you see the difference? Above=bad.

@egreg Yes. I agree.

Scientific word also generate \left ...\right for everything !

makes the code hard to read also when done for everthing

@egreg thanks for your \begin{autobreakeq}, it is very nice. I added it to my common latex file and using it and it is working very well.

@DavidCarlisle I found a solution to my binary tree induction problem of yesterday. Now I just need to clean it up and post an answer here. (Is it good or bad to write your own answer?). Many thanks for your suggestions!

@Egbert Nothing wrong with answering your own question.

@TorbjørnT. Cool! I'll do that then (:

@Egbert self answer is fine

Is it easy to have a macro with a syntax like \macro[option1=x,option2=y,], similar as in tikz?

@Egbert With a key-value package. See tex.stackexchange.com/questions/26771/…

@egreg Great, thanks! I see in the manual of one of them that there's also a guide to key-value methods. I'll have a look at that.

When you start a bounty is there a timeline for accepting it?

I mean, awarding it to an answer?

@ForkrulAssail Yes, seven days, IIRC

@egreg @DavidCarlisle, do you think this question is worth answering, interesting? Should I put a bounty on it, or am I wasting my time as Tobi suggested?

It's not all that important to me, it's just that I found it to be an interesting problem.

@ForkrulAssail I think Tobi's answers tell everything.

@egreg, thanks. How do I close then?

Accepted Tobi's latest.

@egreg Thanks. Exactly. I was thinking in name it \char, as I need it in many places. Why not \char

@leo \char is a tex primitive, if you redefine it, things will go wrong

@DavidCarlisle I just did it yesterday. I didn't get problems, but I'll fixit

@leo Wait until LaTeX3 is officially released, when primitive names will not be reserved any more. ;-) For the moment avoid clobbering primitive names.

@leo Many documents wouldn't be affected by redefining \char; some will and the error message would be funny.

@egreg I see. How do LaTeX3 handle that? Approximately, when will it be released?

ha ha :-)

fixed now

@leo Ask @DavidCarlisle when he'll finish xor. Then a release date might be tentatively given.

Must go. Thanks

Hi folks, how to set a newif macro via an argument of edef?

\newif\ifAutoRotatePages

\edef\IsAutoRotatePages#1{%
	\AutoRotatePages#1%
	\ifAutoRotatePages
		% do nothing
	\else
		-dAutoRotatePages=/None
	\fi}

@Marienplatz you can't make an assignment inside an edef (why do you need too?)

@DavidCarlisle Please see my real scenario below.

% this input file name is filename.tex
% compile it with pdflatex -shell-escape filename.tex

\documentclass[preview,border=12pt]{standalone}

\usepackage{filecontents}

\begin{filecontents*}{temporary.tex}
\documentclass[pstricks,border=12pt]{standalone}
\begin{document}
\begin{pspicture}(4,2)
	\rput{90}(2,1){Marienplatz}
\end{pspicture}
\end{document}
\end{filecontents*}

\usepackage{graphicx,pgffor}

\newif\ifAutoRotatePages

\edef\IsAutoRotatePages#1{%
	\AutoRotatePages#1%
	\ifAutoRotatePages

@Marienplatz Why should \edef be needed here?

@egreg Because I just modified the existing code in my previous question.

@Marienplatz Also, \edef\JobName#1{-jobname=#1} does nothing different than using \def

@Marienplatz An \edef there skips the true branch, because the conditional is false. You seem to believe that \edef does the expansion when the macro is called, but you're wrong: the expansion is performed at definition time.

@egreg OK. Then, what is the solution to make the code works? :-)

@Marienplatz It would be useful to know what the code is supposed to do.

@egreg The code should compile the temporary.tex with 2 different jobnames. The first one is named Rotated and the other one is named Unrotated. The Rotated.ps will be passed to ps2pdf without -dAutoRotatepages=/None, but the Unrotated.ps will be passed to ps2pdf with that switch.

Crystal ball questions seem especially popular today: tex.stackexchange.com/q/142414/2693 and tex.stackexchange.com/q/142364/2693.

@Marienplatz You can't do assignments during a \write.

@egreg So is it really impossible?

@Marienplatz it will be possible but it isn't clear what it is

@egreg @DavidCarlisle Do either of you know of an answer to the comment about why/how the particular sizes corresponding to the fontsize commands were chosen? tex.stackexchange.com/a/24600/2693

@DavidCarlisle OK. I need fewer keystrokes as usual.

@AlanMunn well more or less increasing sequence with an overlap to \magstep scaled fonts in the original cm distribution. So usual reasons: history, accident and luck:-)

@AlanMunn I'll add a comment

@DavidCarlisle Thanks. When the answer is basically "because", argument by authority comes in handy. :)

2 more votes for my aircraft today! (a couple more I'll get a badge:-)

@Marienplatz It's wrong to try TeXing \outputname.\ext in the first place, because the file Rotated.tex will not exist. So the TeXing phase must be taken care differently.

@egreg Good point!

\@percentchar is for %, what macro is for #? \@sharpchar does not exists I think.

@Marienplatz I don't think there is an equivalent for #. You can define something via (untested) {\catcode`\#=12\gdef\@sharpchar{#}} or \edef\@sharpchar{\expandafter\@gobble\string\#}

@NicolaTalbot Here is a good place for your suggestion: tex.stackexchange.com/questions/142428/… ;-)

@MarcoDaniel Thanks. I hadn't seen that, but I see @DavidCarlisle has got there before me :-)

@NicolaTalbot sorry I wasn't "here" only on the main site:-)

@DavidCarlisle Your solution is simpler than mine :-)

@NicolaTalbot of course: I'm not @egreg, I like simple solutions:-)

@NicolaTalbot Thanks for responding!

@Marienplatz :-)

@NicolaTalbot \begingroup\catcode\#=12\def\x{\endgroup\def\hashmark{#}}\x` (just joking)

@egreg lol :-)

@NicolaTalbot I'd say that \edef\hashmark{\string#} is more economic.

@egreg Perhaps we should have an obfuscated TeX competition (although it would be a challenge to beat @DavidCarlisle's xii.tex) :-)

I had my glass of wine on the train! :)

What modification do I have to do to make the \compileA able to receive \filename? See my code below.

\documentclass[preview,border=12pt]{standalone}

\usepackage{filecontents}

\begin{filecontents*}{template.tex}
\documentclass[preview,border=12pt]{standalone}
\usepackage{xcolor}
\begin{document}
\textcolor{\clr}{Hello World}
\end{document}
\end{filecontents*}

\def\filename{frames}


% Compile A
\def\CompileA#1#2{%
\immediate\write18{\unexpanded{pdflatex -jobname=#1 "\def\clr{#2} \input{template}"}}}
%\CompileA{\filename}{blue}% not working
\CompileA{frames}{blue}% working


% Compile B
\def\CompileB#1#2{%

I want to use \unexpanded as a replacement for some \noexpand's to save more keystrokes.

@Marienplatz If you put #1 in the scope of \unexpanded, then \filename will not be expanded. Why not \immediate\write18{pdflatex -jobname=#1 \unexpanded{"\def\clr{#2}\input{template}"}}?

@egreg Good point! Thank you!

@NicolaTalbot Can you imagine any purpose for indexing a beamer presentation?

@egreg I suppose if the pdf is given out to students then they can look stuff up from the index or the handouts may have more extensive text than the screen version.

@NicolaTalbot Handing the presentation to the students is something I can't even think to. Unfortunately people do it.

@egreg Unfortunately some universities insist on it. :-(

@NicolaTalbot It can't be the main material.

@egreg When I apply it to my another project, it fails. Please see the following, it is the simpler one.

\documentclass[preview,border=12pt]{standalone}

\usepackage{filecontents}
\begin{filecontents*}{template.tex}
\documentclass[preview,border=12pt]{standalone}
\begin{document}
Hello \name
\end{document}
\end{filecontents*}

\usepackage{pgffor,graphicx}

\def\Compile#1#2{\immediate\write18{pdflatex -jobname=#1 \unexpanded{"\def\name{#2} \input{template}"}}}

\foreach \outputfilename/\target in {a/Andy,b/Bill,c/Charlie}{\Compile{\outputfilename}{\target}}

\begin{document}
The files we created automatically are:

@egreg I think it depends on the type and length of the course. At the UK TUG training days we just give them handouts. However when I was an undergraduate I found it easier to revise courses where I had to write notes than the ones where we were given a handout. At least when you write the notes as you listen in class the information gets one pass through the brain.

@Surijo: Please don't just review for the sake of reviewing (AKA: robo-review). I read that in your most recent review where the "answer" is really just a bunch of unrelated things (in my opinion).

You must be a machine (literally) in order to be able to review two posts 4 seconds apart...

...in related news:

If anyone wants to test out my new glossaries version 3.13a, the tds can be fetched from my experiment page. It's incrementally working it's way towards a stable release, but it could do with a bit more backward-compatibility testing.

(I'd rather not have my inbox flooded with messages from panic-stricken students who can no longer compile their thesis after upgrading!)

@PauloCereda I've switched back to the black text. It occurred to me that most people will see the thumbnail before seeing the large version and the black font stands out better than the brown in the thumbnail, but thanks for the suggestion :-)

@egreg: Is your question here:

...just that the question be re-opened? If so, it has been. If not, then what is the question? :)

Is there an easy way to make certain subsections disappear without just commenting out the whole subsection?

I was lazy and decided not to use a Question package, and no someone wants to give my problem set a try

@Canageek \iffalse \section{...} .... \fi

@DavidCarlisle Thanks

@Werner wow so @egreg can find the ask a question button?

@DavidCarlisle :) It doesn't exist on the main site, for some reason.

Mac users: when I try to run gs from the command line, I get a dialogue box from the OS (Mavericks) to install X11. This takes me to a page to install XQuartz, which I have already done. What am I doing wrong?

@AlanMunn Not using the one true operating system from Billl Gates?

@DavidCarlisle Oh don't you start. I have enough problems with @Canageek. :)

@AlanMunn I use it but am in denial as I use everything through cygwin and pretend I'm on linux:-)

@DavidCarlisle Every so often I have to work on a Word document, and it's getting harder and harder. Unfortunately there's still no good replacement for Excel.

Mac users: problem solved. I must not have logged out after installing XQuartz.

@AlanMunn fortunately I've only ever had to use a word processor as a document reader I've never had to make a document in such a thing. I once had to make a powerpoint presentation that was enough (and I spent most of my talk moaning about it, serves the organisers right for insisting on that format:-)

@Werner The discussion is about how we should behave for “symbol requests”, not about the single case.

@Werner... like this?

@Sujiro So, is that from when there was suddenly a bunch of questions in the reopen queue? Because that was an unusual case.

@TorbjørnT. Hi man. I read the questions before they go to the queue (First Post or Late Answers). Please, use "Questions" "newest" or "active"... Also, I think @Werner should give more importance to their own life :-)

@Sujiro You are basically trying the patience of goodwilled community here. Just see what the message for once.

@percusse what you think I should to do?

Read the fucking messages and get the point instead of playing hide-and-seek

It's been months already

@percusse take easy man

@Sujiro Don't call people to get a life then. You set the threshold here

Holy patience, Batman! Someone called me "Bot"... Someone thinl I am subverting the review process because he/she is not at the top of the list...

Again read the comment before you tell people what to do. It was a joke I was here.

I am wasting my time trying to responde politely to rude comments from those people just because they want to make themselves feel better or because they want to looks better for someone else!

I don't care what fun you get out of this review button fetish but other people do it for making the place better

My appologies to the other members of this site...

You do it for a strange unknown reason

@percusse Have a good time

@Sujiro Honestly, I haven't looked at any of your reviews, so I'm not criticising them. I was just pointing out that it was an unfair comparison. As long as you look carefully at each question when reviewing them, and take any action necessary, I'm not complaining.

@Sujiro I sincerely doubt that's Werner's motivation. Anyway, maybe you should just try to be a bit more measured as you review. Finding anomalies in others' reviews (especially someone who is one of our top users) is unlikely to be a good strategy for discussion. (Of course, @percusse overreacted a bit to.)

@Sujiro I do. You keep your cool and don't you ever insult others.

@percusse I am not insulting no one.

Also, I think @Werner should give more importance to their own life :-) is worse than a simple joke calling you a bot

And what is this ?

I am subverting the review process because he/she is not at the top of the list...

Are you joking me ?

Do you think anybody cares?

@percusse You are the famous who?

@Sujiro The famous nerd hater.

@percuse are you sure?

Pretty much

hehehe

@percusse If you like, and if it is possible, I shoul to transfer my points to you...

@percusse Self-hatred is big around here.

@Sujiro I hope you get the message. I'll leave you to your witty comebacks

@percusse English is not my mother language. I am using Google Translate to discuss with you, man.

Translate this to your own language then, Don't insult anybody!

or anyone

@percusse Sincerely, I hope that you leave in peace. If I am doing things the wrong way, I think the moderators should do their job.

I don't know whichever is the right one.

@percusse I do not insult anybody here.

@Sujiro You have a very good translator. Moderator job is not exactly chasing you after

@Sujiro I gave you examples above. Use your eyes

@percusse you are wrong man... I am just have a reation to someone that do not respect me.

@Sujiro Although the moderators have some extra power, the site is designed so that all users help to moderate it. That's why those of us with > 20,000 rep are called "trusted users". If someone with > 100k rep makes a suggestion about your reviewing, you should take it as if a mod said the same thing.

Although reputation is not really serious, it is a fairly good indicator of participation on the site, and with that participation comes quite a bit of knowledge.

@Sujiro Yes, that's an unusual case when a large number of historically post-close edited question made its way into the Re-open review queue.

@AlanMunn I understand your point of view.

@Sujiro Again. I don't need to play your game. You get the message pretty much. There is no diverting attention for you. I give you the cases above. Do your button clicking thing but don't insult anyone. Put this into your translator and print the result in case we need it again.

@percusse I am not playing your game. I need to do a picture to you?

@sujiro @percusse Enough. Let's just let this go, ok?

I was accused of subvert the review route queue

@Sujiro I'm just one of the community members here, and just thought I'd mention what I see. I meant no offence, or try to undermine your reviewing...

...if you were in the area, I would even take you out for a beer. A free one. Really.

@Sujiro And? So are other users. Every so often, users who newly gain reviewing privileges seem to go on a reviewing spree. This is not always a good thing, and @Werner (who does a lot of reviewing) sometimes points it out to make such users aware of what they are doing. This is all part of making the site better, and it's not aimed personally at you.

@AlanMunn I second that...

@AlanMunn No problem. Sorry.

Meanwhile normal service is being resumed: England v Australia A starts tomorrow, I'm sure you'll all be watching this channel for updates.

A prime candidate for one of @DavidCarlisle 's picture mode answers. :)

@DavidCarlisle It's somehow not the same without Psmith, though.

@PauloCereda get to work and fix ^^^^^

@DavidCarlisle We'll add that advice to the one from Brent.

Is there a really easy way to specify the text area size in LaTeX? I don't really mind about the page size, it should just fit the text area. Ideally I'd just like to specify that I need a text area of width 25cm and it would give me a PDF of width ~25 cm, text running side to side, as tall as necessary to fit all text (and never broken across pages.)

@Szabolcs tex.stackexchange.com/questions/27041/…?

@Szabolcs \textwidth=25cm \textheight=\maxdimen (and possibly set oddsidemargin to 0pt and hoffset to -1in)

Sorry, I was called away for a while.

I'm trying to make a poster, make the pieces in TeX, then put them together in some vector illustration program.

@Szabolcs in that case you probably want to avoid the page breaker alltogether just \shipout\vbox{first bit}\shipout\vbox{secondbit} then you get each bit on a separate page just with the single box content and then massage the pages into your application

@DavidCarlisle the \textheight=\maxdimen part gives me a Huge page cannot be shipped out error. This is the simple document I tried:

pastebin.com/qUurQ7Lp

I.e. as basic as it gets with those two lines at top.

Is this something easy to fix ... ?

Does anyone know if I can insert an HTML table into an answer?

What does \maxdimen mean exactly?

@AlanMunn You could try unicode tables.

╔═══╦════════════╦═════════════╗
║   ║ A          ║ B           ║
╠═══╬════════════╬═════════════╣
║ 1 ║ Type       ║ Unicode Art ║
║ 2 ║ Header     ║ Spreadsheet ║
║ 3 ║ Auto Align ║ Off         ║
╚═══╩════════════╩═════════════╝

@Kasper It's ok. Thanks. I've just used <pre> ... </pre> Of course, unicode would be your solution. :)

I have got to keep my image alive ;)

I think I got a solution, based on this: texblog.org/tag/fit-page-to-content

\documentclass[varwidth=true]{standalone}
\textwidth=10cm
\begin{document}

Actually tables like that would be my really last option. I just copied this from stackoverflow.

@Szabolcs Also this: tex.stackexchange.com/a/19721/4427

hi @90intuition

hi all

Are there any Ghostscript whizzes around?

@egreg Thanks!

I think I found your markdown for math language. Peter Jipsen made something like that a couple of years ago: www1.chapman.edu/~jipsen/asciimath.html

Or this is the new link I think: www1.chapman.edu/~jipsen/mathml/asciimath.html

That seems quite complete.

Can I get pdflatex to embed a complete font (however large), and not subset it?

Thanks @Kasper I will look at it.

@Szabolcs yes maybe .5\maxdimen might be better but actually once I saw you use case you are better off shipping out boxes and not using \textheight at all

Got it here: tex.stackexchange.com/questions/24002/…

@Kasper rather more than a couple of years ago:-)

@90intuition mathjax is a lot more complete (but a lot slower) than ascimath, it depends what you need.

@DavidCarlisle You can use asciimath also with mathjax output (so that it is compatible with IE and chrome): docs.mathjax.org/en/latest/asciimath.html