@DavidCarlisle -- the problem with figure is that it has its own ideas about where to put something on a page (and not always the current page). but if you know where you want something (e.g. a tabular) and there's room for it where you input it, then there's not always a good existing latex environment.

@santimirandarp -- @DavidCarlisle's answer that \par is needed is correct. if it's not there, and the passage you want to be centered is just enclosed by braces, it won't be centered, because the decision on things like centering and baselines isn't made until a paragraph break occurs. this is a feature of tex itself, not altered by latex.

@DavidCarlisle Consider this: after I changed my profile pic to TikZ, I was in every month the reputation charts before @egreg. This pictures make one look much more competent than one actually is. Does that sound convincing? ;-)

@marmot egreg could use that help, I agree, I'm sure I look very competent already:-) Good night:-)

@DavidCarlisle How could he be in the charts before himself? ;-)

Hi

I have a latex file, in which a very particular page, which is not so important to me every time I compile code for that file. But, in every other latex files that I have, front page opens, as i expect. Can I solve this issue?

Any one knows what happened to micropress-inc.com and if there is any other way to get their math fonts (namely micropress-inc.com/fonts/ifmath/ifmain.htm). An acquaintance of mine is trying to contact them but no luck with email or phone.

@marmot Please do NOT answer this tex.stackexchange.com/questions/459560/…

@KhaledHosny someone asked on tex list the other day and got an answer pointing to Michael Vulis' academic address tug.org/pipermail/texhax/2018-November/023338.html

@DavidCarlisle Well, that is the same person I’m asking for :) so I guess he had no luck with that either.

@KhaledHosny I thought it might be:-)

I'd like to put some big bold red text right on top of a verbatim environment. Any ideas? :(

@NaCl Put it before the verbatim environment, perhaps?

@HaraldHanche-Olsen No, I mean it should overlay the verbatim

@NaCl Oh. Well, that's different. Are you in a situation where you are sure there won't be a page break in the verbatim? That would complicate things.

The verbatim fits on one beamer frame

I see, there is no rather simple solution, is there?

@NaCl If a bit of manual fiddling is acceptable, put the command to generate the overlay text after the verbatim environment. Something like \raisebox{48pt}[0pt][0pt]{\makebox[0pt][l]{\huge\textcolor{red}{foo!}}}. If you need to measure the height of the verbatim, more work is needed.

@NaCl easy:

\documentclass[]{beamer}

\begin{document}
\begin{frame}[fragile]

\begin{verbatim}
abc\\
abc\\
abc\\
\end{verbatim}

\vspace{-3\baselineskip}
\textcolor{red}{on verbatim}
\end{frame}


\end{document}

@UlrikeFischer So long as you don't want other content to appear after the verbatim on the same frame, this will work well. Otherwise, there is the question of adding the right amount of positive space to get back below the verbatim for the continuation.

oh, thanks!

Need to play around with the colour, though, as it appears pink. Might be because of pygmentize

@HaraldHanche-Olsen sure. but without context or mwe it is difficult to know what would be the best solution so you can only offer some options.

@UlrikeFischer No complaints from me; we both offered ideas, he can pick and choose according to his needs.

You both helped me well :-)

@cis I will take a closer look at the arara questions today. :)

@DavidCarlisle oh no

@PauloCereda seriously though, can you see if that is a real option, or is it just a typo by the user using an accented form of the option name? Because if it's a real option and we broke it, we should at least contact the maintainers and work out a plan.

@cfr I took a look at your code. I do find it quite daring to do a tikzexternalize in the middle of a \AddThispageHook ;-).

@MarcelKrüger could you change the travis notification settings in your latex2e fork? Currently it's mailing us twice (probably we should obscure the email in the travis config file but....

@DavidCarlisle I checked yesterday in the whole latex tree and found no mention of this option.

@UlrikeFischer yes seems odd though, the OP didn't answer my query, perhaps hasn't been back to the site having got an answer.

@DavidCarlisle Roger.

@DavidCarlisle: IIRC the document probably lacks \usepackage[T1]{fontenc}

And the option does not exist.

I've seen it happening before. :)

@DavidCarlisle my guess is that the correct answer is "delete the option" ...

@UlrikeFischer yes. :)

And sometimes people forget to use fontenc

Performance question: Is \begingroup\expandafter\expandafter\expandafter\endgroup\expandafter\ifx\csname #1\endcsname\relax\expandafter\@secondoftwo\else\expandafter\@firstoftwo\fi faster or \begingroup\expandafter\ifx\csname #1\endcsname\relax\endgroup\expandafter\@secondoftwo\else\endgroup\expandafter\@‌​firstoftwo\fi?

@Skillmon use l3benchmark to test ;-).

@Skillmon you could use \ifcsname rather than \ifx...\relax these days (unless you specifically need to test for \relax)

@cfr Cwac! :)

@DavidCarlisle I would, but it seems more stable to use \ifx tests

@PauloCereda are you learning a foreign language?

@Skillmon me? I am a humble duck. :)

@Skillmon that's Welsh! :)

@PauloCereda ahh :)

@UlrikeFischer where can I find the syntax?

@UlrikeFischer nevermind, found it

Turns out version 2 is a full 1e-7 second faster than version 1.

@cfr in Royaume Uni.

at a university that bears the name alluding to the vestiges of a long-gone 'empire'

And a simple \ifcsname #1\endcsname\expandafter\@firstoftwo\else\expandafter\@secondoftwo\fi is again 1.9e-7 seconds faster than version 2 of the above.

I hate the fact that l3benchmark exists. Now I benchmark everything...

@Skillmon Er ... cool?

@Skillmon Not surprising

@JosephWright of course not, I just wanted to check how much faster it is.

@Skillmon ooh profiling

@JosephWright still great that it exists, because now I can benchmark everything :)

@DavidCarlisle Sorry for that, I didn't expect the address to be set in the travis config. I am trying to change it now.

@MarcelKrüger it seems to be the only way to have an email other than the registered github account be notified, although @UlrikeFischer has pointed out that we could encrypt it (so it would simply fail in forked repros)

@Skillmon Maybe version 2 is faster, but version 1 is cleaner.

@egreg Do you think so? Because there is only one \endgroup?!

@Skillmon Yes

@Skillmon and both of them are non expandable as the price you pay for avoiding \ifcsname

@DavidCarlisle but both of them don't fail if anybody messed with the macros using them in \csname without testing whether they exist...

Hey guys, I was just wondering: is it possible to know in which column a label is placed when inside multicols?

Just out of curiosity.

'ello, Psmith is now fully operational.

@PauloCereda knowing in which column you are in a table is not easily possible (as far as I know).

/say

Psmith had predicted, the change of policy had the mumps, for four years, and I've never met any of them

@Skillmon oh I see

/8ball is it possible to know which column I am?

@PauloCereda Sure.

oh

@PauloCereda there is a \docolaction command which records the column, so with some work I would say yes.

@PauloCereda you could hack it into it by using array and >{def\curcol{1}}, ...

@Skillmon @PauloCereda asked about multicols not multicolumn.

@UlrikeFischer Oh, I'm stupid

@UlrikeFischer Ah interesting! I was just wondering, I think this is a typical case in which a manual tuning is better. :)

/love

@PauloCereda It is sort of possible, I once wrote an answer tackling this. Let me search

@Skillmon loves me!

@Skillmon oh no worries, it was just curiosity.

/duck

Holy cow

@PauloCereda tex.stackexchange.com/questions/414555/…

@Skillmon Thank you!

@PauloCereda you're welcome

@PauloCereda If you don't intend to write those numbers to the toc much of that answer isn't necessary for you.

@Skillmon I like to learn. :)

/say

We are much obliged to Comrade Jarvis

/fortune

Your fortune cookie: How much deep would the ocean be without sponges?

ooh

ooh (5326 hits so far)

@Psmith but I helped, not Jarvis :(

@Skillmon 'ello, I bid you good day! Type /help to see all my commands.

/shutdown

Initiating shutdown sequence. Cheerio!

'ello, Psmith is now fully operational.

Hi @Psmith!

@DavidCarlisle how should <code>\dimexpr0#1\relax=#1</code> is a building occupied by more than anything else the preliminary sniffing snarl a bull-dog emits before he joins battle

/shutdown

Initiating shutdown sequence. Cheerio!

@Psmith hello!

'ello, Psmith is now fully operational.

@Psmith hi!

That was Kid Brady

@Skillmon you could use \@ifundefined which uses some brilliant combination of ifcsname and testing against \relax so it covers that case while still being expandable

@Psmith ^^

@DavidCarlisle \ifcsname #1\endcsname\expandafter\ifx\csname #1\endcsname\relax test?

Hm why won't you reply...

@Psmith hi?

These petty matters of etiquette are not that bad. The differences are in there, too, in the plains, combined the toughest of muscle with the g and I’d expect that to give himself away

@Psmith this looks great, doesn't it.

\def\@ifundefined#1{%
  \ifcsname#1\endcsname\@ifundefin@d@i\else\@ifundefin@d@ii\fi{#1}}
\long\def\@ifundefin@d@i#1\fi#2{\fi
  \expandafter\ifx\csname #2\endcsname\relax
    \@ifundefin@d@ii
  \fi
  \@secondoftwo}
\long\def\@ifundefin@d@ii\fi#1#2#3{\fi #2}

@PauloCereda he answered before your last shutdown.

@Skillmon ^^

@Psmith oh I think he is ignoring stuff.

/shutdown

@DavidCarlisle as I said...

'ello, Psmith is now fully operational.

@Psmith how about now?

The stranger raised a long neck and large pince-nez, and he protected them

@Skillmon yes but no `\expandafter either

@Psmith good!

texdef --tex latex \LaTeX

LOL

@Psmith you are a good bot

I may manage to pull through

@DavidCarlisle there are \expandafters. (well there is one)

/learn shutdwn Irritating shutdown sequence. Cheerio!

@DavidCarlisle Saved.

/shutdwn

Irritating shutdown sequence. Cheerio!

WAT

@DavidCarlisle is the \@ifundefin@d@ii construct that much faster than \expandafter\@firstoftwo?!

@Psmith hello?

Dere was two fellers in de street sickin' a dawg on to his satisfaction, hitched up the paper's contents, was a few more bars, then nodded to Psmith, has received this extraordinary communication from a dog on to us with still more breezy frankness

@DavidCarlisle you are mean

oh

@Skillmon you are the one with a l3benchmark setp running:-)

@DavidCarlisle :)

/say

@DavidCarlisle Jonathan Kew knows all this means in term of pdf markup.

hm

/say

Comrade Windsor knows certain stout fellows, reporters on other papers, groaned under the portrait the legend, Next Week! See Editorial! and compiling in conjunction a snappy editorial, setting forth the proposed changes

/marmotnews

marmots.org/category/news-from-the-field

ooh marmot news

ooh (5327 hits so far)

/arara

:(

@DavidCarlisle It's 1.5e-7 seconds faster than using \ifcsname#1\endcsname\expandafter\ifx\csname #1\endcsname\relax\expandafter\expandafter\expandafter\@firstoftwo...

@Skillmon You have to become an arara-maniac like me and @clemens...

@cis nah, I'm more the GNU make guy...

@cis Yes I will write an answer. :) Just gimme a couple of minutes, I am eating chocolate cookies. :)

@PauloCereda If the arara-rules will not work, you will never get chocolate cookies any more!

@cis Do not worry, I know my tool. :D

@PauloCereda Haha, very good... I read the answers...

@PauloCereda but can you spell it backwards?

@DavidCarlisle ooh a conundrum

ooh (5328 hits so far)

@cis :)

@DavidCarlisle miktex updates to luatex 1.09 - now probably all tests will fail anyway ;-).

/shutdown

Initiating shutdown sequence. Cheerio!

@DavidCarlisle I'm not sure if you really want to see the results ;-). E.g. from the first with pdftex:

***** .\BUILD\TEST\accsupp-test1.tlg
   28:  Package: kvoptions ....-..-.. v... Key value format for package options (HO)
   29:  (../keyval.sty
   30:  Package: keyval ....-..-.. v... key=value parser (DPC)
***** .\BUILD\TEST\ACCSUPP-TEST1.PDFTEX.LOG
   28:  Package: kvoptions ....-..-.. v... Key value format for package options (HO)
   29:  (D:\MiKTeX2.9\tex\latex\graphics\keyval.sty
   30:  Package: keyval ....-..-.. v... key=value parser (DPC)
*****

***** .\BUILD\TEST\accsupp-test1.tlg
   31:  \KV@toks@=\toks...

@JosephWright ^^^^ it looks as if pathes with slashes are not removed correctly.

@UlrikeFischer boo why doesn't miktex report the paths in a style that tex can use, with / ?

@DavidCarlisle why doesn't texlive report the pathes in a style the windows explorer can use? ;-) (Joking aside: I really have a small script in my editor which replaces slashes by backslashes in a selection to be able to open texlive pathes from the log).

@UlrikeFischer it would be easy enough to normalise \ in the logs in l3build-check.lua I suppose main thing to decide if just to always normalize both ie replace / by [/\] or to have some variable which is / or \ and build the patterns using that. I suppose the latter would be safer in some ways, although would make it more likely to have system dependent tlg, as a \typeout that happened to have a \ or a / might get normalized accidentally as a path in some systems and not others.

@JosephWright ^^

@PauloCereda My coolest arara-rule was (was...) zip.yaml with password and so on... ( matheplanet.de/matheplanet/nuke/html/… ) - but nobody used it.

@DavidCarlisle Just to make things more complicated, that is the output of lualatex: D:/MiKTeX2.9/tex/latex/base/article.cls. which means one can't use luatex to decide about the slashes ...

@UlrikeFischer, @DavidCarlisle Hmm, OK, I'll take a look

@JosephWright but it is not so pressing. The main test suite is done with texlive. I naturally check with miktex too, but at least for luaotfload I don't have many failures due to this differences.

@DavidCarlisle latex.org/forum/viewtopic.php?p=107381#p107381 It goes on.

@cis Cool! I didn't know about it!

And my German skills are terrible. :)

@Johannes_B LaTeX itself could intervene and issue an error message er I suspect that he does not realise how latex is implemented. LaTex can not "intervene" here, the only option would be a major rewrite so that the error is not possible, you can't trap a tex error and return a different error message.

@DavidCarlisle Don't tell me :-)

@Johannes_B I wonder if he means the same x as we mean (a lot people think that the x in a window is meant).

@DavidCarlisle Are the interaction possibilities documented on the LaTeX side anywhere? I couldn't think of an example.

@UlrikeFischer You mean the read button as TeXworks has it? Not sure.

How can a question have a close vote but not appear on the review queue? tex.stackexchange.com/q/459649/2693

@AlanMunn I just reviewed it from the queue, so it is there…

@CarLaTeX Of course not. ;-)

RIP Stan Lee.

@egreg @DavidCarlisle @UlrikeFischer latex.org/forum/viewtopic.php?p=107444#p107444

@Johannes_B I don't have an account so can't comment there:-)

@Johannes_B latex companion or texdoc fileerr for example

@DavidCarlisle @JosephWright Is a mod, he could give you his account details. He is a team member too :-p

@Johannes_B "Since it will be first parsed by LaTeX and LaTeX can know that TeX will issue an error message," No, it is not first parsed by latex and no latex doesn't know an error will be raised.

@DavidCarlisle First time i ever saw this document :-)

@DavidCarlisle I know.

@Johannes_B if you type x or s to a "missing file" error and tex stops rather than goes into an infinite loop, you have used that file though.

@DavidCarlisle I knew about the existence of those files, but never checked what is inside. And i never read the documenation ... until two minutes ago.

@Johannes_B one of the earliest parts of the 2e work, that set of files.

@marmot Let's see if the OP will add an MWE, eventually.

@egreg 2013 :( shouldn't we try to get that fp-exp package fixed? (although I realise you need a source of % questions:-)

@DavidCarlisle Huh?

@JosephWright see

@JosephWright egreg pointed out in an answer 5 years ago that fp-exp.sty is missing a % at eol, and it's still missing

@DavidCarlisle Bugger

@egreg did you know if email to the documented address works?

@DavidCarlisle We might try sending email and if not working, ask Karl to just fix the missing %.

@DavidCarlisle I'd be wary of adopting fp

@egreg can we make it a one line package that inputs l3fp?

@egreg I don't mind sending a mail, but wanted to check if one had already been sent

@DavidCarlisle The syntax is somewhat different. ;-)

@egreg we leave that as an exercise for the user

@DavidCarlisle :) I'll try sending email

@egreg Ok, thanks, ctan lists two: [email protected]; [email protected]

@DavidCarlisle Mail sent also to you in Bcc

@egreg yep thanks

@DavidCarlisle @egreg I think there is another problem in fp. It constantly does (e.g. in fp-basic \edef\FP@tmp{#1} and then \global\let\FP@tmp\FP@tmp. Imho that's why the save stack explodes in tex.stackexchange.com/q/459626/2388.

@Johannes_B I don't think that this stille-post-communicate is really sensible. If he wants to discuss with latex team members he should go where they are and not use you as talebearer.

@UlrikeFischer Probably. But that would be quite a huge task.

@UlrikeFischer There are 51 occurrences of \global\let\FP@tmp...

@egreg and it is imho not only this command.

@UlrikeFischer 57 occurrences of \global\let and 77 of \global alone

Good evening, hi @UlrikeFischer hi @egreg

@Sebastiano Ciao. The problem with 1/2 is only solvable manually: 1\mspace{-1mu}/2 (find out the right value for the backspacing.

@egreg and for line of fraction? I don't want to put any more add-on packages in my code, there are already many of them. I've had an answer but i not like it. I would be delighted to have your answer.

@Sebastiano That's right how it is, in my opinion.

Is there a neat fail-proof way of testing for a digit. \ifnum\number`#1>47 and \ifnum\number`#1<58 gets "tricked" by \0 for example. I'm currently doing 10 \ifx tests, but that feels clunky.

@egreg I'm trying.

@UlrikeFischer I didn't understand what you were talking about.

@Sebastiano sorry I wanted to answer egreg, but my smartphone selected the wrong message.

@UlrikeFischer Don't worry

@Skillmon Hi, and good LaTeX.

@Sebastiano good evening.

@egreg ...but 1/2 it is very big compared to T. I should use \scriptscriptstyle? Or other?

@Sebastiano It is already \scriptscriptstyle

@Sebastiano Of course you should define a \onehalf command and not type all that mess each time.

@DavidCarlisle Hi, @cfr Welcome, I haven't seen you in the chat room in a long time.

@egreg and they all just want to lift it out of one group level by the look of it

@egreg I have understood. But for your opinion in the your answer 1/2 isn't that much bigger than the T?

@Sebastiano They have the same size. I would never use a fraction at a second level superscript. It's very difficult to read.

@Sebastiano Just omit that step and, if you want, explain the step after the main equation. It's a simple application of logarithms.

@egreg I vote and accept your excellent response. Thank you very much for your precious time. At this time I have problem with my line of internet.

Thank you all so much. I'm going to sleep that tomorrow I have 6 hours of class. Greetings and good work.