@LaTeXereXeTaL I have rapid slide transitions (or whatever they're called) which work in Okular, so I don't think Adobe is required for those.

@samcarter @cfr I've experimented with Adobe Reader and the rapid transitions seem to work perfectly. I had forgotten about Présentation.app. I didn't know it was still around.

@cfr I've never heard of Okular but it looks interesting and I will try it. I'm on macOS Sonoma.

@LaTeXereXeTaL I'm on Linux. You might find it rather heavy duty for OS X. I'm using KDE anyway, so I've need the frameworks and libraries anyhow. If you're using a different environment on Linux or a different OS altogether, you might not want to install it all just to try the PDF viewer ;). It would be a bit like installing half of Aqua/Quartz/Cocoa or whatever OS X uses (I forget & it's probably changed) - just to get Preview. Right now it has a horrible bug copy-pasting selections as images.

I have a question: Some days ago, I learned about the macro \symbol{...} which converts an integer to a character. For example, \symbol{97} returns a. I want to read the documentation of the macro \symbol{...}. Which document from texdoc contains the documentation for \symbol{...}?

@cfr Yeah I see there's no way to easily install it on macOS.

@LaTeXereXeTaL :(

@LaTeXereXeTaL Though it isn't as good since they improved it.

Does anybody know how to prevent makeindex from using item_0 not only between two top level items (as the manual says) but also before the first? It works fine for the standard case where the index is a list. It doesn't work well at all if the index is a tabular. The best I've come up with is \\[-5ex] to undo two rows worth of space or thereabouts.

Find of the day:

\renewcommand{\(}{\begin{columns}}
\renewcommand{\)}{\end{columns}}
\newcommand{\<}[1]{\begin{column}{#1}}
\renewcommand{\>}{\end{column}}

... and the user was wondering why they got errors

@egreg Good morning Enrico. Please can you traslate better in English language my question? math.stackexchange.com/questions/4888124/…. Can you put, please, also your answer? Thank you very much.

@Sebastiano Without going into details, the answer you got makes very much sense.

@rdrg109 texdoc source2e (or the latex book of course) but it's just a trivial wrapper arounf \char the definition is \DeclareRobustCommand\symbol[1]{\char#1\relax}

@LaTeXereXeTaL You seem to use Adobe Reader. So, package animate could be an option too. It allows you to animate image sequences playing at well-defined frame rates within a single frame of your presentation. The command interface is similar to \includegraphics and its scaling options. Optional control buttons for play/pause and stepping are available.

@UlrikeFischer ^

@DavidCarlisle without testing: not my fault. One shouldn't have ExplSyntaxOn when the supp-whatever file is loaded at begin document (or the aux file or whatever ...) @AlexG

@UlrikeFischer Ok. Will keep this in mind. Just wanted to experiment with some interface3 and testphase commands within the document body and enable l3 globally in the preamble.

ah I didn't notice the trailing ExplSyntax, so not our fault

@DavidCarlisle seen the ucs issue report? Why does it try to process the options?

@UlrikeFischer no in meetings

@UlrikeFischer where?

@DavidCarlisle github.com/LaTeX-Package-Repositories/ucs/issues/2

@UlrikeFischer odd, I don't seem to have that mail i wonder if gmail filtered it

@UlrikeFischer yes it's in spam:-)

Building dev release

@UlrikeFischer does that really not error in miktex?

@DavidCarlisle it errors in miktex too.

@UlrikeFischer ah it's a feature, I'll fix.

@DavidCarlisle ?

@UlrikeFischer ucs

@DavidCarlisle yes but why a feature? From the new key handling?

@UlrikeFischer no always I think

@DavidCarlisle ah, then I wonder why they claim it works in miktex. But in any case it is good to know that one need to consider options when removing packages like this.

@Skillmon That will be today, so 2024-06-01 PR2

@DavidCarlisle rereading he actually doesn't say that it doesn't error in miktex, he only wonders about the different dates.

@mickep Hi. Thank you for your reply. I wanted a solution on my question and not on other details.

Urrgg, turns out the shortcuts which will make every document, which uses \(...\) for inline math, explode, come from an overleaf template overleaf.com/latex/templates/beamer-template-by-cameron-bracken/… - and the template then does not even use them, just breaks other documents for no good reasons :( I hate templates

@samcarter blame @PauloCereda

@DavidCarlisle Great idea! Makes me feel so much better :)

@samcarter where did that turn up?

@UlrikeFischer on SO, but beware that the question is a big mess, missing MWE and it turned out to be a combination of unrelated problems which were not mentioned in the question stackoverflow.com/q/78215769/2777074

Why does \begin{foo}\end{foo} not fail on \end{foo} as long as \foo is defined even though \endfoo is not defined? I tried to look at the code of \end but didn't really understand it. It seems that if \endfoo isn't defined it gets turned into \relax. Is there some reason for this as opposed to throwing an error?

@AlanMunn \csname end Alan Munn\endcsname is \relax not an error

@AlanMunn well I don't know why it was done this way, but if one would change that now documents doing \begin{small}...\end{small} would error ...

@DavidCarlisle So does that mean that every \end{...} turns into \relax? Or is it just when there no existing \end... defined?

@AlanMunn \end{foo} does \csname end foo\endcsname\endgroup and the rest is down to tex primitive behaviour

@DavidCarlisle oh no

@DavidCarlisle I see also that you can't define \endfoo using \newcommand either.

@UlrikeFischer Yes, that's a good example of the effect in actual (mis)use.

@AlanMunn yes because if you defined \endsection just because it was undefined \begin{section}{Introduction}...\end{section} would change with no warning

@AlanMunn the \begin{small} example is in the latex book if I remember correctly, so documented feature, not misuse

@DavidCarlisle \begingforproblems

@AlanMunn it is sometimes but not always a misuse. \begin/\end do more than simply grouping. They have hooks and they keep track of the @endpe switch. Compare e.g. here the indentation after the list/center:

@mickep you are just jealous that context just has \startfoo \stopfoo and not proper environments \begin{foo} \end{foo}`

@DavidCarlisle For blocks there is in fact some begin and end syntax.

@mickep not in any of the context manuals I have read

@DavidCarlisle Well that is just because you did not read any.

@mickep a response in honour of @AlanMunn (but yes, exactly)

@UlrikeFischer Thanks for the help. You seem very stressed is everything ok with you?

@UlrikeFischer I see. I'm learning all sorts of things today. :)

@AlanMunn I learned that from the tagging code ;-). One of the more frequent errors is that lists structures go wrong because a group hides the @endpe.

I'm reading the document "$ texdoc latex2e" and In section "4.2 Font Styles", \texttt is shown along all the other commands that are used for changing the font style. "Typewriter" is shown next to \texttt. So, the name of command stands for "text typewriter (one more t)". What does the last "t" stand for?

@rdrg109 typewriter type

@yannisl interesting, I have for some reason read it as teletype.

@mickep I have read the TeXbook

@yannisl Ok.

@rdrg109 \tt is the traditional name also shows in latex as \ttfamily along with other two letter abbreviations rm for roman sf for sans serif and they all consistently hav a \..text{...} form which is consistent wven three t in a row is a bit hard to parse.

@AlexG Yes indeed I will experiment with it.

@AlexG AHA! I sense you might be the author. LOL

@JosephWright thanks, so just testing against 2024-06-01 should be enough... I'll see when I manage to update the packages :)

@UlrikeFischer have you any idea who'd write text with funny characters like this?

l.14 \section{grüße}

! Argument of � has an extra }.
<inserted text>
                \par
l.14 \section{grüße}

\documentclass{article}
\usepackage{ucs}
\usepackage[utf8x]{inputenc}

\usepackage[unicode]{hyperref}

\begin{document}

\section{grüße}
Dummy text here

\end{document}

@JosephWright ^ works in overleaf (with latex 2023-06-1) but fails in my texlive 2023 (with latex 2023-11-01) I'm wondering if I should try to make it work our just update the test to say it fails (@UlrikeFischer)

@DavidCarlisle no idea. Do we have germans in the team?

@UlrikeFischer we have well known max and moritz typesetters, they must know German

overleaf has hyperref 7.01b and my updated tl2023 has 7.01h but I think I'm just going to change the test, I can't see keeping hyperref working if ucs is loaded as a requirement? Not sure what changed exactly though

@UlrikeFischer was supposed to be simple, just add a line and push to ctan, who's idea was it to have a test suite that can complicate things by failing?

@DavidCarlisle does it really work? I thought we only made it work with utf8x github.com/latex3/hyperref/issues/248

@UlrikeFischer overleaf.com/read/sgfhrytvbzqp#26c261

@UlrikeFischer It worked in June 2023 ^^ but failed by November

@UlrikeFischer that's what I thought we'd done (and what I propose to change the test to check) but it works on overleaf and I checked in a tlg that seem to imply it worked with ucs loaded as well

I'm going to change the test to just load [utf8x] and not ucs.

@UlrikeFischer github.com/LaTeX-Package-Repositories/ucs/commit/…

@DavidCarlisle hm, I don't see what should have changed here. I see only a rename of the utf command but that should be ok.

@UlrikeFischer I decided I don't care I think we should only claim it works if ucs is not loaded so I changed the test

@UlrikeFischer I could push to ctan (unless you want to look more:-)

@JosephWright latex-dev all gone to ctan?

@DavidCarlisle it seems to be this commit: github.com/latex3/hyperref/commit/… did we miss something in ucs here?

@UlrikeFischer ah... \let\HyPsd@expand@utfvii\HyPsd@expand@UTFviii

@UlrikeFischer let me try that with viii

@DavidCarlisle spot the difference \HyPsd@expand@utfviii \HyPsd@expand@UTFviii

@UlrikeFischer yes works now thanks

diff --git a/ucs.dtx b/ucs.dtx
index a6af263..71215b0 100644
--- a/ucs.dtx
+++ b/ucs.dtx
@@ -536,6 +536,7 @@

   }
  \let\HyPsd@expand@utfvii\HyPsd@expand@UTFviii
+ \let\HyPsd@expand@utfviii\HyPsd@expand@UTFviii
 }
 \fi

@DavidCarlisle you could perhaps define \HyPsd@expand@utfviii directly, the let was only temporary

@UlrikeFischer I think minimal change is good, changing two lines is already twice as much as I'd planned

although I suppose it would look more sensible...

@DavidCarlisle Yes

@JosephWright l3sys-query as well, to keep them busy?

@UlrikeFischer all tests pass again defining it directly, I'll push to ctan, thanks

@DavidCarlisle hadn't you somewhere an answer with l3sys-query?

@DavidCarlisle I'll sort tomorrow

@DavidCarlisle 'Not my fault' ;)

@JosephWright not mine either (I can count to 8 in Latin)

@UlrikeFischer I thought so,but I don't see it

@DavidCarlisle by reading the documentation I managed to get it working ...

@UlrikeFischer the expl3 version or 2e ?

@DavidCarlisle expl3. But it doesn't work in miktex yet?

@UlrikeFischer a miktex problem or just expl3 not updated there?

@UlrikeFischer almost got an automatic math tagging pipleline based on latexml

@DavidCarlisle the tool is not there (yet).