@gusbrs \let\thebeginnotes\theendnotes?

@PabloGonzálezL actually, if I'd known about that yesterday, I might not be entangled in systemd-nspawn. oh, well.

@cfr :-D I almost did that in the MWE!

@gusbrs >^.^<

Hmmm.... My RSS feed pointed me to latex-project.org//news/2024/11/01/issue40-of-latex2e-released, which seems not to exist...

@Rmano hmm latex3.github.io/news/2024/11/01/issue40-of-latex2e-released exists and was supposed to have moved itself...

latex-project.org//news shows TUG as the newest entry...

@Rmano thanks: we'll blame someone investigate

@Rmano latex-project.org/news/2024/11/01/issue40-of-latex2e-released works (with www.) not sure yet what it's showing without

@UlrikeFischer You may recall some time ago I was playing with that multiple option (a la footmisc) for postnotes, which uses \lastkern as a "signal". To make it work I had to store \lastkern at the beginning of the note, otherwise it was lost somewhere, and I didn't know why. I'm taking a closer look at this now, and some bisecting shows the culprit is \MakeLinkTarget. I don't know if this is relevant in any way, if it's avoidable, or if you should care, but... I thought mentioning it.

@Rmano works now

@gusbrs with or without hyperref? And do you have an example?

@UlrikeFischer Both. See this bit: github.com/gusbrs/postnotes/blob/…. \@@_multiple_store_lastkern: is storing the value of \lastkern. I started moving it down, until it went past \MakeLinkTarget, when the tests started to fail. I have two for the feature, one with hyperref the other without (because of footmisc), both fail at that point. I think it is that \smash, as far as I get.

@UlrikeFischer No hyperref example:

\documentclass{article}

\usepackage[multiple]{footmisc}

\begin{document}

\footnote{1}\footnote{2}

\footnote{3}\MakeLinkTarget*{mytarget}\footnote{4}

\end{document}

@UlrikeFischer But, even with hyperref, couldn't it be made to work. It stores spacefactor, why not lastkern as well?

@UlrikeFischer The example fails with phase-III as well.

@JosephWright The current state for the matrix server is that the server is running, but I'm still doing some tests to ensure that we don't overload the server when it actually gets used.

@gusbrs Well every whatsits destroys \lastkern, e.g. \footnote{3}\label{blub}\footnote{4} does it too. But why is your anchor between the footnotes and not inside?

@UlrikeFischer I see, then it is what it is. This is the anchor of the mark (for the backlink), not of the content. It immediately precedes the mark. There's only label setting and assignments in between.

@gusbrs well it looks as if you are checking for the multiple marker too late. You should do that before you set an anchor.

@UlrikeFischer Indeed, you are right. Now that I know what the problem is, I might review this. And thanks.

@DavidCarlisle Jeje

@cfr Ouch...I haven't actually tried act, but it works directly with gh cli so (I think) it's the best option for GitHub actions.

@mbert Are you online?

can I clone a git repo locally i.e. pick up stuff I haven't pushed and not pick up untracked things?

o/w I can use subversion, but I consider that cheating ...

@PabloGonzálezL seems very likely. on the other hand, bash is easier than yaml.

@cfr yes you could just git clone your existing clone but why not just start a new branch?

@DavidCarlisle because I want to put it somewhere else?

@cfr you can just copy clone directory anwhere and then do a git clean on the copy (that's what I do when moving to a new machine, more or less)

@DavidCarlisle I don't lose untracked files if I start a new branch, though?

@DavidCarlisle the amans class does

\let\amans@begindocumenthook\@begindocumenthook
\let\amans@enddocumenthook\@enddocumenthook
\def\AtBeginDocument{\g@addto@macro\amans@begindocumenthook}
\def\AtEndDocument{\g@addto@macro\amans@enddocumenthook}
\def\@begindocumenthook{\amans@begindocumenthook\amans@beginCJK}
\def\@enddocumenthook{\amans@endCJK\amans@enddocumenthook}

Guess what's wrong! ;-)

@DavidCarlisle By the way, they want to typeset the whole document inside a CJK* environment! Good job!

@egreg where did you find that? I had a quick look but couldn't be bothered to search the site

@DavidCarlisle I was lucky to find the right button in the site.

@egreg feel free to answer and pick up a tick:-)

@DavidCarlisle git clean -f doesn't actually do anything for me, so presumably I'm doing something wrong.

@PabloGonzálezL Yep

@mbert I left you a message in texdoc enumext :)

git clone -l --no-hardlinks --single-branch A B?

@cfr perhaps your untracked files are ignored?

Normally, only files unknown to Git are removed, but if the -x option is specified, ignored files are also removed. This can,
       for example, be useful to remove all build products

Oh, new enumext question...

@mickep release packages (or fonts) with bugs, easy way to generate rep points

@DavidCarlisle Oh, you know, with longtable, and the others...

@PabloGonzálezL Haha thanks

@mickep that only has features, no bugs

@DavidCarlisle yes. thanks. some are not ignored, but I didn't test with those. but clone is one command rather than two and avoids copying files only to delete them.

@cfr yes clone is the right command really

@mbert :D

@JosephWright today was the first time I saw use-xspace in siunitx, and I must say I was a bit bewildered :)

Interesting fact: In that usage, free-standing-units wasn't set.

@Skillmon :)