@UlrikeFischer Comparing your answer tex.stackexchange.com/a/273296 to biblatex.def from the current TeX Live, I see that biblatex.def has switched from dx.doi.org to doi.org and is possibly doing something else I cannot understand. Would it be possible for you to take a look and, perhaps, update your answer?

Perhaps, there might in the meantime a better option to break a DOI after slashes, dashes, and periods (rather than after every symbol) when using latex+dvips?

Has anyone used the TikZ perspective drawing library? It seems robust, but I cannot figure out how to properly set the view.

The view is "determined by rotating the coordinate system by -azimuth around the z-axis, and by ⟨elevation⟩ around the (new) x-axis, as shown below."

This is all good, but I cannot seem to get the z-axis to be anywhere except point along the vertical axis of the page.

As a sanity check, I produced an images of the axis arrows in a for loop, setting the azimuth and elevation to be between 0 and 360 deg, in increments of 15 degrees. All the images had z on the vertical axis.

@DavidCarlisle worldbuilding.stackexchange.com/questions/244332/…

@CarLaTeX useful information

@DavidCarlisle :D

@CarLaTeX Pineapples are very dangerous, especially if used as topping.

@egreg Indeed!

@egreg yep, addiction is nothing that should be taken lightly.

Back on Typset, I think I've spotted what is a pretty big issue with their chosen input syntax, at least for maths ...

Try writing in TeX terms $a_{ij}$ and find you need $a_(i j)$ - that space is not at all pretty

@JosephWright In general, their contractions of two letters in math mode are sometimes funny. I'm also not convinced that RR over \mathbb{R} or similar (the latter at least indicating font change) is that great of a choice.

@TeXnician Urgh, I'd missed that

@TeXnician The more I look, the more I think this is going to be a missed opportunity ... trying to avoid escape chars just doesn't scale well

@JosephWright Oh, they do have escape chars (from what I understood, you can avoid RR being blackboard letter by means of some escape char), it's just not built for scalability to a number of math alphabets and use cases, it seems.

@TeXnician Oh, yes, sorry, what I mean is they've decided they don't want to require an escape char for operators, blackboard-bold, etc.

@JosephWright Ah, yes, agreed.

[12/15, 03:26/03:26] auto-install: latex-context-ppchtex (66687) [2k] ... done

a brilliant package ^^^

@DavidCarlisle Oooh

@TeXnician hi, could you please tell me how I can change to mkiv for compilation in TeXLive 2023? My circuitikz test fails...

(I know nothing about context, unfortunately; I tried to maintain compatibility during the last 4 years but now I am a bit swamped...)

@Rmano tl list has a discussion currently about a "show stopper" bug in luametatex and whether they should allow the binaries to update, or wait a year ...

@DavidCarlisle Oops... missed that (somebody interested in being head of department?). I am seriously thinking about suspending the compatibility checks for context until it stabilizes. I have the same exact error reported by @PabloGonzálezL: texlive.net/…

@PabloGonzálezL --- is there a issue open somewhere about that? I do not know if adding that to github.com/pgf-tikz/pgf/issues/990 could help, or if it's already known.

@Rmano try texlive.net/…

@Rmano a (not very circular) circuit in context texlive.net/…

I may have to take over from @JosephWright as the latex project context mole

@DavidCarlisle yes, that works! But the support for the embedded siunitx fails... I have to investigate more (anyway I never liked that trick in circuitikz to be honest.

@DavidCarlisle Oh I doubt it ;)

@JosephWright I thought you'd been lured away to more modern apps

But maybe siunitx is just now supposed to work in context? And it worked by chance?

@Rmano example?

@DavidCarlisle wait... trying to minimize it.

@DavidCarlisle texlive.net/…

...used to work in 2022. But again, probably by accident...

Using \draw (0,0) to[R=1~kΩ] (2,0); works ok. ;-)

@Rmano I don't think siunitx was loaded, in 2022 I get

> \kilo=macro:
->\rm {k}

in 2023

> \kilo=undefined
tex error

the 2022 defn doesn't look like siunitx I guess

Hmmm... funny. I have to look at it!

I really do not know from where that definition was coming. Mmmhhhh

OK, I found something. I have

\processaction
   [\currentmoduleparameter{siunitx}]
   [true=>\def\SI#1#2{#1\,#2}
   \def\ampere{\rm{A}}

and before that

It seems that something changed and the \defs are not executed any more.

Not a fan of that, really...

The removed thing was

\setupmodule
   [current=european,
    voltage=european,
    resistor=american,
    inductor=cute,
    logic=american,
    siunitx=true,
    arrowmos=false]

@DavidCarlisle so I think this is a circuitikz fault but something changed in the module processing thing in context.

@UlrikeFischer When you have a mo, could you look at updating the copy of pdfresources in the L2e repo? There were a lot of failures when I tried, and I'm not confident in checking all of them

@JosephWright ok. Are the failures only in latex-lab or somewhere else?

@UlrikeFischer In the lab

@Rmano Sorry, but you can't (at least not using TeX Live, it now ships LMTX, and you should test against that as well as MkIV is now basically obsolete). David already answered your immediate problem above. But you may also consider to run your action against the latest LMTX (which will probably resolve ConTeXt bugs faster than TL given the discussion on the TL list).

The Island of TeX provides lmtx docker images, though I'm no GitHub actions expert, so can't really help you on how to change base image.

@JosephWright I updated the tests and now check on github (two test can perhaps fail, (../xtemplate versus xtemplate) as I didn't yet understand why it is something the one and sometimes the other)

@JosephWright I have updated the tests, and the tests pass, if you want you can merge. that.