« first day (3883 days earlier)      last day (37 days later) » 

4:13 AM

\begin{frame}{Frame 2}
\begin{frame}{Frame 3}
\begin{frame}{Frame 4}
\begin{frame}{Frame 5}

How to distribute the sections in the headline ?
How can I make the latter two sections appear on the other side of the header ?
2 hours later…
6:07 AM
@JosephWright got something strange this morning. I'm changing the \ohm symbol of siunitx to use \text{$\upOmega$} in my preamble. I get the following two \show calls (seemingly added for debugging?) with siunitx version 3.0.11:
> \l__siunitx_symbol_tmpa_tl=macro:
->\mathrm {Ω}.
<argument> \show \l__siunitx_symbol_tmpa_tl
                                            \show \l__siunitx_symbol_tmpb_tl...
l.85 \begin{document}

> \l__siunitx_symbol_tmpb_tl=macro:
->\mathrm {\protect \text  {$\upOmega $}}.
<argument> ...tl \show \l__siunitx_symbol_tmpb_tl
                                                  \group_end: \prg_return_fa...
l.85 \begin{document}

6:23 AM
@JosephWright with the latest version (just updated) this doesn't happen.
6:39 AM
@Skillmon I remember this came up here in chat before (a few days ago). You might have luck searching for it to get Joseph's response.
1 hour later…
8:03 AM
@gaufler the other side of the header is for subsections
@Skillmon you were lucky it was only \show not \tracingall that got left in.
8:25 AM
Here is an interesting one we don't quite understand
\settoheight{\varheight}{$\overline{P}$}% Capture height of \overline{P}

These have the same height


Came up in a discussion. Why is ^A not placed the same in both the green and red example. I've shown below that they have the same height. What are we missing?
@daleif er well, it's the difference between these two, so it's just a matter of perusing appendix G .....
\accent\fam0 _
..\fam1 P
^\fam1 A

\accent\fam0 _
..\fam1 P
^\fam1 A
Should mathchoice then end up with the same height? or are there no height? I cannot really read this
@daleif and removing any suspicion that it's a color node getting in the way


\settoheight{\varheight}{$\overline{P}$}% Capture height of \overline{P}

I already knew the colors did not affect anything.
@daleif but extra braces fix it, I wonder if we dare change vphantom?


\settoheight{\varheight}{$\overline{P}$}% Capture height of \overline{P}

@daleif yes but I didn't and needed to make sure that I wasn't to blame:-)
8:41 AM
@DavidCarlisle always the first thing to do.
@DavidCarlisle who wrote \vphantom?
I just do not understand what goes wrong here? I always assumed that a phantom always had the same characteristics as the original.
@UlrikeFischer not me (I think?)
@daleif I think it's placing the superscript on the subsubscript choice, which seems.... odd.
@DavidCarlisle I don't know. The only two changes entries say
% \begin{macro}{\phantom}
% \changes{v1.0p}{1994/11/18}
%         {(DPC) use \cs{expandafter} instead of \cs{next}}
% \changes{v1.0p}{1994/11/18}
%         {(DPC) colour support}
@DavidCarlisle hmm, that seems plausable. Hmm, there might be people relying on this wrong behavior.
8:52 AM
@daleif it doesn't get it wrong for a general mathchoice, this is right
\mathchoice{X}{x}{{a\over b}}{{X\over Y}}^n
{X\over Y}^n

@UlrikeFischer two excellent changes, don't you think?
@UlrikeFischer what was the tex "feature" we had to work round not long ago were tex-the-program has an "optimisation"x that doesn't correctly repack boxes whose size has changed? (@PhelypeOleinik)
@DavidCarlisle is it this problem? tex.stackexchange.com/questions/165499/…
@UlrikeFischer yes exactly and @wipet's answer is better than the accepted one I think. ( I just voted for it:-)
2 hours later…
10:31 AM
@egreg (@DavidCarlisle) I am thinking (again) of having a look back over breqn and seeing which bits might be useful longer-term - ironically, I think that everything except the automatic breaking is probably 'on the list' ;)
@Skillmon Indeed - was only present for a day or so
@DavidCarlisle To be fair, v3 is a lot faster than v2 partly as the tracing is a lot lighter (relatively)
@JosephWright while pushing a few people to adapt to the pdfmanagement I came again across "overprint/knockout" which is e.g. in colorspaces. We should have some code and interface for it. But where at best? in l3color?
@UlrikeFischer There are a few of those things at the 'trapping' end that we need to think about - I'm not sure they are strictly colour-related, but at the same time I don't have a good suggestion for where they do live
@UlrikeFischer Perhaps in l3pdf?
@JosephWright perhaps we should rename l3opacity in l3graphicstates.
@UlrikeFischer That's a possible, yes - there are a few other things that we need. There's a reason I just put it in the trial area for a bit of experimentation
@UlrikeFischer I guess we should list out all of the stuff that needs to go in there and see how it looks
10:53 AM
@Rmano Hi, sorry missed the message yesterday (having some busy time). You can switch what we informally call the "strict" mode, formally halt-on-error, see overleaf.com/learn/how-to/… for details. Other than that, as you pointed out, it's very difficult to grab the errors.
@JosephWright perhaps also the text parameters from the pdfrender package.
11:08 AM
@UlrikeFischer Yes, so like you say l3graphicstate looks the way to go
11:25 AM
@yo' looks like tikz is going to change here so this particular case will go
@egreg, @DavidCarlisle (@daleif) github.com/josephwright/mathtoys - Have to see where I get to
@JosephWright do we allow chemists to play with our toys?
@JosephWright you do know that people ignore sentences like "this code should not be used in production" and blame you for errors anyway?
@UlrikeFischer Yes ;)
@UlrikeFischer afterpage
11:29 AM
@UlrikeFischer Currently no actual code so I'm relatively safe - it might not get anywhere, but I would like to see if I can extract some of the bits of breqn I like the idea of
11:41 AM
hi there! is any chance to update MikTex packages from command line without opening console all the time to verify if there are some packages updates? Or maybe a task scheduler in windows with some parameteres?
12:04 PM
@Mafsi you can use mpm on the command line docs.miktex.org/manual/mpm.html
@yo' the good news is that this problem will be fixed soon: github.com/pgf-tikz/pgf/issues/1025 !
@Rmano yay!
12:27 PM
@UlrikeFischer Thanks a lot, @UlrikeFischer!
@samcarter_looks_forward_TUG'21 I think I've found another strange but funny package that maybe can be added to the `PackageOfTheDay` repository.
Would you prefer me to add a new comment, or edit the one I wrote before?
I'm asking because this package has nothing to do with calendar-related packages :D
@LukeTheWolf Whatever is easier for you
@samcarter_looks_forward_TUG'21 It's the same thing, but I think it makes more sense to put a new comment, detached from the previous one :)
@LukeTheWolf I just saw your edit - thanks! Just leave it as it is.
1:14 PM
@samcarter_looks_forward_TUG'21 wow, I feel honored --- two packages by me on the list!
Although thinking well one idea is from Randall Munroe and the other one is... picture mode ;-)
@Rmano both picture mode and Randall Munroe are a guarantee for success :)
1:30 PM
@samcarter_looks_forward_TUG'21 I don't think that afterpackage should be in your package list. What it tries to do can be done now with proper LaTeX hooks, without redefining internal commands like \@popfilename. (I didn't test if it breaks with a current latex, but I wouldn't be surprised if is does).
@samcarter_looks_forward_TUG'21 the list at your github is rather short?
@DavidCarlisle :-O listing at celtic music now, so I hope to learn more from entry #3 ;-)
@DavidCarlisle there is an issue.
@UlrikeFischer I assumed that since pineapple encouraging packages were banned, @samcarter_looks_forward_TUG'21 had had no submissions.
1:39 PM
@DavidCarlisle you could change the documentation of typewriter to shorten the list.
@DavidCarlisle At the moment I'm just collecting. I'll keep doing the ducks of the year until end of this month and then switch to package of the day.
@UlrikeFischer thx. I'll keep that in mind
2:06 PM
@DavidCarlisle I wish I could shorten github.com/josephwright/mathtoys/commit/…
2:32 PM
@JosephWright use context
@DavidCarlisle :)
@PauloCereda there was a rubber duck on the bbc news last night, something to do with understanding comets or asteroids or something.
@DavidCarlisle ooh
@DavidCarlisle Local newspaper also had duck news today: two men wanted to buy ducklings from a farmer, they celebrated the deal with drinks, got into a fist fight and the police had to come to stop the fight - I hope the ducklings were not harmed :)
@UlrikeFischer Out of curiosity I just tested afterpackage and it seems to still work with current latex.
2:39 PM
user image
@PauloCereda /quack
@JosephWright /quack
3:37 PM
@DavidCarlisle @UlrikeFischer see what you have done?
@DavidCarlisle ;-) but you mentioned the package ...
3:51 PM
@UlrikeFischer can you read Frank's message or is it somewhat low resolution?
@DavidCarlisle very low resolution ;-)
2 hours later…
6:19 PM
@PauloCereda Wrong language, but cute logo: meta.stackoverflow.com/a/408478/2777074
@samcarter_looks_forward_TUG'21 ooh a duck
3 hours later…
9:19 PM
Oh no! updating homebrew and it deleted the cask duckstation
(I have no idea what this is, I just thought it was funny.)
1 hour later…
10:41 PM

« first day (3883 days earlier)      last day (37 days later) »