« first day (3879 days earlier)      last day (45 days later) » 

2:00 AM
@DavidCarlisle dvipdfM, I see. And 👎🏻 to 40 years of legacy cruft in the TeX ecosystem.
5 hours later…
6:53 AM
@DonHosek er well if you choose to start with meafont sources, it's a bit hard to avoid legacy tools:-)
7 hours later…
2:13 PM
@DavidCarlisle I see my \textsuperscript change broke a document ;-)
@UlrikeFischer seen my final comment under that answer?
@UlrikeFischer but it's not your fault:
2020-04-09  Frank Mittelbach  <Frank.Mittelbach@latex-project.org>

	* ltfloat.dtx (subsection{Footnotes}):
	Set non-zero baselineskip inside \textsuperscript and \textsubscript
	so that link boxes drawn by hyperref have a sensible height (gh/249)
@UlrikeFischer shh
Or \text_expand:n, which does try to work properly by expansion? — Joseph Wright ♦ 41 mins ago
@JosephWright guess how I found the bug of \text_expand:n with an argument that expands to nothing in two steps?
@Skillmon Er, I do wonder ;)
3:29 PM
@JosephWright I guess "do these two token lists expand to the same thing" isn't completely answerable expandably even with \expanded and \pdfstrcmp and friends?
@JosephWright "you are asking the wrong question" is probably the correct answer.
@DavidCarlisle Huh? It is provided the two sets of input work by expansion
@DavidCarlisle \text_expand:n is the best I can offer in this context
@JosephWright well thats what I mean, I mean arbitrary stuff. so "expand to the same thing" isn't well defined of course.
@DavidCarlisle Oh, sure, if we are talking any input that might contain assignments or whatever, certainly if you just use \edef
THIS IS TEH BEST TWITTER ACCOUNT twitter.com/shouldhaveaduck
@DavidCarlisle quack
@PauloCereda dinner
3:38 PM
@DavidCarlisle oh no
@DavidCarlisle My objection is more to the fact that we have both dvipdf and dvipdfm. Is there any compelling reason to keep the former around?
@PauloCereda wow, ducks and mint --- which one will reproduce faster to occupy all the available space is anyone's guess! ;-)
@DavidCarlisle oh no
3:54 PM
@PauloCereda Although could be an evil (?) plan: gressinghamduck.co.uk/recipes/…
@Rmano OH NO
@DonHosek dvipdf isn't in texlive, I had forgotten all about it, although I just noticed I have a copy in cygwin so I guess it's also in linux releases but it's not distributed by the tex distributions, we can't stop it being distributed by everyone
4:12 PM
user image
ooh a capy bike
@barbarabeeton ^^
@DavidCarlisle ooh
@PauloCereda out hunting for ducks
@DavidCarlisle oh no
@DavidCarlisle Hmm, my dvipdf is in /usr/local/bin on my Mac and appears to be a wrapper around dvips+gs I have no idea where it came from. I'll have to look on the other Mac which doesn't have texlive installed on it and see if it has this script. It does call the TeXlive dvips so perhaps a warning could be built into dvips if it's called with -Ppdf
@DonHosek I have it too. But since its in /usr/local/bin I guess it wasn't installed with MacTeX.
@DonHosek I assume it's installed as part of Ghostscript.
@DonHosek Ah, but MacTeX does install Ghostscript.
5:32 PM
@AlanMunn Yes, so it seems on Linux:
[romano:~] 4s % wajig whichpackage /usr/bin/dvipdf
ghostscript: /usr/bin/dvipdf
From #CTAN:

David Carlisle submitted an update to the carlisle bundle.

Version:  2021-06-13
License:  lppl

Summary description:  David Carlisle’s small packages
@DavidCarlisle ^^ ooh
5:46 PM
Seems pgfmath can not handle with

\pgfmathprint{3<\x<5 ? 1 : 0}

but with

\pgfmathprint{3<\x && \x<5 ? 1 : 0}
@PauloCereda somebody broke ltxtable :(
@DavidCarlisle @UlrikeFischer?
@cis there are not many languages that can evaluate 3<x<5 as a boolean (python can but I don't know anything else)
@PauloCereda the longtable author
@DavidCarlisle Ah ha.
6:10 PM
@PauloCereda Do you also have a bundle? A duck bundle.
@mickep soon :)
@PauloCereda Talking about secrets, do you have any nice MP drawings yet? :)
@mickep ooh haven't got much time, but it's a WIP :)
@PauloCereda Very good! I had some fun last week preparing the geometry part of the course I'll teach this autumn. It is a pure joy to do it in MetaPost.
@PauloCereda probably you saw the picture mode whale and decided to give in, you can't compete
6:13 PM
@DavidCarlisle ooh
6:39 PM
@DavidCarlisle Haha
Yay the sheep person is here
@Plergux quack
@samcarter_looks_forward_TUG'21 oooh! love the floppy ears. like bunny-sheep :p
@PauloCereda baa!
The NewComputerModern-Book looks nice.
6:58 PM
@PauloCereda one for you: abntex2ppgsi I know you like abntex classes
@DavidCarlisle who's afraid of longtable-man?
@DavidCarlisle oh no, someone tweaked something dreadful to something more dreadful.
! Undefined control sequence.
l.1 \ducklove

? H
The control sequence at the end of the top line
of your error message was never \def'ed. If you have
misspelled it (e.g., `\ducksya'), type `I' and the correct
spelling (e.g., `I\ducksay'). Otherwise just continue,
and I'll forget about whatever was undefined.

@Plergux ^^^
@Skillmon :D
2 hours later…
8:52 PM
@PauloCereda Ohh! Nobody can be without duck after seeing this twitter account!
2 hours later…
10:48 PM
Is there sth. like this?

\ifnum\x=11\or12 yes \else no \fi
    \ifnum\x=11 1\fi
    \ifnum\x=12 1\fi
    % ...
@cis ^^^
@PhelypeOleinik Uiuiui
@cis What does that mean? :-)
@PhelypeOleinik So there is only the hard way.
@cis It's not as hard as it looks, but you can also do:
\int_case:nnTF { \x }
    { 11 } { }
    { 12 } { }
  { yes }
  { no }
10:55 PM
@PhelypeOleinik Ok, seem I should start to learn this strange exploration syntax. But I don't know where.
@cis My (personal) suggestion for expl3 is at the bottom of this answer: tex.stackexchange.com/a/597030/134574
@PhelypeOleinik thx
@cis You're welcome!

« first day (3879 days earlier)      last day (45 days later) »