« first day (3827 days earlier)      last day (1089 days later) » 

7:22 AM
Good morning! A quick check: I want to do \tikzset{a/.try=1, b/.try=2, ...} (using .try for a whole tikzset command). Is it possible to compact this? Telling at the start to use .try for the rest of the command?
 
@Rmano Hmm... is it possible to use "for x in y" in \tikzset?
 
@Plergux hmm, I suspect no. There is a kind of .handler facility, but I thing works just for one level...
 
@Rmano drat
 
Correction to the message before: I want to do \tikzset{a/.try=1, b/.try=2, c/d/e/.try=3} --- it should work for nested keys too.
 
7:40 AM
@Rmano gzip
 
7:51 AM
 
@Rmano it's best not to have users
 
8:14 AM
@DavidCarlisle definitively
 
@Rmano not sure I'd call these people users. more like abusers.
 
@Plergux that reminds me... "ab-normal?" (Young Frankenstein)
 
@Rmano I've not seen Young Frankenstein (it's disgraceful, I know) so I'm not quite getting the reference. :p
 
@Plergux :-O
"Put... the candle... back in!"
 
8:29 AM
@Rmano Yeah, I know. It's super weird cause I'm a massive Mel Brooks fan but I've seen like half of his movies twenty times and the other half I keep forgetting XD
 
I had asthma when I was a teenager, and my parents forbade me to watch it because it always triggered an attack (by too much laughing)
 
8:42 AM
@Rmano O.O Mel Brooks tends to have that effect on people :p I think I will never forget when I first saw Robin Hood because my youngest sister laughed so much at the bit where Big John fell in the "river" that I thought we'd have to resuscitate her XD
 
 
2 hours later…
So far, I've had a grand total of one bug report for siunitx v3 (thanks @UlrikeFischer); so either the code is excellent or ....
Now I've got to make the tricky call of when to go for it: I was expecting more howls of protest
 
@JosephWright which reminds me, I had to revert yesterday to v2 for a document (and it took me only three tries to get the correct syntax \usepackage{siunitx}[=v2] ;-)) as this errors: \num[round-integer-to-decimal]. What is the replacement?
 
@UlrikeFischer Ah, I must have missed that one in the emulation file
 
@JosephWright you mean I found a second bug? ;-)
 
@JosephWright ZOI? :)
 
10:36 AM
@UlrikeFischer I decided that it's not really something that's needed: for physical measurements, the different between an integer and a decimal is just one of what you right down in terms of number of digits
@UlrikeFischer Looks that way
 
@UlrikeFischer It's ZOI again :)
The Zero one infinity (ZOI) rule is a rule of thumb in software design proposed by early computing pioneer Willem van der Poel. It argues that arbitrary limits on the number of instances of a particular type of data or structure should not be allowed. Specifically, an entity should either be forbidden entirely, only one should be allowed, or any number of them should be allowed. Although various factors outside that particular software could limit this number in practice, it should not be the software itself that puts a hard limit on the number of instances of the entity. Allow none of foo, one...
@JosephWright ^^ it's either zero bugs, exactly one bug or infinite bugs. :)
 
@UlrikeFischer You might want to use drop-zero-decimal if you want to avoid rounding adding a (visible) decimal part
 
@JosephWright No, I want two decimals (that's for money, and it looks odd if the ,00 is missing.) I guess minimum-decimal-digits=2 is right? Would that work with something that starts as an integer?
 
@UlrikeFischer Yes, that would work
 
@PauloCereda I think you could change that to "computers always find a way..." :p
 
10:40 AM
@UlrikeFischer It's independent of rounding, so a good choice
 
@JosephWright did minimum-decimal-digits exist in v2?
 
@UlrikeFischer No, it's a new idea I had
@UlrikeFischer Essentially, it's a generalisation of add-decimal-zero
 
@JosephWright a good idea. Typically I want to control the number of decimals. Is there something to replace 1,00 by say 1,-- ?
 
@UlrikeFischer No, that would be a feature request :)
 
11:03 AM
@UlrikeFischer github.com/josephwright/siunitx/issues/413: possibly zero-decimal-as-dash?
 
@JosephWright ;-)
 
@UlrikeFischer See my comments
@UlrikeFischer It's an easy option so I've put down for v3.1, which is basically meaning 'I'll probably add it as soon as v3.0.0 ships'
 
@Plergux ooh :)
 
 
1 hour later…
12:15 PM
@PauloCereda counter-ooh!
 
 
3 hours later…
3:35 PM
A LaTeX related licence question law.stackexchange.com/questions/64400/…
 
3:56 PM
@AlanMunn interesting one.
 
@Skillmon I think it's a common misconception about package licences (but of course I'm not a lawyer). The package/class is effectively an extension of the software you use to write the document. So in the same way that MSWord has no claim to the documents you write with it, package/class authors have no claim to the things you write with their class/package.
@Skillmon So the licence affects (re-)use/distribution of the class itself only.
 
@AlanMunn I'd agree, but just as you said, I'm no lawyer myself, so I don't want to make definitive statements on this :)
 
@Skillmon Exactly.
@Skillmon We do have on LaTeX using lawyer on the site that I'm aware of, Keks Dose.
 
@AlanMunn is he still active? Haven't seen answers by him for a while now.
 
@Skillmon No idea, but "last seen 8 hours ago" means he's still around to some extent.
 
4:05 PM
@AlanMunn We could talk @JosephWright into mod-pinging him to point him to that question... :)
 
For a few days now, I get the following error when trying to update MacTeX with tlmgr:
/Library/TeX/texbin/tlmgr: signature verification error of /tmp/aLCeg3nB4s/Ia11_pI4zc from mirror.ctan.org/systems/texlive/tlnet/tlpkg/texlive.tlpdb: cryptographic signature verification of
  /tmp/aLCeg3nB4s/5Z1hlqsTOX
against
  mirror.ctan.org/systems/texlive/tlnet/tlpkg/…
failed. Output was:
gpg: Signature made Fri Apr 23 01:54:13 2021 CEST
gpg:                using RSA key 4CE1877E19438C70
gpg: BAD signature from "TeX Live Distribution <tex-live@tug.org>" [ultimate]
Does anyone else have this and how would one resolve this?
 
@Szabolcs you might try a different mirror or reinstallation. But I have no real experience with these kinds of errors. Maybe you want to either wait for someone more knowledgeable in this chat, or (which I'd say is better in this case) ask a real question on the tex.sx.
 
@Szabolcs I haven't encountered this error, but as the text says, just try again, perhaps? Does it happen every time? One possibility is to choose a specific site to use for the repo instead of the mirror. I don't know how the mirror assignment works, but maybe the one you're getting assigned to is out of sync somehow?
@Skillmon I actually think this is a better question for the TeXLive list rather than the site.
 
@AlanMunn Yes, it has been happening every time for a couple of days now. I would have expected that it would fix itself after a few day. Perhaps just wait more?
However, TeX Live Utility works fine. Only tlmgr doesn't. I am not sure if tlmgr is always trying the same mirror, but TeX Live Utility isn't.
Probably just a bad mirror ...
After updating "infrastructure packages" with TeX Live Utility, now tlmgr works fine too.
 
4:24 PM
@JosephWright and what is the replacement for \SIUnitSymbolDegree?
 
@Szabolcs Ah, that makes sense. I only use TeX Live utility and it always checks for updates to the infrastructure.
@Szabolcs You can do that using tlmgr by sudo tlmgr update --self
 
Yes, I did that, but even tlmgr update --self threw the same error.
 
4:45 PM
@Szabolcs Weird. It's strange then that TeX Live utility succeeded.
 
5:16 PM
@UlrikeFischer At the moment there isn't one. With the updated printing approach, most things come out as 'the symbol'. I'm not really happy with the older approach. If it's needed, I can add these back in as part of the v2 support - do you need it?
 
@JosephWright well I have this in one of the documenta (I found the error while trying to debug something else). I can easily add something, but it should then work for version 2 and 3 (side question: how do one force the loading of version 2 if it is in the middle of a large class??)
\DeclareSIUnit\degreeFahrenheit{\SIUnitSymbolDegree \text{F}}
\DeclareSIUnit\degreeCelsius{\SIUnitSymbolDegree \text{C}}
 
@UlrikeFischer I'll give it a bit of thought: the change of printing approach makes it non-trivial
 
@JosephWright every definition to get the correct output would be fine. This code is old, it hasn't to stay this way.
 
5:33 PM
@UlrikeFischer I use
\DeclareSIUnit\degreeCelsius{\siunitx_print_text:n{°} C}
in v3 (with a bit of redirection) - so you want
\DeclareSIUnit\degreeFahrenheit{\siunitx_print_text:n{°} F}
@UlrikeFischer You could use a check for \siunitx_print_text:n I guess
@UlrikeFischer I'm thinking
 
5:52 PM
@JosephWright yes, no problem. I can simply provide \SIUnitSymbolDegree with this definition.
 
@Szabolcs see the archives of the texlive list and use the repostory option to force it to pick up https version of mirror.ctan.org
 
 
1 hour later…
7:08 PM
@UlrikeFischer just load it earlier with \RequirePackage{siunitx}[=v2].
@AlanMunn might be, but the general public will not search the TL mailing list.
 
@Szabolcs http looks wrong, didn't they change everything to https? Oh, @DavidCarlisle said that already ;-)
 
7:59 PM
May 6 '20 at 17:02, by David Carlisle
@UlrikeFischer so slow
 

« first day (3827 days earlier)      last day (1089 days later) »