« first day (2521 days earlier)      last day (2409 days later) » 

1:27 AM
Scrap, looks like everybody else is AFK. I'll come back tomorrow or something…
Wouldn't hurt to loiter around for a bit to see if anybody shows up, though… (Sits down.)
 
 
1 hour later…
2:28 AM
@DavidCarlisle @egreg it was a really tough decision, i just chose egreg's because...it was more expanded out, and easier (somewhat) to understand. actually, the answer still feels like voodoo, and i feel bad using it.
plus, i know @egreg lives for the ticks
5
 
 
4 hours later…
6:53 AM
@bertieb You probably want to use \supercites then (note the extra s). That allows you to have separate prenotes/postnotes for each citation: \supercites[42]{citeA}[3]{citeB}
@bertieb Forget that, I was wrong. \supercite(s) doesn't allow for pre/postnotes, so they are discarded.
 
Quack!
Shaky bus time! :)
 
7:12 AM
@PauloCereda Hey pal, what's up?
@PauloCereda I have a (not so quick) question.
@PauloCereda How does Arara parse the rules in the preamble?
@PauloCereda Is this some kind of config syntax, like INI or JSON?
 
@HenriMenke Hi Henri!
@HenriMenke ooh a tricky question.
@HenriMenke Before the conditional, it's YAML. After the conditional, it's MVEL. :)
 
When typesetting opposite pair of functions f\colon X\rightleftarrows Y ? g, is there an appropriate colon-variant to be used for "?"?
 
@PauloCereda Conditional were introduced in v3.0?
 
@HenriMenke no, version 4.0. Version 3.0 only relies on YAML, which is basically a map.
 
@PauloCereda So it's MVEL now, I reckon.
 
7:19 AM
@HenriMenke Sorry, I think I wrote it badly. Let me rephrase.
@HenriMenke: the part before the conditional keyword is YAML. The part after it is MVEL.
Directives are still YAML.
 
@PauloCereda Brrr
2
 
@HenriMenke Sorry. :)
 
@PauloCereda I'm trying to reimplement Arara as an exercise for myself and I was trying to keep it simple and idiomatic :)
 
@HenriMenke the YAML part is basically a map, in which everything you want is put inside a key/value structure, which is purely syntactic. After the conditional keyword, I need to have semantics, so I went for MVEL as a means to interpret what is there and come up with a boolean result. :)
@HenriMenke Oh that sounds fantastic!
@HenriMenke I will think of something and poke you later on, I have shaky bus woes in a few minutes. :)
 
8:02 AM
@PauloCereda I've sent you an invite to my private GitHub repo.
 
@HenriMenke in emacs lisp I hope? @PauloCereda should have used that in the first place!
@user89 well that is true, but you shouldn't feed the trolls:-)
 
8:31 AM
@PauloCereda I bet Psmith knows more about cricket:
 
9:19 AM
When I installed MacTeX, it added a certain directory to the system PATH. Where is this set? It is not in ~/.profile.
 
@TorbjørnT. Ah, good thought though!
 
9:35 AM
@Szabolcs Look in /etc/paths.d/.
@Szabolcs Also check man path_helper.
@Szabolcs And /etc/profile.
 
@HaraldHanche-Olsen It is indeed there (paths.d). Do you know when these paths become active? In a Terminal session they are. But if I start a shell from some other application (in non-interactive mode), then they are not.
 
@Szabolcs Assuming you are running bash? It reads /etc/profile on startup if it is a login shell, or is invoked with the --login option. See man bash for the exact conditions.
 
The --login option does cause it to pick up these paths. Thank you.
 
 
3 hours later…
1:00 PM
@Szabolcs The path is set in the /etc/paths.d/TeX file
 
yo'
1:14 PM
@PauloCereda Are you around please with couple minutes spare?
 
 
1 hour later…
2:38 PM
@yo' Hi Tom! I am in SP, but I always have time for my friends! :)
 
yo'
@PauloCereda running fedup and getting messages like this:
Downloading failed: The GPG keys listed for the "Fedora 21 - x86_64" repository are already installed but they are not correct for this package.
Check that the correct key URLs are configured for this repository.


 Failing package is: libpurple-2.12.0-2.fc26.x86_64
 GPG Keys are configured as: file:///etc/pki/rpm-gpg/RPM-GPG-KEY-fedora-21-x86_64
 
@yo' Hm hold on, let me check.
@yo': what's your underlying system, F21?
 
yo'
I thought I would find something on the net, but nothing really relevant so far :-/
@PauloCereda well, moving from F20 to F21
(don't say I'm quite late with this, I know)
 
@yo' I see.
@yo': I believe little to nothing can be done in this specific scenario. If I recall correctly, fedup has been discontinued and I believe there are no repositories with proper support anymore. :(
 
yo'
@PauloCereda and how do you upgrade then?
 
2:49 PM
@yo': prior to F24, I always did a clean install of my systems. From F24 on, dnf has introduced a system upgrade.
 
yo'
@PauloCereda ah ok. So my luck is low and you seem to imply I should forget this and purge the hard drive?
 
@yo' Nowadays, dnf takes charge of it. I might ask on the #fedora channel about upgrading an old distro and report back to you. :)
@yo' I think newer versions have better support for upgrades. :) I will investigate that with the Fedora devs and report back to you, don't worry. :)
 
yo'
@PauloCereda ok. well, it seems that if I yum reinstall the problematic package, I get to another one. The problem is that it takes about 5 minutes to get there and I have no idea how many are there.
 
@yo' Oh I see, it might be a slow server. I don't remember if yum has a dry-run option, let me check.
Hm they suggest the --downloadonly flag... But that's not what I want.
@yo': I have to go offline for a moment, I will be back soon!
 
yo'
@PauloCereda but yum is not failing, it's fedup who is
@PauloCereda ok
so no, unfortunately the problem persists even after a reinstall. Something is broken with the keys :-(
 
 
3 hours later…
5:48 PM
What is the correct LaTeX3 regex to match \\ (the newline command)?
\regex_count:nnN { \\\\ } { foo\\bar } \l_tmp_int \int_use:N \l_tmp_int yields 0
\c{\\} is the correct one
 
@yo' back from the shaky bus. Tom, is ignoring the key a valid alternative?
 
quack
 
6:04 PM
Quack!
 
yo'
6:46 PM
@PauloCereda well, if you tell me where to tell yum to ignore it, as it's not me invoking yum directly, it's fedup invoking it... (And moreover I'm on my phone only and on the way to bed as I'm really tired)
 
 
2 hours later…
9:02 PM
@Skillmon Yes, that's the right one: \c{chars} matches a control sequence with name chars, that is \chars; the backslash must be escaped.
 
@egreg I just tried to match the literal and forget that it differentiates between control sequences.
 
@Skillmon \\\\ matches two “ordinary” backslashes.
\documentclass{article}
\usepackage{expl3}

\begin{document}

\ExplSyntaxOn
\cs_generate_variant:Nn \regex_match:nnTF { nV }

\tl_set:Nn \l_tmpa_tl { \\ }

\regex_match:nVTF { \\\\ } \l_tmpa_tl {TRUE} {FALSE}\par

\regex_match:nVTF { \c{\\} }  \l_tmpa_tl {TRUE} {FALSE}\par

\tl_set:Nx \l_tmpa_tl { \token_to_str:N \\ }

\regex_match:nVTF { \\\\ } \l_tmpa_tl {TRUE} {FALSE}

\end{document}
@Skillmon ^^^^
This prints FALSE TRUE TRUE
 
10:08 PM
@egreg I know you like Bruno's regexp code, but I was thinking more of using some variant of \newunicodechar to map the inputenc utf8 straight to the ascii text, I suppose space would have to be handled separately but...
 
@DavidCarlisle That's essentially redefining \u8:...
 
@yo' sorry, I could not find the proper key for fedup.
Grrr lost two USB sticks on the shaky bus...
 
@egreg possibly. I didn't get as far as details of the code:-)
@PauloCereda thesis gone with them?
 
@DavidCarlisle thesis is safe. :)
 
@DavidCarlisle The list of substitutions needs to be provided anyhow. But it would be expandable.
 
10:12 PM
@egreg yes as I commented to the OP you need a long list anyway so it doesn't make any difference as far as usage
@PauloCereda can you explain Java to me?
 
@DavidCarlisle any time. :)
 
I got something to work by not following the documentation, but I don't know if the documentation is wrong
 
@DavidCarlisle Hm.
 
@PauloCereda hang on, I'll collect some links...
 
@DavidCarlisle Poked you in H.
 
10:32 PM
@UlrikeFischer Thanks for suggesting the prison duck, it came in handy for tex.stackexchange.com/a/393233/36296 :)
 
10:43 PM
user image
3
@samcarter Wonderful!
 
@egreg ooooh
 
11:12 PM
@PauloCereda thanks:-)
 

« first day (2521 days earlier)      last day (2409 days later) »