« first day (1543 days earlier)      last day (3383 days later) » 

cfr
1:08 AM
@yo' ;)
Nobody having problems with CTAN? I get the same thing from 2 machines running 2 different versions of Linux on 2 different networks.
 
yo'
1:19 AM
@cfr well, Nobody is here. :D
 
@cfr I get the same error you reported earlier with a search
 
cfr
1:40 AM
@PaulGessler Good. That makes it a SEP rather than an MP.
@yo' True, but they were when I first asked.
 
 
5 hours later…
7:05 AM
@JosephWright sorry, he wasn't happy with my answer. tex.stackexchange.com/q/224159 Are you taking a look? I asked him in my comment to be patient on what you say...
 
@LaRiFaRi You've got this right
@LaRiFaRi 'Traditional' units don't follow any particular rules, so there is a limit on what can be done
 
@JosephWright I guessed so. You could write that stuff in your package... But: why?
@JosephWright Yes, I understand.
Well, write that as an answer and he will accept. (as he says)
 
@LaRiFaRi It's for example possible to define settings for a named unit, so with \DeclareSIUnit[per-symbol = p]\mph{\mile\per\hour} (untested) you can convert to a particular form, but if he's defined \mile as mi that won't help
@LaRiFaRi You've already answered: I don't know what I can add
@LaRiFaRi There is a get-out which is a bit evil!
 
@JosephWright allright. I will put some quotes from you to my answer and done...
 
@LaRiFaRi The whole point of siunitx is that it deals in consistent formatting. For example, if you say \mile is mi then it should always be mi
 
7:10 AM
@JosephWright You don't have to explain that to me :-) I am one of your packages votaries
 
@LaRiFaRi The only get-out I can think of is you could define \mile as a 'plain' macro not a unit. It could then do a look-ahead for \per and if found for \hour, and depending on the outcome print mi (if the look-ahead fails) or mph (if it succeeds).
 
@JosephWright I was thinking about that to. But I can find no pleasure in programming that right now.
...about that, too. uh, still too early for English...
 
\makeatletter
\protected\def\mile{%
  \@ifnextchar\per
    {\mile@aux}
    {mi}%
}
\protected\def\mile@aux\per{%
  \@ifnextchar\hour
    {mi\per}%
    {mph\@gobble}
}
\makeatother
\documentclass{article}
\usepackage{siunitx}
\begin{document}

\si{\mile}
\si{\mile\per\minute}
\si{\mile\per\hour}

\end{document}
@LaRiFaRi ^^^
A bit risky but will more-or-less work
@LaRiFaRi Should I post that as a separate answer?
 
Why not... I was just adding your quotes to mine, but I stop that.
@JosephWright If he really is in need for that due to his XML stuff, this would be a valid (even if hackish) solution
 
@LaRiFaRi Option b I guess is to do \let\savedSI\SI (or similar) and then just check the argument grabbed. I'll post an answer in a bit.
 
7:19 AM
\sisetup{per-mode=fraction} does not work for your first MWE
@JosephWright Well, have fun! :-) Thank you.
 
@LaRiFaRi Yes, I know: it is a hack
 
7:37 AM
Ayo!
@JosephWright What's the best way to log dates, like a journal entry? At first I thought of using \today but then all the dates will be the same...
 
7:58 AM
@Anthony \logentrydate{2015}{01}{21}, with a suitable definition for \logentrydate.
 
@egreg So this just takes care of formatting then?
Oh also, I'm not using the journal class, if there is one. :P
 
@Anthony Whatever you want. It's not a TeX question: if you want that today's date is logged, it's something your text editor program should do. When you compile a file, TeX can access only the current date, not guess what was the date when the document was typed in.
 
Yeah. Thanks!
 
yo'
8:40 AM
Morning! This has to be a dupe, but I didn't find it ... :(
1
Q: defining drop shadow as makro not allowed?

JürgenIs it really not possible to define drop shadow (with or w/o options) as a makro? \documentclass{scrartcl} \usepackage{tikz} \usetikzlibrary{shadows} \newcommand{\test}{drop shadow={shadow xshift=1ex}} \begin{document} \begin{tikzpicture} \node[rectangle,fill=red,drop shadow={shadow xshift=1ex}...

 
Morning! (Night here. :P)
 
9:02 AM
@yo' It's the old 'keyval methods do no expansion' :-)
 
yo'
@JosephWright well, indeed, but where ? :)
and needed to say, I've done the \edef dirty works before.
 
9:15 AM
@yo' Hmm, we have several questions on this but I'm not sure if we have one 'totally general' one
 
yo'
@JosephWright well, that's the problem. (That's also why I decided to give an answer rather than comment + close vote)
 
@cgnieder Call me crazy, but i would have said the citavi question i rolled back had an answer before that was gone after the rollback. But i didn't really pay attention.
 
@yo' Perhaps we need to set up a general question/answer for duping?
 
yo'
@JosephWright yep. And zap them all! :D
 
@DavidCarlisle Odd think with that issue from yesterday: testing on my Windows machine, the script does stop properly if the errorlevel isn't 1. There might be something odd going on with os.exit that only shows up on Unix systems. Will test later today on my Mac.
 
9:20 AM
@JosephWright I read the comment about \XeTeXcharclass when the train was approaching the station: \afterassignment was obviously the answer. Too bad people keep asking XY questions.
 
@egreg Well yes
 
yo'
Readings from Mathematics: If you travel back in time and paradoxically become your own grandparent ...
101
A: Can you be 1/12th Cherokee?

Hagen von EitzenThis depends on the model. Instead of arguing that we have only $46$ chromosomes and cross-overs or whatever the mechanism is called are not that common, let us assume a continuous model. That is, a priori, everybody can be $\alpha$ Cherokee for any $\alpha\in[0,1]$ and the rules are as follows ...

 
Chuck Norris is 1/12 Cherokee. — Asaf Karagila 2 days ago
 
yo'
@Johannes_B LOL
 
@egreg Doen't line-end normalisation predate web2c: Knuth says something about it in The TeXbook I think
 
9:32 AM
@JosephWright Knuth refers to a different thing. With TeX implementations that didn't use Web2C (OzTeX, for instance), files coming from DOS ended up with \par between any two lines.
 
@Johannes_B I don't think it had but I'm not sure either
 
@egreg Ah
 
@JosephWright tex itself makes the whole thing system dependent and says the implementation must report the "lines" with trailing white space removed. So it worked on systems that had a record based structure without a specific line separation char at all, so I think it is web2c that decided to uniformly normalise line endings on the platforms it supports
 
9:47 AM
@JosephWright you got mail:-)
 
9:57 AM
@DavidCarlisle, @egreg One of you going to tackle the web2c question?
 
@JosephWright @egreg's answered in comments so just make that an answer I think
 
 
2 hours later…
yo'
11:55 AM
@PauloCereda Quack Morning! :)
 
@yo' Hi Tom!
 
yo'
my shirts arrived today, yay! :D
 
@yo' TeX.sx? :)
 
yo'
@PauloCereda no, other funny stuff :)
 
@yo' Kiss me I'm Czech type of thingy? :)
 
@yo' Holy shirt, Batman!
I so want the first one!
 
yo'
@PauloCereda it's really funny, isn't it? :) and they're super comfy as well
it's the French winter sales thingy, and unfortunately they already didn't have the ones I wanted in my size :(
 
 
1 hour later…
1:19 PM
Oh, my! A tabularx inside a footnote! With a footnote to the table!
3
 
/facepalm
 
@yo' Yes, kind of! :)
 
yo'
I was looking for the Chinese/Japanese tea box drawing with a lady that carries the same tea box with the same lady ..., but I didn't manage to find it
 
@PaulGessler I should flag this image as offensive. :)
@egreg Burn the document! :)
 
1:30 PM
@PauloCereda Send the author to the Spanish Inquisition!
 
@PauloCereda quite! :-)
 
@egreg excellent use of my code I assume (haven't seen it:-)
 
1:51 PM
@DavidCarlisle does it work? :)
 
yo'
@PauloCereda seeing the message above, probably not :p
 
@yo' :P
 
2:31 PM
someone here who talks bahasa, indonesian, indon, bahasai, bahasam, malay, meyalu?
getting tedious to share my non-knowledge here: tex.stackexchange.com/a/224234
 
yo'
3:22 PM
I think I've got a result! (well, a partial one: we've got two conjectures (C1) and (C2) and I think I've got a proof that (C1) => (C2))
this means I'll have something to sumbit to teh conferencez
 
Anybody with knowledge on LaTeX and tablets? latex-community.org/forum/…
Just read this question tex.stackexchange.com/questions/213258/… that was on the top of main and thought, that's gotta be a dupe. There was some patching involved. Turns out, i answered that question.
 
3:37 PM
@JosephWright fixed a few more typos in ltnews, but nothing requiring a rebuild I think:-)
 
yo'
3:47 PM
Is there a way how to install vanilla TL2012?
 
@DavidCarlisle I'm sure there will be more :-)
 
yo'
4:13 PM
@DavidCarlisle ah thanks
 
@yo' that's bad news, it means now you have only got one conjecture
 
yo'
@DavidCarlisle Well, the point is, (C1) itself is true certainly, we just don't have a proof, and (C2) is the interesting one
 
@yo' :-)
 
5:03 PM
@JosephWright new luatex I see....
 
@DavidCarlisle Yes
@DavidCarlisle Will be interested to see what impact this has once I install TL2015 pretest
 
yo'
5:50 PM
Do you have any trick how to switch different versions of TeXLive quickly?
 
@yo' Path priority. :)
 
yo'
@PauloCereda well, I'll see if I manage to put that in different options in TeXworks :)
 
@yo' oooh :)
 
yo'
@PauloCereda well, you can create pdflatex10.sh with something like PATH=.../2010; pdflatex $1 $2 $3 $4 $5
[root@toheshiba ~]# cd /usr/local/texlive/
[root@toheshiba texlive]# ls
2010  2011  2012  2013  2014  texmf-local
^^ testing can begin :D
 
@yo' LOL
 
yo'
5:55 PM
@PauloCereda It's important, I have to see what I can do before I agree it will be TL2012-compatible
anyways, the TCS+ talk starts in 3 minutes and I need to get a coffee yet. See you later!
 
6:35 PM
@yo', @JosephWright I don't think saving the original optional argument so a KV option handler can use it instead of the mangled one works. Just saving the tokens doesn't help you also need to make them safe in the existing option handler, and I'm not sure that's possible in any compatible way:
\documentclass[foo={a,b,c}]{article}

\stop
 
@DavidCarlisle Yes, my concern too
@DavidCarlisle Read kvoptions-patch?
 
@JosephWright of course not:-)
@JosephWright see the start of the history section in that:-)
 
@DavidCarlisle I like the introduction :-)
@DavidCarlisle Ah
 
@JosephWright noticed that, I got to about page 4, then I thought "I think I wrote something like this", skipped to history section and saw that confirmed..
 
7:09 PM
@JosephWright Show off. :P
 
yo'
7:20 PM
@DavidCarlisle what's wrong with this (other than unused options warning)?
\begin{filecontents}{aaa.cls}

\RequirePackage{expl3,l3keys2e}

\ProvidesExplClass{aaa}{2014/01/18}{0.1}{Class for typesetting CTU theses and alike}

\keys_define:nn { aaa } {
	mykey .code:n = \def\bbb{#1},
	myotherkey .code:n = \def\bbbccc{#1},
}

\keys_set:nV { aaa } \classoptions

\LoadClass{article}

\end{filecontents}

\let\origdocumentclass\documentclass
\def\documentclass[#1]{\def\classoptions{#1}\origdocumentclass[#1]}

\documentclass[ mykey = {foo, bar}, myotherkey = baz ]{aaa}

\begin{document}
 
@PauloCereda The book arrived!!!
 
@Brent.Longborough Yay! :)
@Brent: I should never doubt the Royal Mail. :)
 
@PauloCereda I just spoke to my "manager" at home. She said the wrapping was a bit torn, so she opened it, then thought the book looked very interesting, and has read a lot of it already! I know it's Tschichold, but Google hangouts fell over before I could discover whether it's in English, Portuguese, or German?!?
@PauloCereda Thank you. A really wonderful present in any language.
 
@Brent.Longborough Believe it or not, it's Portuguese! :)
@Brent.Longborough My pleasure, my dear friend. :)
 
@PauloCereda Ah, that's why she kept referring to "A Forma do Livro" LOL
 
yo'
7:25 PM
@Brent.Longborough what an euphemism :)
 
@PauloCereda As we normally talk in Portuguese, I couldn't tell whether or no she wqas translating the title !!!
 
@Brent.Longborough LOL
 
Unbelievable. I was in the middle of shutting down for bedtime, but I had tocome bak to say thank you.
 
@Brent.Longborough No Welsh? I'm disappointed. :)
@Brent.Longborough <3
 
OK, I will now amaze you all by turning into a pumpkin in Comic Sans. Goodnight all...
 
7:29 PM
@Brent.Longborough G'night, Brent! :)
 
7:43 PM
@DavidCarlisle Do you think it would be OK to change ltvers to do \everyjob\expandafter{\the\everyjob\typeout{\fmtname (i.e. keeping anything present at the start of latex.ltx)?
@DavidCarlisle Context: there's one thing outstanding from lualatex.ini: the 'quote the job name' business. My feeling is that the code itself there is web2c not format related, so still belongs in the .ini, but it would be easiest handled if we leave anything present in \everyjob at the start of the format building alone.
 
@yo' yes if you don't use the standard option handler at all it's a bit safer. But most classes want to pass options to article (for 12pt or twoside or whatever) and I'm not at all sure that the behaviour in most cases is documentable:-) will think about it.
 
@DavidCarlisle Heiko's code uses e-TeX a lot, and it's quite dense
 
hhh
This problem of not getting references compiled in OSX is appearing again. How is it with your OSX? Can it be caused by some update in OSX? (I just updated everything with Texlive Utility and did not fix the problem so now puzzled how to approach this problem of not getting references compiled).

http://tex.stackexchange.com/q/212540/2956
 
@JosephWright what a shock:-)
 
yo'
@DavidCarlisle well, you can allow the class to handle it itself (including \LoadClassWithOptions and stuff. The classes would have to choose: either completely what it was until now, or completely the new thingy.)
 
7:56 PM
@DavidCarlisle I've added a couple of .ini files to DropBox, but the LuaTeX relies on fixing that last point
 
@JosephWright probably, was just reading it now,
@JosephWright how come dropbox told me that you'd added the ini files, did you explicitly share that with us as well as giving the http address by mail?
 
@DavidCarlisle I shared the folder: I think if you grab them while logged into DropBox it 'remembers' that you have access
 
@JosephWright anyway they look lots nicer than the existing ones!
@JosephWright everyjob looks OK, probably should flag it on team list though
 
@DavidCarlisle Yes, just thought I'd ask first in case there was anything to watch out for
 
@JosephWright no, don't think so, other place of course to do it would be the end (where ltpatch.ltx does/did it) but so far we've avoided having a cfg file at that point so doing it at the start is OK I think.
@JosephWright do you think it would be worth starting the extended allocations at (say) 266 rather than 256 so leaving 10 for special uses (etex did something similar for its counts for the local/global stuff. In particular could then steal 256 and 257 for use in \newmarks and \newxetexcharclass which would save using \newcount for those which uses up a classic register and changes all the numbers in the test logs....
 
8:10 PM
@DavidCarlisle Sounds fine to me
 
8:51 PM
@JosephWright mail sent, not sure who's around apart from us though:-)
 
@DavidCarlisle Indeed
@DavidCarlisle Like I said, plan sounds solid to me and does leave space for a 'second round' for LuaTeX-specifics (some stage in the future)
 
@DavidCarlisle Ah yes, that's the one
@DavidCarlisle Odd think thought is presumably for pdfTeX it's done at the web2c level not in the original sources
 
@JosephWright I think it's in the web change file though (or at least I think I recall seeing something about that once) so presumably not code shared with luatex anymore
 
@DavidCarlisle Probably
 
9:35 PM
@DavidCarlisle Great irony of course is that for the end user of LuaTeX/LuaLaTeX it actually acts the same :-) (ConTeXt MkIV of course different)
 
@JosephWright one suspects for the developers, it's only really the last of those that matters;-)
 
@DavidCarlisle I've had some interesting direct mails, to be sure
 
@JosephWright re bug reports?
 
@DavidCarlisle Indeed
@DavidCarlisle And re blog posts
 
@JosephWright ah
 
yo'
9:41 PM
... when you think about something for 3 days, and the solution is a graph on 10 vertices with 24 edges ...
 
9:52 PM
@yo' you could use this in your paper ^^
 
Anyone know what 'Xing' is?
Just stumbled across it on the CTAN website
 
@JosephWright context? (it's crossing in US english)
 
Hi
 
@JosephWright not this? ctan.org/author/xing
 
good evening here
 
9:56 PM
@DavidCarlisle No, see logo at bottom of CTAN page
@barznjy Hello
 
I have a question about figures in latex;
How I can reduce the space between the figure and its caption?
 
@JosephWright www.ctan.org? don't see anything
 
@JosephWright A name?
 
@barznjy depends on class/packages but in article you can \setlength\abovecaptionskip{2pt} (default is 10pt)
 
I use the following
\begin{figure}[htp]
\centering
\includegraphics[scale=0.53]{fig1}
\caption{System model}
\label{fig:fig1}
\end{figure}
 
10:00 PM
@DavidCarlisle Perhaps it's just me, then
@DavidCarlisle Ah, reloading it's off on the left-hand side and 'grows' when you go near it
 
@JosephWright I see it in view source: xing.com
 
@DavidCarlisle what about if I use IEEE style?
 
yo'
@barznjy do you prepare a conference or journal paper?
 
conference
@yo' it is for IEEE conferences
 
yo'
@barznjy then the answer is simple: don't do it or it can be a reason for rejection.
 
10:14 PM
@yo' may be you know that some of the conferences restricted the number of pages, therefore, you don't want to have an extra page for one sentence.
 
yo'
@barznjy I do know this. Try \usepackage{microtype} that one hasn't been blacklisted yet I think
or make the figure itself slightly smaller
 
@yo' but how they know if it is used or not? because you are usually submit pdf file
 
yo'
I wouldn't dare juggling with the page layout
@barznjy well, if you submit a pdf file then you're probably safe.
 
@yo' all IEEE conferences accept only pdf submission
 
yo'
You can try putting \vspace*{-1.5\baselineskip} between the figure and the caption (separate it by an empty line from the figure)
 
10:18 PM
@yo' so, could you please give me some detail of how I make the space between the figure and its caption smaller ; for example for this case

\begin{figure}[htp]
\centering
\includegraphics[scale=0.53]{fig1}
\caption{System model}
\label{fig:fig1}
\end{figure}
 
yo'
@barznjy you know how it works: Create a MWE and then it's simple. Without it, it's impossible
 
@barznjy see the comment on the line above
 
@DavidCarlisle I used your command but nothing changed
 
@barznjy Yo's command will work in any class
 
yo'
@DavidCarlisle well, it's just a very dirty trick, these work the best.
 
10:22 PM
@yo' whereas mine is pure and elegant and not working:-)
 
yo'
@DavidCarlisle OTOH, there're no bugs in your packages whatsoever
 
@yo' of course not
 
@yo' It works, great ..... many thanks for both of you
 
yo'
@barznjy you're welcome. However, if you prepared a MWE next time, it would have taken us together half of the time and effort to solve the problem
 
@yo' but less opportunity for mutual insults
 
10:25 PM
@yo' what is MWE ?
 
yo'
@DavidCarlisle sure. Now excuse me, I should leave my office, it's 11:30 PM
 
minimum, working example
 
yo'
174
Q: I've just been asked to write a minimal example, what is that?

Juan A. NavarroI was posting some question about this strange thing LaTeX is doing when I try to compile my thesis. Someone asked me to provide a minimal example that reproduces the problem. My thesis is now a few hundred pages long and spans along ten different source files! How am I supposed to know what or w...

 
@yo' @DavidCarlisle Thanks again.
 
hhh
11:00 PM
Can someone help me to debug the problem where I cannot get the bibtex compiled? I coved this earlier in this question but the same prob reappeared -- is there someone in OS X that could test this on their machine?
 
11:21 PM
@JosephWright: Do you have an interest to add tbtags as discussed in the following link?
@Bugbusters Yes I pinged Joseph to see if he wanted to add a tbtags option to mathtools to match:-) — David Carlisle Apr 18 '13 at 20:06
 
11:43 PM
@hhh Maybe it's the common problem of the corrupt cache of Biber.
63
Q: Biblatex/biber fails with a strange error about missing recode_data.xml file

Alan MunnI'm trying to use biblatex and when I compile my document I get an error like the following: Windows data source C:\Users\<username>\AppData\Local\Temp\par-5061756c\ cache-890efc00b3ca6b775c7d44a325c1349fb2a3a3bd\inc\lib/Biber/LaTeX/recode_data.xml not found in . Mac data source /var/folders...

 

« first day (1543 days earlier)      last day (3383 days later) »