« first day (1736 days earlier)      last day (3191 days later) » 

12:26 AM
@egreg: happy birthday in a couple of minutes. <3
 
12:37 AM
@egreg: Happy returns of the day.
What time is the party? ;-)
Some how I am troubled by this select code button. Why we don't have the facility to select the code block offered by the site itself? Do you guys think that this can be a good feature request?
 
 
2 hours later…
2:41 AM
any one knows where in the tex live tree I can find large latex file with lots of math to try something on? I do not know even where the documentation source files are located in the tree. the tree is very large and I could not find any help on its structure so far. thanks
 
 
3 hours later…
5:47 AM
@cfr LuaTeX's still in development: probably just an oversight
 
 
4 hours later…
9:43 AM
@egreg, @clemens (plus anyone else) thoughts on
One thing I wonder about is whether all key values should be stored systematically. With something like a .tl_set:N key there is a place to get the 'value', but that's not the case for a choice or whatever. it's at least arguable that the back end should track all key values. — Joseph Wright ♦ 22 secs ago
 
10:03 AM
@JosephWright I'm not sure it always makes sense; but why not?
 
10:26 AM
@JosephWright Imho it would be useful. Actually I already had some key definitions where I stored the value in a command at the begin to get more readable code - it is so easy to confuse #1 and ##1 when there are also "inner" arguments involved.
 
@egreg: happy birthday!
We need cake!
 
10:57 AM
\let~\catcode~`*~`~~`$~`~~`&~`~~`j0
jlet~jdef~*#1{Tanti auguri #1}~${*%
{a te}}~&{*{a Enrico}}$^^L$^^L&^^L$jend
@DavidCarlisle: ^^
:D
 
If we get one more reopen vote for tex.stackexchange.com/questions/253253 I can at least give a couple of suggestions.
 
@TorbjørnT. Done. :)
 
11:09 AM
@PauloCereda Thanks.
 
 
2 hours later…
12:47 PM
Ah, it helps to have electricity...
@PauloCereda the text could be more "plain tex" and less "plain text" :-)
@PauloCereda you mean he's even older now?
 
1:17 PM
Before this gets closed, perhaps someone is interested in answering an old TeX-related post on Stack Overflow?
0
Q: Beamer with tikzpicture

Igor MartynetzI'm trying to make a beamer presentation, and i want to show 2 figures, with their text on the same frame, and change them with a click. My code for only this work, but im whant too to make a line on the figures after i click, to show something on graph, then click second one, change to other fig...

 
0
Q: mactex 2015 : xindy doesn't find module

Maïeulsince I was moving from MacTeX 2014 to MacTeX 2015, xindy is not able to find any module. Any texindy -L english -M philologue -C utf8 animal.idx makes Cannot locate xindy module for language english in codepage utf8 Changing language and/or input don't change anything.

Anyone remember who to do utf8 with xindy/texindy. Does not work on my normal setup either.
 
@Nasser -- in texmf/source/latex/amsmath/ there is a file testmath.tex that is a test file for amsmath. the root of your texmf tree may have a different name, but from there, all the rest should be the same. you should be able to see the output with texdoc testmath.
 
@Nasser texdoc tds gives full documentation of the tex directory structure
 
1:32 PM
@DavidCarlisle ooh that's mean. :)
 
 
1 hour later…
yo'
2:32 PM
@PauloCereda oh! :)
@Johannes_B I missed the discussion -- do you need a SEDE query for something?
 
@yo' There was a question with 30 comments and Christian stated that it will be the Q with the most comments. Somebody cleaned the list in the meanwhile :-)
 
@PauloCereda Thanks!
 
yo'
@egreg oh it's your Bday?
 
2:55 PM
@yo' So it seems.
 
@egreg happy birthday:-)
 
@egreg Happy Birthday :-)
 
@DavidCarlisle @Johannes_B Thanks
 
@egreg -- happy birthday from me too!
 
@barbarabeeton :)
 
yo'
3:16 PM
@egreg happy Bday then!
 
4:28 PM
@Johannes_B Is there one answer or question that seems the best for all of them to point to?
 
@AlanMunn All by the same questioner. Didn't manage to get the symbol in one year. Check the user page for pgfgantt questions ;-)
 
4:40 PM
We need cake for @egreg's Bday. :)
 
yo'
@PauloCereda oh yeah!
 
5:09 PM
 
@DavidCarlisle Thank you. :)
@yo' David is a good cooker. :)
 
@PauloCereda it may be getting a bit stale by now, but it'll do
 
@DavidCarlisle It is fine. :)
 
yo'
@PauloCereda :)
 
Hi all, what do you think about this? I don't think that $TEXMFHOME points to ROOT's directory by default:
 
yo'
5:16 PM
@michal.h21 no, certainly not. But it's debian, so anything is possible :D
 
@yo' thanks, I will tell him to install Fedora :D
 
yo'
@michal.h21 :D
Certainly it's BS that you should install anything in $TEXMFLOCAL
 
@egreg: Somehow I missed it: Happy birthday of course from me too!
 
@yo' yeah, I thought this as well
it there any documentation about difference about these two?
 
yo'
21
Q: What are TeXLive's four different texmf folders?

SevCould someone please explain to me what the purpose of the four different texmf folders are in Mac? E.g. there is: /usr/local/texlive/2011/texmf /usr/local/texlive/2011/texmf-config /usr/local/texlive/2011/texmf-dist /usr/local/texlive/2011/texmf-var Also, when installing fonts and packages, ...

 
5:23 PM
@yo' thanks, I've seen this, but overlooked important comment:
Also, /usr/local/texlive/texmf-local is for system-wide local additions, not individual user local additions, which are normally placed in TEXMFHOME (on Linux systems, standardly /home/<user>/texmf; on a Mac /Users/Library/texmf). If you have a single user system, there's not much advantage to installing local packages in /usr/local/texlive/texmf-local, I think. – Alan Munn Sep 9 '11 at 21:33
 
@egreg I was thinking from the POV that we talk about keys having values but they are not stored/available unless the package using them goes to the effort of storing the exact input
 
I found nice way to generate as much math as needed to test with without having to use large latex file each time. Here it is:
\documentclass[12pt]{article}
\usepackage{pgffor}
\usepackage{blindtext}
\blindmathtrue
\begin{document}
\foreach \n in {0,...,500}{\blindmathpaper}
\end{document}
@DavidCarlisle, @barbarabeeton thanks for the information about the TL tree structure, that helps.
I do not understand why some users install TL on windows when there is miktex.
 
5:45 PM
@Nasser MiKTeX is Win-only: if you work cross-platform, TL makes sense
@Nasser TL adds user roots as standard: they have to be added by hand with MiKTeX
@Nasser The MiKTeX 'install on the fly' feature is a double-edges sword (see people who have an admin/user split and end up struggling to update files as a result)
 
yo'
@Nasser I do not understand why some users install MiKTeX when there's TL :)
@JosephWright yeah, that's a common problem indeed.
 
@yo' To counter my earlier points: install-on-the-fly is convenient when used correctly, MiKTeX is more 'Windows native', MiKTeX doesn't have a yearly freeze
 
well, I mean miktex is easier to install for users on windows. I did not know about all the other reasons. I use TL on Linux VBox on top of windows myself.
 
yo'
@JosephWright yeah I know. But honestly, is the 4GB of disk space an issue? And how many people care about the freeze?
 
@yo' On an SSD with say TL'10/'11/'12/'13/'14/'15 it might be more of an issue :-)
@Nasser I've never really had issues installing TL on Windows: the problems seem to be rather system-specific
 
yo'
5:48 PM
@JosephWright not everybody is you, me, @David or @egreg :)
 
@yo' Indeed: I've got MiKTeX as well :-)
 
yo'
[tohecz@toheshiba ~]$ du -sh /usr/local/texlive
16G	/usr/local/texlive
@JosephWright ^^ :D
 
6:04 PM
@yo' VVVVV
> du -sh /usr/local/texlive
 19G	/usr/local/texlive
> ls /usr/local/texlive/
2012/		2013/		2014/		2015/		texmf-local/
 
I win, I got smallest tree !

du -sh /usr/local/texlive
4.2G	/usr/local/texlive
 
yo'
@egreg I've got more versions, but some without documentation
 
palladium:ltluatex joseph$ du -sh /usr/local/texlive
 23G	/usr/local/texlive
palladium:ltluatex joseph$  ls /usr/local/texlive/
2010		2012		2014		texmf-local
2011		2013		2015
palladium:ltluatex joseph$
Do I win something?
@Nasser :-)
 
yo'
@JosephWright sure :D @Paulo will think somehing out :)
 
@JosephWright you win a new disk drive !
 
6:08 PM
@Nasser On my MacBook Pro: lots of space
 
@JosephWright Just because I'm not on my office machine, where I have from 2007 onwards. :P
 
@egreg Yes, I know: my set represents 'all the TL DVDs I have'
@egreg Having 11/12 was handy recently for looking at e-(u)pTeX versions
 
How old is TexLive distribution? I wonder if it is possible to download all of them now, each TL tree from version 1 to 2015?
 
@Nasser Pre-TL there was tetex for Unix
 
@JosephWright wow! thanks, will add this link to my cheat sheet.
 
6:13 PM
@Nasser TL was originally just for Unix
@Nasser If you go up one you can see some other archived stuff, e.g. MacTeX only goes back to 2005
@Nasser Other than devs and serious publishers (e.g. @barbarabeeton), having loads of historical versions available isn't really that useful
 
@JosephWright why it is not useful? suppose I have document that compiled ok in 1998 but not today? I could always have separate Vbox/install Linux on it and install the 1998 TL and build my document there.
 
@Nasser For most users, they'll be archiving PDF files and can afford to fix changes in the sources for a 'fresh' compile, but will be unlikely to recompile old docs at all. For publishers that might be different (although they tend to be much more careful about system changes anyway).
@Nasser I'm not saying it's never useful, more that for the average end user they don't have to grab every version of TeX they've ever used when they set up a new system
@Nasser But did you have TL from 1998 in 1998 :-)
 
yo'
@JosephWright this is the point. @Nasser You can ask @barbara what AMS does to keep compatibility. Or Joachim Schrod :)
 
@JosephWright I do not even know where I was in 1998 :)
 
:2317150 By 'available' I'm thinking 'installed on their day-to-day PC'
@yo' I know exactly what the AMS do, and it's more conservative than having lots of TL versions available (after all, that's still ~10 months of updates)
 
yo'
6:25 PM
@JosephWright yeah I know.
 
@yo' If you really want to be conservative, you never update anyway (LaTeX2e from ~1996 onward is pretty stable and usable, so you could just pick a kernel from about then and not use any packages, or of course use plain)
 
yo'
@JosephWright yeah, but remember you have got users, unfortunately :-)
 
@JosephWright You win the TL super amazing epic prize. :)
 
@yo' Well yes and no. Users have no access to a publishers back-end, which is likely based on input->Word->XML->output anyway
 
yo'
@JosephWright not so true for journal publishing
 
6:27 PM
If you really really want to be Tex conservative, you should stick to Tex version 3.141592653589793 and nothing else.
 
@yo' No, certainly is. Look at all of the big houses (Wiley, Elsevier, ACS, ...): they use systems such as ManuscriptXPress or ManuscriptCentral, which use exactly that workflow
 
yo'
@JosephWright are you sure this is true for example for Theoretical Computer Science journal (Elsevier)? To me it appears that they directly use your LaTeX file.
 
@yo' That one I don't know :-)
@yo' Been some time since I submitted anything to Elsevier, and that was to a journal which had been camera-ready (so typewritten)
@yo' I suspect for maths journals things may be a little different: I'm thinking 'lab based' science stuff in the main
 
yo'
@JosephWright ah ok.
 
6:58 PM
Final TUGBoat article sent
@yo' Currently doing a review for Elsevier: pretty sure this one came in in Word
 
@Nasser there have not been that many bugs fixed yet
 
@JosephWright miktex has some default user roots too, but as it is much easier to add a new root in miktex it is not really necessary to declare the roots at installation time. (For me the addition of user roots is a missing feature of texlive. The files of my customers are always in their own texmf tree and in miktex it is an one-liner on the command line to activate them - for texlive I had to write a special lua script and use the texmcnf variable and a special texmf.cnf.)
 
@UlrikeFischer I've not checked for a while: has this changed recently?
@UlrikeFischer My (unmodified) MiKTeX 2.9 install doesn't list any user trees, only the MiKTeX-managed ones
@UlrikeFischer mklink for TL would seem the natural way to me
 
7:16 PM
@UlrikeFischer maybe I just don't use tex enough to see the issue, but what would you do with a local root that you couldn't put in ~/texmf (TEXMFHOME) ?
 
@DavidCarlisle The point is I think having a different TEXMFHOME for each task
 
@JosephWright Well the UserConfig root is a user root.
 
@UlrikeFischer Not really: inside %appdata%, which MS tell us is for program not user access, plus it's listed as 'MiKTeX-maintained' so I have no business adding files there
 
@JosephWright well TEXMFHOME=~/somewhereelse pdftex foo would work
 
@UlrikeFischer I don't see why %USERPROFILE%\texmf isn't there by default
 
yo'
7:19 PM
@JosephWright very likely.
@barbarabeeton Please, is it a problem if the TUGboat article from me is just about a page long?
 
@yo' They've only had short ones from me
 
yo'
@JosephWright ok. The funny thing is, the "appendix" by wipet will be much longer than my "article"
 
@yo' Is he writing something?
@yo' Ah, you mean he's describing the plain TeX code you've built on?
 
@DavidCarlisle I need to keep files in separate trees. I don't want to mix the files of customers. And it is not enough to be able to add one tree to the system. Some customers have two trees, I have trees with various versions of biblatex, trees with fonts, and I have my own packages where the development code resides in their own tree - and I'm constantly registering and unregistering this roots.
 
yo'
@JosephWright yep, that's it.
 
7:24 PM
I just found big problem with pdf2svg ! with a pdf file that is ok, it generates corrupted svg images. pdf was created by lualatex or pdflatex. Same problem. This happens when I use \usepackage[OT1]{fontenc} in the pdf file
I wonder if pdf2svg questions are allowed here
I have MWE also.
 
@UlrikeFischer I've never tried it, but from the comment in texmf.cnf isn't more than one tree allowed for TEXMFHOME:
% Per-user texmf tree(s) -- organized per the TDS, as usual.  To define
% more than one per-user tree, set this to a list of directories in
% braces, as described above.  (This used to be HOMETEXMF.)  ~ expands
% to %USERPROFILE% on Windows, $HOME otherwise.
TEXMFHOME = ~/texmf
 
@JosephWright When you run initexmf --edit-config-file updmap you will add your configuration there. This is your tree (and you should backup it), so you can use it.
 
@UlrikeFischer Yes, I see that but those are system-maintained files and the description is UserConfig not 'Local TeX tree' or 'User writeable' or ...
@UlrikeFischer I'd view it more like texmf-var
@UlrikeFischer BTW, the tool-tip says specifically 'you should not install your own files here'
@UlrikeFischer On another matter, you said you've needed to access l3keys key values in some cases. Do you have a demo of the exact requirements?
 
@JosephWright Yes but how do you add roots and remove them in an easy way? In miktex I can do initexmf --register-root=path/to/root and initexmf --unregister--root=path/to/root which can be done easily in scripts and shortcuts.
 
@UlrikeFischer I'm downloading TL'15 onto my Windows system here now for some testing (on my Mac at present, and tests on OS X might not translate)
@UlrikeFischer I did say I've never tried that (yet) ;-)
 
7:40 PM
@JosephWright No texmf-var is the UserData root. And I only said that you can use UserConfig - it is ok for local files as you need to backup anyway - but due to the easy way in miktex to add roots there is not really a need to predefine a lot of local roots.
 
@UlrikeFischer I only feel it's odd that Christian doesn't pre-define one user tree which appears without selecting the tickbox and which isn't in a hidden location; as I say I'd have it as %USERPROFILE%\texmf but other places are possible (several programs I use add similar things inside %USERPROFILE%\Documents)
 
@JosephWright I could look it up. But it is really simply that I prefer code with "do something with \l_current_key_value_tl" then "do something with #1". Also it would be fine to be able to show the key value with \tl_show:N when debugging.
 
@UlrikeFischer Ah, that's slightly different to what I meant: good job I asked :-)
@UlrikeFischer \l_keys_value_tl holds (or should hold) the value whilst processing: I was thinking of after you've processed keys, to represent the tree structure of the key data (say if you wanted to serialise it)
 
@JosephWright: Did you saw my comment here: github.com/plk/biblatex/issues/257? Imho \rangelen is really broken ...
 
@UlrikeFischer Yes, I did see: I'll fix it in a bit (really should get this review I'm doing finished)
 
7:51 PM
@JosephWright No need to hurry.
 
@UlrikeFischer Oh, there's always a reason to hurry :-)
 
@JosephWright Just in case: I do have local (lua+bat)-scripts that do the "register-root" thing also for texlive and they work on my system (but need some polishing). But in miktex it is much easier...
@JosephWright Well imho 95% don't need a local tree and from the rest a lot prefer some other location then the default so one could debate if the texsystem or the user should add the localtexmf tree. I never used the texlive default :-).
 
@UlrikeFischer Fix in hand
@UlrikeFischer Needs to be fixed in etoolbox not biblatex: port some expl3 code [the go-to for truly robust solutions ;-)]
@UlrikeFischer No? What's wrong with it?
 
#
nothing, only that I prefer a speaking name (UF...), a special location (on another partition) and that the tree is used with miktex too.
 
yo'
8:06 PM
@PauloCereda What are the little things in front of the ducks?
 
@JosephWright Really? I thought it would be enough to change the catcode of & to 4 before the definitions (everything worked fine after I repeated the definitions in the preamble where the catcode is 4 again).
 
@UlrikeFischer That might well also work but I'm trying to stick to PL's original code as far as possible: he liked catcode-3 & as a marker
@UlrikeFischer The problem really comes from \ifblank, which also uses catcode-3 & as a marker (I missed this as for testing I tend not to mess with the catcode of & and do testing in a preamble). So the bug might show up with any code that uses & in this way, and it can be fixed using a new more robust definition of \ifblank.
\documentclass{article}

\usepackage[]{biblatex}
\makeatletter
\long\def\ifblank#1{%
  \expandafter\ifx\expandafter\relax\detokenize\expandafter{\@gobble#1?}\relax
    \expandafter\@firstoftwo
  \else
    \expandafter\@secondoftwo
  \fi
}
\makeatother
\begin{document}

\rangelen{1}
\end{document}
@UlrikeFischer ^^^
@UlrikeFischer It's also easier for me to update etoolbox quickly :-)
 
@JosephWright I see. A deeper bug ;-).
 
@UlrikeFischer Yes
@UlrikeFischer Minimal demo of the issue
\documentclass{article}

\usepackage{etoolbox}

\begin{document}
\catcode`&=3 %
\ifblank{?&}{T}{F}
\end{document}
Only happens if you have &_3 and at least one more token in the argument
The other token has to be before the &: those after have no effect
So my biblatex logic was correct :-)
@UlrikeFischer I'd appreciate a double-check on my fix: if it works for you (try a few things) then I'll send to CTAN
@UlrikeFischer Just done a quick test on my Windows 7/TL system. Doing Z:\>set TEXMFHOME={Z:\Library\texmf,Z:\texmf} does search both locations in the order given. (I'm on Parallels hence the slightly odd locations: Z: maps to my Mac home directory.)
Anyone else willing to test my etoolbox update?
 
8:57 PM
Decided to go for it :-)
 
9:18 PM
Hello @DavidCarlisle
 
9:42 PM
@JosephWright was just mailing you... (sent)
 
@DavidCarlisle: Does \BAnewcolumntype support column types with an argument, as \newcolumntype, i.e. \BAnewcolumntype{Q}[1]{foo}?
 
@JosephWright @davidcarlisle added it so we may as well use it! hey I added the definition as "someone" added it undefined to the text:-)
 
@DavidCarlisle :-)
@DavidCarlisle Feel free to revert
 
@ChristianHupfer possibly. It was a long time ago, a very long time ago....
@JosephWright no:-)
 
@DavidCarlisle In a galaxy far away?
 
9:50 PM
@JosephWright otherwise known as Manchester
@ChristianHupfer "support" isn't a word I'm sure I'd apply to any of blkarray
 
@JosephWright That's the easy part :-) Now add two more roots and then remove the texmf (and only this) from the list. I can test etoolbox tomorrow if necessary.
 
@DavidCarlisle That's not a galaxy, that's a parallel universe ;-) Thanks anyway
 

« first day (1736 days earlier)      last day (3191 days later) »