@Speravir Do you have a MiKTeX installation only for the user? Do you know were the MiKTeX system is installed in this case? If you install it as admin (as I did) it is installed in c:\Programme/MiKTeX 2.9\ With the user installation I suppose something like Eigene Dateien\MiKTeX 2.9\?
@Kurt Yes, I have a (one-) user installation, but I chose an individual folder outside of “C:\Programme” (for international users: German version for “C:\Program Files”). And … if I remember correctly, I still could have installed in in program files folder. We should ask @UlrikeFischer (and would need perhaps one of our moderators for noticing her) to be sure, but I think the only difference is, that for a user install some MiKTeX variables are simply not used. (continued)
@Kurt (Cont.) Te output of initexmf --report on my system is:
MiKTeX: 2.9
Invokers: explorer/cmd
SystemAdmin: yes
PowerUser: no
OS: Microsoft Windows 7 Home Premium Edition, 64-bit Service Pack 1 (build 7601)
Root 0: C:\Users\Steffen\AppData\Roaming\MiKTeX\2.9
Root 1: C:\Users\Steffen\AppData\Local\MiKTeX\2.9
Root 2: C:\LaTeX\LocalTeXMF
Root 3: C:\LaTeX\MiKTeX
Root 4: C:\ProgramData\MiKTeX\2.9
Root 5: C:\Program Files (x86)\MiKTeX 2.9
UserInstall: C:\LaTeX\MiKTeX
UserData: C:\Users\Steffen\AppData\Local\MiKTeX\2.9
UserConfig: C:\Users\Steffen\AppData\Roaming\MiKTeX\2.9
@Kurt But the last three variables Common* are not used here!
@Kurt In other words: The paths are not existant on my system!
From Karl:
FYI ...
TL'12 is frozen forever.
I updated the various web pages, etc., for the upcoming TL13 pretest,
but it is not nearly ready yet; indeed, the current svn Master is
completely broken. Please don't bother reporting it, anything about it :).
I'm working on it, really.
I hope to have an initial pretest build by tomorrow night, but we'll see.
Sorry, was busy. My normal (not admin) user returns with `initexmf --report`: `C:\Users\Internet>initexmf --report MiKTeX: 2.9 Invokers: non-existing/explorer/cmd SystemAdmin: no PowerUser: no OS: Microsoft Windows 7 Home Premium Edition, 64-bit Service Pack 1 (build 7601)
@PauloCereda It looks nice, thanks. Personally I'd prefer a lion in an earlier dynasty, if we use an ancient lion. BTW, actually I don't know any kungfu either.
@Speravir Ah, so you can choose were to install a user-based MiKTeX? Interesting. Depending on problems with the rights for writing files I supposed, that the user based installation is not able to use c:\Program Files, because the normal user has no rights to write there ...
@Kurt As I said above I didn’t try. By default at least the first configured user has “almost” admin rights (in the output marked as “SystemAdmin: yes”), but I guess the infamous User Account Control (UAC, German “Benutzerkontensteuerung”) comes in between. And yes during the installation you can select where to install, maybe there is something like “Extended Installation Settings” or so, I don’tremember exactly.
@Speravir No problem, I just want to understand the difference between the both kinds of installing. I see the advantage if more than one user can use MiKTeX on a computer (for example in the computer pool of a computing center), but I do not understand the reason for the problems with updating as admin and as user ...
In the manual is no information about that and I do not know another place to find this informations. (Schade -- In English?)
@Kurt I didn’t want to critic you. :-) Updating by the admin would be enough. The user(s) need only a refresh of their own FNDB and format+map files, and the latter only, when the had done this once before. This could happen because of automatic installation on the fly or manual adding packages into an own local tree – or simply execution of these refreshing steps as user (clicking on buttons or command line). This is BTW only “Hörensagen” from Ulrike and perhaps in parts wrong.
@GonzaloMedina Add the related question just now. @StefanKottwitz @MartinScharrer @JosephWright If possible remove this revision: tex.stackexchange.com/revisions/107473/3 (Gonzalo, I hope, I am right.)
@JosephWright: I have not read the siunitx manual thoroughly. Is there an option to take the first n digit after decimal without rounding? I mean \SI[<options>]{2.56}{\meter} becomes 2.5 m and \SI[<options>]{2.00}{\meter} becomes 2.0 m.
@PauloCereda It only seems like I am: the commits are only corrections of old typos in the manuals and introducing new ones so I'll have something to do for the next one ;)
Sorry for interrupting. I got an error when setting precision=1 as follows.
\documentclass[tikz,border=12pt]{standalone}
\usepackage{siunitx}
\begin{document}
\begin{tikzpicture}
\foreach \n in {0.0,-0.5,...,-2.0}{%
\draw (0,\n) node[anchor=west] {%
% errors occur when precision=1 but not for other values.
\pgfmathparse{\n}\pgfmathprintnumberto[precision=1]{\pgfmathresult}{\temp}$\SI{\temp}{\second}$%
};
}
\end{tikzpicture}
\end{document}
\SI cannot parse num \cdot 10^num produced by pgfparser.
@DavidCarlisle If you want to rung dandelion natively on Windows, it seems until i sort a script for /f "delims=" %I in ('kpsewhich dandelion.lua') do texlua %I test example1.tex is the best plan
@DavidCarlisle All of our Windows scripts work natively: people really do get the wrong idea :-) (I have gow for things like the GNU find utility, as gow is much ligher weight than Cygwin)
@JosephWright what's strange? it was a new test (that I wrote) that replaced 1 by ^^@ in the \TYPE but I couldn't work out what to put in the expects block
@JosephWright I was actually planning to test how dandelion reacted to control chars being written as ^^ syntax but picked on ^^@ out of habit and of course that one is rather special:-)
@DavidCarlisle Did you try texlua `kpswehich dandelion.lua` save example1.tex?
(I'm always using kpsewhich as texlua doesn't seem to do a search, so doesn't find the script)
I get
%%% ! Text line contains an invalid character.
%%% l... \TYPE{^^@
%%% 234}
%%% A funny symbol that I can't read has just been input.
%%% Continue, and I'll forget that it ever happened.
%%%
%%% 234
@JosephWright conversely I haven't installed it locally just running in the gh checkout so kpsewhich wouldn't find it for me but I just cd to the examples dir and use ../dandelion.lua
@JosephWright It's sort of convenient but I'm not sure this is the best forum for dandelion discussion (a) it's almost TeX related so off topic, and (b) we are probably boring people...
Psmith, the TeX bot, in fixed font mode: Let's take a look at the last fencing results:
- CADET 7 JUNIOR WORLD CHAMPS REPORT - DAY 2
- FENCING WORLD CLASS PROGRAMME 2013-2017
- BOARD ELECTIONS 2013 RESULTS
- LEVEL 1 AND 2 FOIL, EPEE, SABRE - ESSEX, EASTERN
- SAFEGUARDING AND PROTECTING CHILDREN COURSE - NEWBURY, SOUTHERN
- INTERNATIONAL RESULTS ROUND UP
- GB TO HOST FIE SATELLITE EVENT
- GOOD LUCK TO OUR FENCERS THIS WEEKEND
- INTERNATIONAL RESULTS ROUND UP
- SPORT & RECREATION ALLIANCE CLUB SURVEY
@JosephWright Okay, it turns out the problem has nothing to do with the expl3 part. It's because they're using ` & ` as a separator but have just done \DTLsetseparator{&}. The weird expl3 stuff threw me. Thanks.
@PauloCereda I've installed a text editor and configured the email. I can't work out how to sync the calendar with my google calender (not sure if it's possible).
I need to see if I can connect my tablet to a data projector. It'll be a lot easier to take a tablet than a laptop with me when I have to do presentations.
@Speravir The problem is, I checked it some days ago, that the updating manager of user admin shows other packages to be updated than the updating manager called with normal user. So there seems to be a difference. Exactly as you said, I thought that updating with admin should be enough. That the update of formats and FNDB is needed for user and admin I understand.
@PauloCereda Oh no, it would create some kind of apocalypse with vim and emacs fighting for supremacy, but ultimately merging into some kind of hybrid monster unleashed on the world in a mad rampage!
@topskip I had a remote mouse for my laptop which was quite useful, but it was so big and chunky, there's not much difference between holding that and my tablet :-)
@PauloCereda LOL. I wouldn't be as good as the duck story :-)
@DavidCarlisle I kept asking her to make the pictures wider so they were the correct aspect ratio :-) Each picture spans a double-page spread with the text above. (I've not tried that before, so I hope it turns out okay. It should do as it's a saddle-stitch binding.)
@topskip I'll have to see if I can find one for the Arnova.
@JosephWright "I'm always using kpsewhich as texlua doesn't seem to do a search" - I recall reading something Hans Hagen wrote about why texexec sort-of replicated Kpathsea but would actually use the codebase - one of the stranger decisions in the Context world. And I guess texlua follows Hans's choice.
I think I've found a "showstopper" bug in the newest `babel` version 3.9b: A line break is incorrectly added before `\foreignlanguage`. Can someone test the following example and confirm the wrong behaviour?
\documentclass{article}
\usepackage[ngerman,english]{babel}
\begin{document}
% babel v3.9b incorrectly adds a line break before \foreignlanguage Some text in a foreign language: \foreignlanguage{ngerman}{Man soll den Tag nicht vor dem Abend loben.}
What is the state of the art in automatic hyphenation of compound words? In German, a hyphen between the word components should have a much higher priority than a hyphen within one of the word components. Any new development on that? In the new engines perhaps?
I could imagine having hyphenation in three passes (instead of the current two): once, without hyphenation at all (rather useless for German), once, with a special hyphenation pattern with hyphens only between word components, and finally, the normal hyphenation pattern.
@StephanLehmke I am sure I have seen this implemented around the time tex3 came out with two sets of hyphenation patterns for German. The trouble was it didn't really work well as if you got to final pass you couldn't give higher priority to the compound breaks (and if you could do that you wouldn't need two passes). I suppose the solution of you know putting spaces between words isn't an option?
@Karl'sstudents Because \pgfmathprintnumbertomacro uses math mode internally. It's not a good idea to mix the number parsers of PGF and siunitx. In this case, I would just stick with the siunitx one and let it to the rounding:
@StephanLehmke yes but are there any sufficiently simple rules for splitting compounds that you could avoid a full hyphenation trial and just apply a regex to add compound word marks. (Showing complete ignorance of German here...)
@Karl'sstudents Typo in my answer: I meant \pgfmathparse{\n}\SI[round-mode=places, round-precision=1, round-integer-to-decimal]{\pgfmathresult}{\second}
@DavidCarlisle Yes this would probably be feasible as a preprocessing step. What exactly is the effect of the compund word mark on hyphenation? I thought it only breaks ligatures?
@StephanLehmke If you use the T1 encoding mark then (in principle) you can make it a letter for hyphenation and have hyphenation patterns that use that letter in the patterns.
@DavidCarlisle Ah, OK. So I need new patterns on top of inserting the characters. I can see how this will yield more correct hyphenation of compound words. But how does it prioritize hyphenation at word boundaries?
@lockstep Sorry for the delay - I can confirm the bug with a freshly docstripped version of babel from CTAN (v3.9b), pdftex's dvi path on a lightly modded TL 2010.
@NicolaTalbot It's a fantastic plugin, written by Tim Pope. You can have all your plugins inside .vim, very well organized, with no need of recreating the vim structure.
A novice was trying to fix a broken Lisp machine by turning the power off and on.
Knight, seeing what the student was doing, spoke sternly: “You cannot fix a machine by just power-cycling it with no understanding of what is going wrong.”
Knight turned the machine off and on.
The machine worked.
Emacs understanding is a bit similar
@lockstep You'll just have to apply the cool features of v3.9c...
@lockstep I don't know, but I unpacked the dtx for babel base, english and german, then did pdftex -ini -etex -output-format=pdf -jobname=pdflatex39 latex.ltx and with pdftex -fmt pdflatex39 bug the result was as shown (I added \raggedright to be sure about spaces)
This SO question here asks for an emacs shortcut for creating a latex block, and one solution (proposed by me, but just coming from the emacs documentation) is to use C-c C-t.
Is there something similar that would create an 'inline' block without creating any new lines, like for turning
blah b...
@lockstep I put them in ~/Library/texmf/tex/latex/base and ~/Library/texmf/tex/latex/babel-contrib; this has the good effect that when I use the standard pdflatex they are not found, because the "generic" subtree comes first, while with the newly created format they are used because "generic" is not looked for.
@DavidCarlisle Riiiiiight; the author(s) of colortbl and longtable should communicate better. I wonder why there is such a disconnect...?
@DavidCarlisle: It must be difficult being a package author and being associated with TeX.SE. Everyone always complains here about something not working to their liking, rather than asking for your bank account to deposit some well-deserved funds...
...I guess @PauloCereda is partly to blame for being to sole provider of funds around here.
@Werner longtable was written with the right hemisphere, while colortbl that's more artistically inclined with the left hemisphere. (Or the other way around.)
@Werner haven't time to look yet but it used to work, the examples in the colortbl documentation are set with longtable. My plan is that @PauloCereda offers another real money bounty so that @egreg fixes it while I drive home.
@egreg Please add \babelhyphenation{whatever} to your test file. Is this macro (new in 3.9) recognized? (I followed the steps described by you and didn't get the offending paragraph -- however, also an "undefined control sequence" error.)
I am still confused with HV's comment here. What is the relationship between \ignorespaces and \pstVerb invoked in the preamble? Is it prohibited to use \ignorespaces in a macro invoked in the preamble?
@MarcoDaniel It's a little tiresome. You need to use one of the VCS repo conversion tools to create the initial git repo, fork that to Github, and then set up scripts to import changes to the source repo, apply them to your conversion repo and export them to Github.
If you go through all that for the projects Marco is interested in, I'm sure he'll be very grateful to you :)
When I give matplotlib a annotation string such as
'$\frac{A}{B} = C$'
and I specify a fontsize of 18, the A and B are rendered at 12.6 pt, while the C is rendered at 18 pt. I want A, B, and C to all be the same size. How do I do this?
In a LaTeX document, if you give the commands
\begin{...
I can't quite see what matlab is doing less latex works than I would expect, however this sort of works to make the first expression larger:
So after a bit of trial and error and discussion in comments
most of latex is enabled but you need to switch out of math mode, use a lower level font size ...
@Werner ah I don' t think it will help though (having looked):- as that one was about publish (as latex) whereas this one is about latex-like text in plotting (which I think doesn't use text at all)
This SO question here asks for an emacs shortcut for creating a latex block, and one solution (proposed by me, but just coming from the emacs documentation) is to use C-c C-t.
Is there something similar that would create an 'inline' block without creating any new lines, like for turning
blah b...
When I give matplotlib a annotation string such as
'$\frac{A}{B} = C$'
and I specify a fontsize of 18, the A and B are rendered at 12.6 pt, while the C is rendered at 18 pt. I want A, B, and C to all be the same size. How do I do this?
In a LaTeX document, if you give the commands
\begin{...
