@JosephWright @DavidCarlisle Can you take a look, some files of babel in the subversion repository are severely broken, e.g. english.dtx, Texts at line ends are cut off.
@HeikoOberdiek yes there was an edit across the whole repository to remove white space from ends of lines (following a comment from you I think) but in that one directory it acted in a bizarre way, so I reverted the checkin for babel only in all other directories it seemed fine.
@WillRobertson Hi, would you please take a look on this topic tex.stackexchange.com/a/175572 ? I read the manual of verse and there is nothing written on the numbering over multiple stanzas. In the only example (on page 13) the numbering on the last line is missing. Or is this best practise for verse numbering? Don't know. Maybe you can clarify that a bit. Thank you very much for your work.
@WillRobertson Hope, you don't mind asking you this way here. Couldn't find an issue tracker or alike.
@DavidCarlisle Yes it was my suggestion )and I have seen the removal of white space in another directory. I wanted to see, if there are some last-minute changes. (There were, e.g. ltnews21.tex, thus I made a new release of latex-tds. Happily babel is no longer part of latex-tds.)
I would like to put in my resume
N years experience in Java
where N would be generated based on the date provided in the LaTeX file.
Preferably in rounded years, i.e. If the beginning date was 2012-06-04 and the output was generated on 2014-04-04, then the N would be 2.
@JosephWright Sorry for asking here. Could you link me to a post on this side or the page in your manual, explaining the different spacing in this MWE: `\documentclass{article} \usepackage[locale=DE]{siunitx} \begin{document} $0,7 = \num{0,7}$ \end{document}`
should I always use \num for any decimal in my formulae?
@HeikoOberdiek @DavidCarlisle Hopefully have now got the hang of this business: stray spaces should be stripped from all of the L3 stuff (at least all of the .dtx files)
@LaRiFaRi If you want the 'auto-formatting'
@LaRiFaRi Note that as it stands your spacing will be wrong in the simple 0,7 case
@LaRiFaRi \num is relatively slow: I'm hoping to improve it but it needs some time!
but why is the spacing wrong for the simple case? Every "normal" user (strange to call people, not knowing siunitx, normal...) would just type 0,7 in an equation
are the always set wrong? Why is the so?
@JosephWright can't find 'auto-formatting' in your documentation. Which approach would you recommend? Of course, correct spacing (and as a plus, the grouping) would be nice without using \num in every formula
In German the comma rather than the period is used as a decimal separator and the dot is used as a thousands separator: 12.345,67. The comma creates some unwanted space when used as in $12.345,67$. Is there a light-weight package or some macro to support this? (I'm not looking for automatic forma...
@JosephWright and an other topic, as you seem to like xparse (or write it?): Couldn't solve this issue with normal macros or with xparse tex.stackexchange.com/q/175577 But not my problem. Just in case you wanna try.
@DavidCarlisle Really getting the hang of this 'sweep over the whole repo' thing :-) Nex step: install GOW for my Parallels Windows installation on my Mac to avoid line-ending disasters!
@JosephWright so that means you have a unix-based oprating sustem running windows in a virtual machine over that and then install a set of unix-emulating tools in the virtual machine? makes sense to me:-)
@JosephWright sure it sounds almost logical:-) meanwhile I've just been "upgraded" to an office365 mail server so if I you send me mail it'll probably be stuck in Seattle somewhere.
@DavidCarlisle Mainly for the LaTeX2e stuff that I need the Gnu tools on my Windows VM: the build scripts are now Win-only, so I have to checkout the files using SVN on Windows and that means CRLF line-ends. For L3 work, I can get away with most things!
@JosephWright we could set the eol property in svn so unix line endings were used whatever platform.
@JosephWright which is why I have to do a balancing act to get it to work under cygwin. for latex2e-public not so bad as there are only 4 or 5 bat files, but for l3 there are dozens of the things, and I don't have a working setup yet....
@DavidCarlisle L3 assume something Unix-like, but hopefully not too much beyond that (we had some bash-only stuff I'd picked up by copy-pasting, but that got removed)
@DavidCarlisle Anyway, I'm going to do some revisions on them to try to bring more into line with the L2e ones Frank wrote plus to have more similar behaviour on Windows and Unix
@DavidCarlisle Cool
@DavidCarlisle Start with something simple :-) (l3kernelmake unpack, perhaps)
:15362717 $ make unpack
echo "Unpacking files"
Unpacking files
cd /tmp/l3test/test/l3kernel ; \
for I in l3.ins l3doc.dtx ; do \
tex $I > /dev/null ; \
done
@JosephWright so trying top level make check: what could go wrong:-)
@DavidCarlisle Like I say, I'm planning to do some revisions of all of these, hopefully this week: let me know if there is anything I need to consider for Cygwin
@JosephWright although longer term it would probably be good to just have one set of build scripts rather than you having to maintain make and bat in parallel
@JosephWright I was wondering if there is a simple enough implementation of a bat interpreted for linux or mac (ie without invoking all of wine) but basically the bat files only need some tests and some loops, it ought to be able to run them in linux...
@DavidCarlisle The aim was to avoid requiring more stuff than absolutely required (otherwise I guess writing Makefiles that work on Windows would be an option): like I say, Lua is perhaps the best plan as it allows a move to 'just needs a modern TeX system', which we assume anyway.
I know the TeX Live scripts have all gone that way
@JosephWright yes lua sounds sensible but since we assume perl anyway it would be more fun to do some perl to convert and execute the bat files on the fly ie automating this tldp.org/LDP/abs/html/dosbatch.html
@JosephWright that does seem to be the sensible route
@JosephWright speaking of lua do you understand the luabase packaging stuff? I removed it all in my inputenc trials to avoid having half of contrib as a dependency, but I may have removed too much and if you load fontenc later it gives me a wrap over the knuckles for not using luatexbase interface and using the primitive calbacks directly.
@HeikoOberdiek It was clear enough what it's doing, but if we were to make inputenc work better on luatex out of the box it would be good to pull in some lua without the following. It's easy to do that but then you don't get the benefit I assume the package loader gives of loading multiple cooperative packages, so some middle way I think, but luainputenc currently does:
@topskip So how would you see this in relation to going beyond LaTeX2e: are we wasting our time entirely? Should we instead be defining some input format to be translated to Lua (much as Jonathan Fine would like but with s/Python/Lua/g)?
@JosephWright No, I don't think so. There is a huge area for "classic LaTeX documents" (say homeworks, books, and more advanced things). I believe that should stay the same, while admitting it's not always the perfect way to go
@JosephWright yes, I was a bit in a hurry doing the slides, of course I had them prepared right at the conference during the (polish) talks
@JosephWright My customers would never want to touch (La)TeX code, that's why I hide it completly. But this also means, it's limited to the functionality that I provide, and I know many people who would not accept that limits.
Apostolos Syropoulos sent me a new version of the Asana Math font but now, I can't test it, because LuaLaTeX is still using some old stuff. Don't know, where that come from. I guess, Luaodfloat (is that written correct...) has built its database and I need to refresh this.
found it. Just replaced the fonts in D:\MiKTeX\fonts\opentype\public\ Don't know, if this is the good way, but thus, I do not need a refresh of the font database.
@GonzaloMedina the site is very slow today the "notification" box menu just shows white for a while if I click on it, even though it went red to say I had a mesage, but it picked up eventually
Hello guys. Anyone know metafont (read: know metafont better than a total of 45 minutes)? I'm trying to find out if there's a smarter way to write the path in this answer:
Disclaimer: I have never used Metafont before, and I also have only ever touched feynmf once. Please comment if I did any of this completely wrong.
Since you never even provided a complete MWE, I just made this stuff up.
I used the \fmfcmd macro to define a path, which I named otimes, to dra...
It seems really convoluted but I haven't been able to figure out a nicer way.
I want to create an animation about roller coaster. For a simple track, for example, a circle, I can determine the position of the center of its wheel easily.
However, for any parametric curve, I have no idea to determine the coordinates. Let's take a general case as follows.
Given the para...
Hi, @PauloCereda. Is there any hope for compiling with knitr via arara? I tried writing a rule for this, but I can't seem to get things working, and I'm not sure what I'm doing wrong.
@percusse Great! I'm looking for further plot examples. Such as from real life, like from a thesis. And your great solutions here, with some flesh around.
@HenriMenke I have looked at the manual, but I haven't been able to figure out what I'm doing wrong. There are more details in the question that I linked to.
@AdamLiter Oh my, I missed that completely, sorry. One of the things I can already point out is that I did a terrible mistake in the manual. :P Try ^(\s)*%\s+ instead of ^(\\s)*%\\s+. I suspect we will get another error, but at least we move on. :)
You know you get a weird serendipidipididpdiipdipdipdity when you realize that all this emacs / vim thingy is irrelevant if you are not coding like crazy :) I can manually fix 6 mistakes in the code by Find/Replace while the dude is fine tuning his emacs.
@PauloCereda Thanks! It at least finds the directive now when arara myfile is called, though I still get the same error if arara myfile.Rnw is called. The new error is A problem was detected in the 'knitr' task. The 'knitr.yaml' file located at '/Users/adamliter/Dropbox/Local-texmf/arara/rules' requires a valid list of arguments. Could you fix it? Note that the list of arguments can also be empty, but it is still necessary to define it explicitly (with 'arguments: []').
(Incidentally, does the full path have to be specified? arara couldn't find the rule when the path was ~/Dropbox/Local-texmf/arara/rules, so I had to give it the full path.)
@PauloCereda Huh, interesting. So it runs and says SUCCESS, but there is no .tex file that gets written to the directory, which is what running knitr should do...
I've always been vaguely aware of 'raper' as an alternative to 'rapist', as a vaguely wrong sounding, possibly archaic formulation - nowadays, it's most often heard from speakers of English as a second language, for whom the construction is slightly irregular compared to other similar verbs. But...
@percusse “Rapist” always recalls this quotation from the TeXbook: But computers are notoriously bad at hyphenation. When the typesetting of newspapers began to be fully automated, jokes about “the-rapists who pre-ached on wee-knights“ soon began to circulate.
By the way I didn't know that ELL is also that notorious in terms of competition.
oh this one is ELU
i meant ELU ... meh
when two reviewers rant about a paper (one of them is yours truly), this always happens; a third reviewer pops up and writes stuff like this :
This paper is well organized. The presentation of this paper is clear. The related research methods are clearly presented and explained. The mathematic derivations and proofs are rigorous. Previous work in this paper has been reviewed and the references are adequate. This is a solid and publishable manuscript.