@daleif well no rush really except I wanted to warn you in case some early testers contact you, and the whole point of the dropbox release is to avoid pushing latex to ctan yet, but if you could get an update on to ctan before that (we will do it in a month or so, probably) so that it is in the TL2015 pretests that would be good! thanks,
@DavidCarlisle, btw will those changes work even on systems without the latex update? There are users out there who manually update packages without updating the core system.
@daleif yes the \textsubscript one I just wrapped your existing code in \@ifundefined{textsubscript} so in an existing format it just does what it always did, and in a new one it does nothing. The test to doublefloats I didn't suggest a change but the test is probably testing the wrong thing on new formats (but the change looks very suspicious anyway, i couldn't see why memoir was patching that:-)
@JosephWright no @daleif meant (I think) if he changed the \textsubscript code in memoir would it work if someone updated memoir without updating latex. but answer is still yes.
@daleif as I mentioned to Ulrike on latex-l an easy way to test memoir for the new format without actually building the new format is to get latexrelease.sty and make:
@JosephWright BTW: is \tl_mixed_case:n f- or x-type expandable? interface3 marks it with a hollow star but the description says it's x-type expandable.
@clemens I've altered the code so it's f-type expandable. I've made some more adjustments over the last few days: should all be correct on the next CTAN update.
@1010011010 I am, in passing, reminded of the \FLUSHacrostic approach in my answer at tex.stackexchange.com/questions/121440/…. It certainly is not a solution to your problem... but the notion of artificially adding space to force flushness gets me thinking about your issue.
@nuttyaboutnatty yes. Of course if you don't want to do that you could just do \date{\small 1st march 2015\endgraf} and it would work, but it would be wrong\date like \title or \section should just take text not formatting instructions.
@nuttyaboutnatty yes it's safe enough in article, of course the problem with such things is that adding font size into the string breaks if you use a class that does more eg use the same string in two places, eg on a titlepage or in a footnote or end of the document, if \@date is plain text such a class can set the same text in different ways in different places, but if it forces \small that doesn't work.
@JosephWright working back from the above I reach a MWE of
\documentclass{article}
\usepackage{expl3}
\stop
LaTeX2e <2015/01/01>
Babel <3.9l> and hyphenation patterns for 79 languages loaded.
(/home/l3svn/latex2e-public/build/local/article.cls
Document Class: article 2014/09/29 v1.4h Standard LaTeX document class
(/home/l3svn/latex2e-public/build/local/size10.clo))
(/home/davidc/texmf/tex/latex/l3kernel/expl3.sty
(/home/davidc/texmf/tex/latex/l3kernel/expl3-code.tex (./etex.sty)
! LaTeX Error: Missing \begin{document}.
See the LaTeX manual or LaTeX Companion for explanation.
Type H <return> for immediate help.
there is a clue in the path shown to etex.sty ... some time back i was experimenting with an etex.sty that detected 2015 and skipped the allocation code, forgot it was still there. But it leaves \reserveinserts undefined so expl3 `\RequirePackage{etex}% \csname reserveinserts\endcsname{32}%` tries to typeset 32..
@JosephWright could do something eg define it to be \@gobble, but I think it means keeping to the plan of saying if you load etex you get the old etex package code is safest.
@JosephWright so I think all the uses of \textsubscript I saw except memoirs use \providecommand rather than \newcommand to define it so are fine.
@DavidCarlisle Yes: I might raise this on LaTeX-L for exactly this reason. We could argue that we do have a position: \jobname should quote any spaces.
Really, my end goal is figuring out how latex renders integral symbols from a font
I've been told LM Math is a good approximation of the default math font in latex, but typing out an integral sign in libre office looks totally different to in latex
@Ell because LibreOffice Math sucks soooo much (sorry for stepping in your discussion, I had to say this). Needed to say, it's heavens compared to M$ W@#$
@ell seems like Lm Math has a much more pronounced slope to its display math integral than its inline math one. (A slightly odd choice but anyway) the above shows your image alongside the output from the following. It looks like you have the LM inline glyph.
@Ell sorry that's what the ?? meant, it meant it reported an unnamed unicode slot when I looked up the characters as reported by \showoutput in the log file.
@Ell getting to the edge of my opentype knowledge but it may be that you have to access it via the opentype MATH table properties, just as in classic tex you just access a normal integral and the tfm file metrics lead you to a larger one if needed similar to the way ligatures work
@Ell but anyway if you add \showoutput to the above test file, the log says: .....\l_fontspec_font so tex says it uses that character whatever it does in reality:-)
@Ell I know it's an issue in mathjax that it can't use the opentype math fonts such as stix as most of the extension glyphs are only available via opentype properties which the browser can see but are not exposed to javascript, so it has to re-package the fonts with all the gyphs encoded simply in the private use area
> Well, the engine already reads OpenType/CFF files perfectly. What it doesn't do is handling ‘OpenType Layout’ tables.
FreeType 1 comes with a set of extensions that are used to load and manage OpenType Layout tables. It even has a demonstration program named ftstrtto to show its capabilities. However, this code is no longer maintained, and we strongly advise to not use it.
For FreeType 2, we have decided that the layout operations provided through these tables are better placed in a specific text-layout library like Pango.
@yo' Norwich only in the week (at work): at the weekend I'm at home near a place called Wellingborough. It's only about 50 minutes from here on the train to St Pancras
I have yet to decide whether we'll go Sunday afternoon, or Sunday overnight. One possibility is to take off at 1am in the morning and be at 7-8am in Darmstadt, enjoying jam-less night highways.
@DavidCarlisle I think the price will not be too much higher than flying as whilst I might get a cheap-ish flight the 'get to Heathrow' part is expensive
@DavidCarlisle I checked from Standsted but this seems to not be an option
@JosephWright getting to heathrow is always a problem for me (I often used to get queries at work when I travelled more) from oxford there is a bus every few minutes and the cost is basically nothing, but for me to get from home to oxford by public transport means a 25mile taxi which costs more than flying to FRA so usually I drive and dump the car at the airport but it's still a major part of the cost.
The Jabref bibtex editor shows a preview of the article including the abstract and personal notes metadata. It looks a bit like this:
My question is: is there a way to show nice-looking formulas in these previews (preferably using Latex syntax) or are these fields strictly text-only?
@JosephWright nonono, I'm a bash person, and I don't mind terminal commands. It's more the philosophy: commits, pulls, pushes, adds all doing the same thing: putting some file somewhere.
@yo' the main problem I have with git and mercurial though is for many projects they really are not distributed in that sense so you have to do twice as many operations as you would in svn, first commit to a local repository then push those changes to a central repository
@yo' I've used git for a couple of projects but I've been using svn/cvs since they existed and rcs before that, so I'm not saying it's wrong, just that it takes some getting used to, unlike switching from cvs to svn where the main thing to remember is to type svn instead of cvs.
@yo' is it on github, or are you hosting it directly? (either way can't you just pretend to be someone else and see if you can clone the repo and do whatever)
@ChristianHupfer Look at golatex.de/minitoc-t14777.html Im Kindergarten ist es nicht so schlimm, da gibts eins mit der Schippe drauf. Und hier, Erwachsene Leute spielen die beleidigte Leberwurst.
This is fixed in the next release of LaTeX (the above was generated with the pre-release code), but meanwhile you can use the code in the reply to the latex bug report at
http://www.latex-project.org/cgi-bin/ltxbugs2html?pr=amslatex/4393
@JosephWright: Do you think it is possible to convince the StackExchange team to hand over my TUG membership I 'won' recently to someone other who ranked lower in that competition? I think, it was T. Sturm who decided that Sean Allred should 'win' the draw?
@ChristianHupfer I guess it really comes down to talking to the TUG office in the first instance as I don't know what the rules are about 'rep transfers'
@ChristianHupfer I'm considering doing a second master's in astrophysics, but if I want to do it "the cheap way" (locally at my university here), I'd be stuck with 40 ECTS worth of quantum field theory which seems like a bit of a deal breaker to me... How do you look back on that decision? Did you have to plow through all of that as well? :-(
@ChristianHupfer Basically the master's here is just QFT and nothing else... Did you have that too or did you do other stuff as well? What can you recommend? The rest is all optional...
@1010011010: You don't need QFT do understand astrophysics, as long it is not concerned with Cosmology. QFT is quite difficult, but interesting. I think, the eduction system of your university is not comparable with 'my' university. And I dropped out of the system 10 years ago
@1010011010: Theoretical astrophysics concerning stars, their evolution, magnetohydrodynamics and a lot of computational fluid dynamics and numerical simulations
@1010011010: At that time German university education was much better, in my opinion. No need to do mandatory courses. Either you had acquired the knowledge or you got into a pretty pickle being in an examination. Nobody told you which courses to get, apart from some fundamental one, as quantum mechanics or particle physics
@1010011010: I recommend General theory of relativity, however
@ChristianHupfer Doing such an enormous subject as pre-graduate means you will skip on other parts. Here you can take introductory courses on e.g. condensed/soft matter, cosmology, CFD, astrophysics, etc. as undergraduates...
Ah that's one thing that I think undergraduates should always have access to but my university refuses to change the programme. :-(
@ChristianHupfer But you do like atomic physics? In my experience the undergraduate courses were almost identical... except atomic physics yielded no points at all. :(
@ChristianHupfer What you say reminds me of a friend of mine who finished his master's last year. He was so glad that weekend finally really meant "weekend". As in, no deadlines, nothing.
@1010011010: That's not what bothered me... too much people on the campus, most of them thinking they were important. I liked the silence of my home, I studied most time there ;-) But of course, I could not bypass any visit to the campus
@ChristianHupfer “upvotes to silly questions” why does that bother you?
The only thing in your list which bothers me from time to time is questions getting closed much to soon... the rest I simply don't care. (I've only had one bad edit to my answers IIRC so that doesn't count)
This is svn but any command line is the same.
Depending on your configuration you probably need
pdflatex --enable-write18 filename
to enable \write18 to be a hook to execute shell commands.
this writes to a temporary file using the shell escape, then inputs it
\documentclass{article}
\be...
@yo' not sure what you mean, it has the version number of the last checkin so doesn't record any local edits after that. Actually i can't recall exactly what I suggested i just recalled the question:-)
@mikeonly off the top of my head probably not, but after all these decades the answer can't be that there are only two:-) You can always make new bibtex styles anyway with makebst or just editing an existing one.
The Jabref bibtex editor shows a preview of the article including the abstract and personal notes metadata. It looks a bit like this:
My question is: is there a way to show nice-looking formulas in these previews (preferably using Latex syntax) or are these fields strictly text-only?
This is fixed in the next release of LaTeX (the above was generated with the pre-release code), but meanwhile you can use the code in the reply to the latex bug report at
http://www.latex-project.org/cgi-bin/ltxbugs2html?pr=amslatex/4393
