TUG 2015, this years meeting of the TeX Users Group will be in Darmstadt, Germany, from July 20 to 22.
Who would like to go there? Does anybody plan to make a presentation?
April 10 - bursary application deadline.
May 1 - deadline for abstracts for presentation proposals.
May 15 - dea...
@PaulGessler I suppose all. It's not a DANTE meeting. Some may be in French, since Darmstadt is surely considered a French city by some French people, and they never speak English at a conference at home.
@PauloCereda I ask because: I consider driving overnight from prague. Starting at 1am and arriving in Darmstadt on Monday 7am. Why? You get empty highways and save one night in a hotel.
A few minutes ago, when I viewed profile pages on my phone, the tabs were all in Portuguese:
On the mobile interface:
On the full site
But it didn't happen on my desktop, and now it's back to normal. Other users report that the site is normal, but Seth says it's still happening, so I've ...
@TorbjørnT. Now it became normal. I noticed it when I visited karlkoeller's profile page. So I doubt it is Italian and may be the site is very (over) intelligent to identify the location in the profile page! Phew.
But it doesn't happen always! I am lost, I think :)
@JosephWright test failed, looked like I missed one case in fltrace which was "error free" but made it skip two blocks at a time (there are unmatched brackets in messages which meant finding the matching braces was a bit painful:-)
@JosephWright the tests now pass, but running full suite again....
Hi there. Can anyone explain to me what does lax-permissive mean?
I am going crazy with the license descriptions. I want to use my first choice WTFPL but FSF has a warning about it that I also don't understand gnu.org/philosophy/license-list.html#WTFPL
basically lax permissive licenses don't prevent someone from taking the code and making a proprietary release of it (whether direct re-use or with improvements)
I was wondering. Can one locally disable line breaking after \mathbin elements? Eg I have a macro that typeset its two arguments as,say [#1,\allowbreak #2], #1 and #2 are either simple symbols or in the form of A+B, A-B. I'd like this macro to not allow line breaks after + or - (or similar). Ideas?
@DavidCarlisle Nope. I actually would like to promote such usage as long as it won't have repercussions when they attempt to patent it which I think FSF is mentioning that .
oops, hmm, I'd like to still have that feature. I'll add {..} for now, this is just the first pass. (got some nice Emacs macros out of this rewrite though)
@DavidCarlisle, nice. BTW: is there a keystrok to address a person instead of cliking on the name that pops up?
@daleif yes but different from replying. That is shown as an @ but the actual markup is a : followed by the id of the comment and it makes a back arrow to the referenced comment (and the referenced comment highlights if you hover over the reply)
@daleif the previous message was a reply, this is a new one starting with a ping to you, note the lack of an arrow at the left
@daleif That's why you need Psmith's reports from the world cup (currently taking place in Australia and New Zealand:-) @PauloCereda where is Psmith when we need him?
@egreg yes I know (a voice at back of my head was telling me that as I wrote it, but I was distracted (day job, you know:-) so carried on anyway....) @daleif ^^^^
@daleif The sad answer is no, you can't. Either you give up to stretchable/shrinkable space or disallow breaking after a binary operation in the whole formula. If the output must be in brackets, I'd go with the former strategy.
\tracingall
$a+b x+\nobreak y$
\bye
----------
...\mathon
...\teni a
...\glue(\medmuskip) 2.22217 plus 1.11108 minus 2.22217
...\tenrm +
...\penalty 700
...\glue(\medmuskip) 2.22217 plus 1.11108 minus 2.22217
...\teni b
...\teni x
...\glue(\medmuskip) 2.22217 plus 1.11108 minus 2.22217
...\tenrm +
...\penalty 10000
...\glue(\medmuskip) 2.22217 plus 1.11108 minus 2.22217
...\teni y
...\kern0.35878
...\mathoff
@egreg although perhaps modifying the inner token list is cheating?
@egreg it is as you might have guessed intevals (using only [] syntax). Lickily it seems the original author have made everthing using two arg macros, så it is easy to convert into using say the interval package. I'll add it on the next pass
@DavidCarlisle A good addition would be that each mathchar carries its own penalty, respecting groups. The penalty might be automatically assigned from the type, unless overridden.
@ Just need to write the mlist_to_hlist callback. As for Frank's questions the other day about paragraph line breaking, it would be good if luatex exposed a lua wrapper for the existing function so you could modify it rather than rewrite from scratch.
@egreg ah it does supply that: node.mlist_to_hlist so you could have a callback that did the standard thing then went through and changed some penalties....
@DavidCarlisle I have found the way for my problem, and I confirme playing with \count\footins=0 \count\footins=1000 solve the problem of even pages too short
but in my case it doesn't matter, because as it is for // typesetting, the right page is many time not full
I am looking for a way to count the number of words in a LaTeX document. I know of several word counters. What I'm looking for though is which ones are precise and would give me a breakdown to main text, captions, footnotes and exclude any formulae (or better yet: count them).
@JosephWright Yes, that's true. I'm trying to fit into tight limits ... I wish they'd just suggest a word counter they use, that would make it simpler.
The good answer is probably that word count is a stupid metric. Specify a LaTeX template instead and ask to fit in a certain number of pages using that template.
@Szabolcs a long while ago when people were discussing different tools to count words on comp.text.tex I posted this (almost all the tools fail) ctan.org/tex-archive/macros/plain/contrib/xii
@JosephWright OK. I wonder if we are done for 2e now. (Or if I should do that comma accent stuff, or if we should cycle round again and start over....)
@egreg yep, bit strange one, if you happen to have quite more whitespace on a page thanks to a large formula. Also, hardcoded (probably to save a csname I suppose). Anyways, I redefined it.
@yo' -- that's a good question about amsthm's settings for \topsep. i've tried digging in amsclass.dtx (texdoc amsclass for easier reading) and its history, and can't even find a default setting for \topsep; it's always environment-specific. i'm putting the question on our list -- may end up being an archaeological dig. but i agree, it's counter-intuitive, and on the surface, looks wrong to me. i'll try to get some opinions from my colleagues, but can't promise anything soon.
@barbarabeeton don't worry really, I copy-pasted the code for proof and made the change myself. I need to say though that I've always been a bit unhappy about the fact that proof doesn't use \@thm internally, it would make things a bit more systematic for people who do modifications to the design
@yo' -- not "worrying", but i do get bothered by inconsistencies and unnecessary kludges. note -- i said unnecessary. i've seen kludges that were the best that could be done at the time, hence "necessary". but the ones that i accept are properly documented. this isn't. hence, at the very least, questionable. bah, humbug.
@yo' -- clearly you've never encountered ebenezer scrooge, protagonist of charles dickens' "christmas carol". (a mainstay of our local theater group, adapted from the novel.) meaning: "bosh" or "nonsense", or even "don't bother me".
@yo' -- i "have" found a reason to consider the present behavior a bug in amsthm. the value for \topsep is established in quite a different way in the ams document classes (which have theorem handling built in) from what's in amsthm, and the result is such that there wouldn't be any notice taken here where everything is processed with the ams document classes. really does need to be looked at.
I would like to use LaTeX to make some short math snippets and auto-include them in some other figure which is made with some other program. The other program can import PDF. What is the best way to proceed when all I need is a PDF that contains only the equation I typed?
I found the standalone package, there's just only one problem:
The crop is always tight, which makes it hard to align pieces in the other software. Is there a way to make alignment easier? For example, get some information about where the baseline is. Or do not crop tight, instead make the result PDF as tall as a full line, even if all it has is an $x$ (i.e. something that doesn't extend high or below the baseline)
I'm pretty sure someone must have asked this before but I couldn't find it yet ...
Notice on this screenshot that x^2 has the baseline lower than x. What I'd like is have enough information so I can align such pieces after they have been separately generated.
@yo' I'm creating figures with Mathematica. Mathematica's typesetting is not always good enough for the requirements I have ... I want to make a Mathematica package which will make it easy to use LaTeX-generated labels and annotation on figures. Sometimes I need to make sure that several annotations are at the same height in the figure, even though they're separate pieces.
@yo' If I could e.g. get the PDF file with some accompanying information that says "the baseline is at 10% from the bottom of the PDF" or "the baseline is 2 pt above the bottom", I could use this for alignment.
@Szabolcs ok, it still doesn't define well the problem. TikZ can export the baseline height for you, but it can't include this information in the PDF file, since this notion is unknown to PDF
@Szabolcs how should one document, isolated from the rest, know how big the rest of the documents are? The only way is to define your own \mystrut as a zero-width rule with height and depth equal to the biggest thing you'll be producing. Then stick \mystrut in the body of each document.
@PaulGessler It doesn't. All these are single-line documents. All I need is the baseline position relative to the PDF bounding box. I don't need it embedded in the PDF, I just need to automatically capture this information.
alternatively, this would be good: if two documents use the same font size and only inline math (no display math), then they should have the same height, which would fit something like $M^2 g_2$. Even if the actual content is only as tall as $x$.
@percusse How would I make sure everything gets put in the correct place then without a lot of manual fussing with coordinates? I haven't use pgfplots so I'm not exactly sure how it goes. Maybe it's not as painful as I imagine to use a figure generated with something else and add stuff on top with pgfplots ...
@Szabolcs It is actually easier than you might think provided that you save the mathematica figure without any axes etc. In the manual it is pretty straightforward with examples. Search for Using External Graphics as Plot Sources
This creates the graphics and types DEPTH:<value in pt> in the terminal and in the log file. You can then parse the value from the log and use it. It's the depth of the box (i.e., how far below the baseline it goes). Remember that the pt used here is 1in = 72.27pt
standalone could actually offer this feature, and it would make a reasonable feature request I think. It's like 2 lines of code in the package to put there the message.
@GonzaloMedina It depends on how the command is defined. Since \item actually has no argument, …
% latex.ltx, line 4455:
\def\item{%
\@inmatherr\item
\@ifnextchar [\@item{\@noitemargtrue \@item[\@itemlabel]}}
@GonzaloMedina Note that \LetLtxMacro is needed only for macros defined with \DeclareRobustCommand or with \newcommand and taking an optional argument.
@yo' -- if you have a color printer, it will almost certainly remain colored. if your printer is black/white only, then it's likely that the text colored by hyperref will be in shades of gray.
@David: clearly @barbara is referring to Pink Floyd's A great day for freedom: Now frontiers shift like desert sands / while nations wash their bloodied hands / of loyalty, of history, in shades of grey
seriously, hyperref has some good option settings for doing reasonable things with links in print. i just don't use them every day, and would have to look them up.
@yo' -- the context is from the scene where the solicitors come around asking for contributions for a fund to aid the poor in the dark, cold season of winter. scrooge declines to add his name, and one of the solicitors says, "then you wish to remain anonymous?"
@yo' -- if @paulo's hint doesn't help, i'll try to look it up, but first i have to find some information in the tikz manual on how to divine the numbers to be given with a \clip directive.
hi; Is there a way to quantify the quality of pdf file generated by pdflatex or lualatex on different systems? I am trying to see which system produces "better" pdf, but do not know other by looking at them to guess.
@Nasser almost always tex just uses integer arithmetic to avoid any system differences it is allowed to use floating point in a few limited cases but always very localised (results of floating point glue stretch calculations are never accessible to tex macros for example)
@DavidCarlisle humm. using integer arithmetic would be very slow compare to floating points, but latex is pretty fast. So it must be doing something very smart to be able to do this.
@Nasser integer arithmetic is fast, it doesn't need to emulate floating point in integer it just works in integer that's why the basic unit is scaled point sp. all calculations are integer values in sp. But anyway the author of Tex does know something about optimising that kind of algorithm.
@yo' There are online opensource services which offer an environment for translations, so translators can log in and translate messages without the need of being computer-savvy. :)
@DavidCarlisle -- oh, what a mess! i disclaim any involvement in this. both packages, amsmath and blkarray use box0 all over the place! and to be really obnoxious, they define that box in different ways: amsmath (actually, amsgen.sty) has \def\boxz@{\box\z@} and blkarray (which is dreadfully hard to read!) has \chardef\BA@first@box=0. (i don't know which of you was more obscure, you or michael downes.) i think any attempt to clean this up is beyond my pitiful intellect.
so what shall we tell the person who asked the embarrassing question?
\DeclareTextComposite{\`}{LY1}{i} {236}
\DeclareTextComposite{\`}{LY1}{\i} {236}% again
\DeclareTextComposite{\'}{LY1}{i} {237}
\DeclareTextComposite{\'}{LY1}{\i} {237}% again
\DeclareTextComposite{\^}{LY1}{i} {238}
\DeclareTextComposite{\^}{LY1}{\i} {238}% again
\DeclareTextComposite{\"}{LY1}{i} {239}
\DeclareTextComposite{\"}{LY1}{\i} {239}% again
@egreg not buggy, just documented as HIGHLY VOLATILE, use at your own risk!!!!!!!!
@yo' No but you would if you changed T1 to LY1 (which people seem to do, not sure why really, it was mainly for Y&Y (which did not support virtual fonts)
There is always something that catches me unprepared. For instance I still don't know why a \vbox to \textheight{...} doesn't fit in the page. Or, I really didn't get the way how \begin{theindex} inserts the chapter title.
:20273201 well, \enlargethispage, but it's an ugly hack (and works only on pages with no foot)
@barbarabeeton I was looking at the code just now, didn't recognise any of it, basically I just ran diff on the tracingall output with and without amsmath to find the clash. I wrote that thing over a couple of weekends 20 years ago and hardly looked inside the source since.
@yo' you probably did the first instead of the second, serves you right for using vbox in latex:-)
\documentclass{article}
%\showoutput
\begin{document}
\begin{tabular}{p{1cm}}\vbox to \textheight{\vss}\end{tabular}
\begin{tabular}{p{1cm}}\mbox{}\vbox to \textheight{\vss}\end{tabular}
\end{document}
@DavidCarlisle I enjoyed describing it in my book.
@yo' but a tabular p adds a strut to the paragraph but \vbox without \leavevmode doesn't start a paragraph so you get a dangling strut in addition (vertically) to the cell content. with \mbox/leavevmode the strut is horizontally adjacent to the vbox
The solution of LOF+LOT+BIB in TOC with numbered chapter does not work in my case.
Admittedly I used a template of some other person - when I produce a Pdf with his code all above mentioned sections are numbered. However, when I produce the Pdf of my code, these sections remain unnumbered - whic...
Hey guys, quick question. I am using R to create graphs that are converted to Tikz using TikzDevice. My question is that should I use R to create the title/caption
@masfenix it really depends on your goal. Also, I don't know if you can make the proper figure numbering if you export from R. I've never been fan of that thingy