« first day (1298 days earlier)      last day (3633 days later) » 
02:00 - 23:0023:00 - 00:00

2:39 AM
@egreg Not feeling well? What happened? How are you now? Hope it has become al right. Take care.
 
 
2 hours later…
4:50 AM
I don't think this should be closed: tex.stackexchange.com/q/179451/3954 Many other similar questions, hundreds I'd say, with no code provided by the OP have been left open and received answers.
Besides I think this should be reworded Questions like "Please do this complicated thing for me" tend to get closed because they are either "off topic", "too broad", or "unclear". That's not really true; in my opinion, most of those questions are perfectly clear; they are not too broad in fact, it's generally the opposite, and many of them aren't off-topic.
 
 
1 hour later…
6:12 AM
@GonzaloMedina The Powers, I guess, would say 'off topic' as such questions are only ever applicable to the OP, and that's not really the SE model (questions should add value to 'the community')
@DavidCarlisle, @HeikoOberdiek It seems that if the team take more interest in LuaTeX we will accumulate a long list of bug reports!
 
6:28 AM
@JosephWright Spammer 52127/ling is back (but new email address).
 
6:53 AM
@HeikoOberdiek Zapped
 
7:10 AM
A nice info for you all : thehackernews.com/2014/05/…
 
 
1 hour later…
8:40 AM
@HarishKumar A pain at my left side. Quite certainly a pulled muscle.
 
@JosephWright I just signed up to the luatex list:-) (am already on xetex)
 
@DavidCarlisle Good plan
 
@JosephWright \read15 to \foo is different in luatex:(
 
@DavidCarlisle So it does
 
@Manuel do you mean is there a reason now not to change or was there a reason in 1982 not to define it that way. They are rather different questions:-)
 
9:08 AM
@Manuel LaTeX3 plans currently not finalised but likely to use active chars or similar at the document level (more control over what happens, etc.)
 
9:21 AM
I love this site. <3
It helps me figure out all the weird problems I encounter.
 
@DavidCarlisle Taco is right about the \read thing: Knuth just fixed \csname\endcsname for exactly this 'space adding' reason :-)
 
9:33 AM
Hey fellows, I'm having a problem with increasing my buf_size in texmf.cnf , have no idea how to put the command in cmd :(
 
9:53 AM
@JosephWright no reply on texlive to my (your:-) \typein patch :(
@JosephWright yes although in that case it could "bump" into following text whereas in \read there is always a trailing =
@JosephWright so having got the \escapechar fixes, do we update our luatex binaries by hand since texlive won't pick that up to 2015 I guess, or do we just live with the tlgs as they are for now?
 
@DavidCarlisle That was my question about the 'engine-dependent tests' plan: this sort of thing means the tests will only pass with particular setups
 
@JosephWright yes I think we just go with "current texlive" I don't really want to start building luatex by hand, it's much easier for us all to be using the default (or at least pretest) tl binaries.
 
@DavidCarlisle Yes
 
@Negin What TeX distribution?
 
@DavidCarlisle Reasonable position but we do have to decide how to handle pretest period
 
10:04 AM
@JosephWright well next time I suggest we don't schedule a major latex update during a pretest period, so it won't be so critical:-)
 
@DavidCarlisle OK, fine
 
@JosephWright that is, if one of us is on the bleeding edge and has updated and finds some tests fail we can just decide whether to update the tlgs straight away (or push back to get the binary behaviour reverted) and if we do check in the tlg the rest of the team have some time to update their texlive (or live with the new known failures) which isn't critical unless you're building a "final" ctan release
 
@DavidCarlisle I can live with 'current TL release': for the pretest period I can simply swap between systems as required (expl3 releases are quite likely during the pretest, based on experience to date)
@DavidCarlisle I think 'one of us' = me :-)
 
@JosephWright well I'm on TL2014 as well mostly:-)
 
@DavidCarlisle Ah, 'Joseph or David' then :-)
 
10:11 AM
@egreg That LuaTeX approach pleases me more. But what I think is logical is that, e.g., \Usuperscript is not the same as ^ but in fact ^ is just a shorthand/frontend/user-interface to call \Usuperscript. But it's fine to know that :P
 
@JosephWright I have even run the l3 test suite once:-)
 
@DavidCarlisle :-)
@Manuel Like I said, L3 approach is likely to be based around 'saved' versions of the correct-catcode-tokens (pdfTeX and XeTeX require these) with 'document level' use then as you suggest relying not on catcode but on equivalence to internal versions
 
@DavidCarlisle I don't really mean anything. It was just a thought to see what you all think about it. In any case, the question is equivalent to this two: Why did Knuth do it in the past that way (which in my opinion just restricts freedom)? And is there a reason not to change that in next LaTeX releases?
@JosephWright Yeah, I was answering both of them.
 
@Manuel that's actually a minor difference that doesn't affect choice of user interface, in classic plain TeX \sp is defined by \let\sp^ but it is then essentially a primitive and you can redefine ^ any way you want, the fact \sp is defined in terms of the original meaning of ^ is an implementation detail.
 
@Manuel Can't change LaTeX2e
 
10:16 AM
@Manuel There is clearly no way you could change the meaning of ^ or $ in latex2e, you would break 30 years of legacy documents.
 
@JosephWright @DavidCarlisle And wasn't my intention.
 
@Manuel it might be different in latex3 but the latex3 top level interface isn't designed, so everything might be different, it might look like <article><title>....
 
@DavidCarlisle :-)
 
@JosephWright And I like to see that LaTeX3 approach that you say. I think it's more logic, an easily controlled if all of them are command, and then symbols are a front end for that (just remember I know nothing, I just started using LaTeX 2 years ago, but I hadn't really done anything in latex, I just use it).
 
@DavidCarlisle One real question there is whether to follow ConTeXt on the section/item business (they've deprecated but not removed an approach similar to LaTeX2e in favour of one based on a start/stop-XML-like one)
 
10:20 AM
do u guys if i ask a very simple math question here? I only get troll replies in the math chatroom
 
@Manuel As @DavidCarlisle says, the document layer is 'to be decided': I'm speculating
 
*do u guys mind
 
@DavidCarlisle That's in fact one more reason to look for this “uniformness” isn't it?
 
@hb20007 Ask away
 
@JosephWright html is having such troubles over this, trying to move away from h1-h5 :-)
 
10:20 AM
@hb20007 Can't promise useful answers
 
@JosephWright Yeah :P
 
thanks
is there a functional form we can convert x / (x^2+3) to in order to sketch it?
 
@DavidCarlisle I said to be decided
@hb20007 One for @egreg :-)
 
@Manuel what was that comment refering to?
 
@DavidCarlisle About the <article><title>
 
10:23 AM
@Manuel yes well not everyone would be happy with an xml syntax:-) Personally I haven't hand written a \ and {} syntax document for years but I would imagine that latex3's default syntax will look more like latex2e than xml, but part of the design is that you can program the styling and argument handling without committing to a particular top level syntax.
 
@hb20007 When I did curve sketching last, the procedure was to break down into 'parts', e.g. spot that at large values this falls to zero
 
@Joseph I see. So what else can we infer?
 
@JosephWright Surely not!
 
@egreg You're the real mathematician here
 
@JosephWright I just pretend I suppose:-)
 
10:27 AM
@hb20007 What do you mean by “functional form”? That is a functional form.
 
@hb20007 Been a while since I did this
 
@egreg I mean a different form that would make sketching possible. Or just would u sketch that?
 
@hb20007 When I did curve sketching, a few years ago, there were a set of rules for doing this sort of thing by hand: look for the maxima, limiting behaviour, etc. Regrettably, I can't think of all of them now!
 
@hb20007 I can't see of a different form. Actually it's quite a nice function to sketch. It's odd (that is, f(-x)=-f(x)), its derivative is (3-x^2)/(x^2+3)^2, so it has a maximum at the square root of 3 and a minimum at the symmetric point. The curve tends to zero at both infinities. The derivative is 1 at 0.
 
@JosephWright @hb20007 the one I remember was type plot x / (x^2+3) into google and look at the picture
 
10:33 AM
@egreg Ah yes, that's the procedure I remember!
@egreg Surely
\documentclass[tikz]{standalone}
\usepackage{pgfplots}
\begin{document}

\begin{tikzpicture}
  \begin{axis}
    \addplot {x / (x^2 + 3)};
\end{axis}
\end{tikzpicture}

\end{document}
 
@egreg @JosephWright THANKS you guys are awesome
 
@hb20007 No problem: I didn't actually do anything!
 
@Joseph: did David already break LaTeX today? :)
 
@PauloCereda No, he found a bug in LuaTeX
 
@JosephWright they say it's the intention that matters
 
10:37 AM
@JosephWright ooh so he's the new Lua expert. :)
 
@JosephWright An undocumented feature, I'd say. ;-) What LuaTeX should do is to allow \endlinechar having any value and normalizing it a usage time (if not in the interval 1..127 use no char).
Which is what TeX does (with different bounds)
 
@egreg That sounds like what Taco will implement
 
@JosephWright :)
 
10:57 AM
@DavidCarlisle Ok. The first post I made was just a comment about those catcodes which I don't fully understand, to see your thoughts, nothing more.
 
kan
Hello!
 
@Manuel You should consider when TeX was born and what was available at that time: roff used (and still uses, it's the program that formats man pages) “shifting codes”
There are two display modes\&.  The first, multiline mode, is the
default\&.  It only works if the \fBTERM\fP parameter is set to a valid
terminal type that can move the cursor up\&.  The second, single line
mode, is used if \fBTERM\fP is invalid or incapable of moving the
cursor up, or if the \fBSINGLE_LINE_ZLE\fP option is set\&.
Knuth thought it was the easiest way for typing documents.
 
hello
@egreg what do you do IRL ? I've been always seen you on that website ^^
 
So... TAOCP will be released in ebook form. But I still prefer my real books. :)
 
@hb20007 You're lucky you didn't catch them in cricket mode.
 
11:10 AM
@StephanLehmke David was probably busy breaking LaTeX for the n-th time. :)
 
12:07 PM
@melostap I teach math.
 
@egreg in which field, if I may ask ?
 
@melostap Algebra and Didactics of Math
 
@egreg great!
 
@melostap That's why I'm able to sketch a graph without bothering to actually sketch it. ;-)
 
@egreg: holiday today? :)
@egreg: how are you feeling today?
 
12:19 PM
@PauloCereda Yes, S. Zeno. I'm quite fine, thanks.
 
12:37 PM
@egreg I see. ^^
 
@egreg It's XeTeX, Version 3.1415926-2.3-0.9997.5 (Web2C 2011)
 
@Negin On what OS?
 
@egreg Windows 7
 
@Negin Well, it seems to be quite outdated. Are you on MiKTeX 2.8?
 
@egreg where can I find that out?
@egreg It certainly is outdated :D, but that's all I've got on university system!
 
12:49 PM
@Negin Anyway, exhausting the buffer size can mean a programming error. What to do on Windoze is beyond my expertise, I'm afraid. ;-)
 
@egreg I'm fine with increasing it on mac and linux but have no idea about how to do it on windows! :D the same problem's here ;)
 
@Negin Judging from the doc at miktex.org, initexmf --edit-config-file=texmf.cnf might do.
 
@Negin you did not answer my comment on the question: what program is giving the error?
 
@DavidCarlisle by what program u mean my editor? it's TeXmaker
 
@Negin no I mean what program is trying to read your big eps and running out of buffer space. actually given your comment above I guess it is either xetex or dvipdfmx
 
12:57 PM
@DavidCarlisle oh I got it, it's xelatex
 
@Negin well xelatex doesn't need to open the eps file so why is it filling any buffer, what happens if you use \includegraphics[bb=0 0 100 100]{mylargefile.eps}
 
@DavidCarlisle I get the error : ! Unable to read an entire line---bufsize=200000.please increase buf_fize in tekmf.cnf.**ERROR** file ended prematurely Output file removed.
 
1:20 PM
@DavidCarlisle Sorry couldn't remove my previous message, the process exites normally!
@DavidCarlisle really interested in how this optional command works, any tutorials available?
 
@Negin thought it might:-) but that os presumably the wrong size. Your EPS file should have a %%BoundingBox: a b c d comment (otherwise it's invalid EPS) and the numbers you pass to the bb options should be those. I woudl guess it is at teh end of the file (as tex was just trying to skip lines until it found it and died on a long line. but we should move to teh question. (try texdoc grfguie or texdoc graphics for documentation)
 
@DavidCarlisle I was wondering increasing buffer manually apart from my question ;) but they pretty mixed up. I'll keep commenting there
@DavidCarlisle Thank you so much it's really useful
 
1:41 PM
@Negin You might try running eps2eps on the offending file, that should move the bounding box information at the correct place.
 
@egreg How is it now? BTW how did you pull it? ;) May be your long tours on your silver wing has this effect!
 
@HarishKumar I'm fine, thanks; still a bit of pain, but bearable. I really don't know how it happened: it started on Monday morning, the tour had been on Saturday and just around 300km, a trifle. ;-)
 
@egreg Just 300km? Uh! You are getting old ;)
 
2:00 PM
@HarishKumar I'm planning one to Cortona, around 600km. It was an important Etruscan city in the eighth century b.C.; very nice town, with an important masterpiece of medieval art, the Annunciation by Beato Angelico.
It's an older town than Rome. Some people say that the last Etruscans rejoiced seeing Hannibal beating the Romans at the Lake Trasimeno battle, which can be seen from the town walls.
 
2:32 PM
@egreg On silver wing? ;) You seem to be fond of travelling to new places. A born tourist! That city looks fascinating though.
 
2:44 PM
@egreg can you believe it:(
Both answers are equally acceptable, but I like the \gloss macron defined by egreg. — Sverre 41 mins ago
 
@DavidCarlisle Yes, I can. :P
 
3:03 PM
@HarishKumar It is! I've been there for two weeks several years ago for a summer course during my PhD. Another fascinating city nearby is Perugia, also of Etruscan origin. Some parts of the Etruscan walls are still existing, in particular one of the doors (it has a small Roman addition, the balcony on the left).
It is between 2300 and 2400 years old.
 
Hi, small question.. I got this \newcommand{\code}[1]{\lstinline[basicstyle=\ttfamily\normalsize, prebreak=]{##1}} inside \IfFileExists. It gives me more errors than I have fingers and toes. It works fine outside \IfFileExists. What's happening here?
 
3:23 PM
@1010011010 You can not use verbatim (or similar like listings) inside the argument of another command
 
@DavidCarlisle Why is it fine to declare environments like this one: `\newenvironment{code}{%
\VerbatimEnvironment
\minted@resetoptions
\setkeys{minted@opt}{}
\begin{center}
\begin{minipage}{\linewidth}
\begin{VerbatimOut}{\jobname.pyg}}
{%
\end{VerbatimOut}
\minted@pygmentize{r}
\DeleteFile{\jobname.pyg}
\end{minipage}
\end{center}}`?
That's the minted verbatim environment
It's also embedded inside the same IfFileExists
 
@1010011010 it wont work in general inside another command, it's not possible, in a verbatim envionment you can make % an ordinary symbol and } etc together with special parsing looking for \end{verbatim} (or whatever) to restore things, If you have a macro \iffileExists or anything else that goes \foo{\verb| ...|} the the macro argument is scanned finding } before the verbatim command is executed so it is too late to change any meanings, the characters have already been read
 
@DavidCarlisle Oh, actually that makes sense. Thanks. :-) I'm assuming there's no real workaround to it?
 
@1010011010 Try \newcommand{\code}{\lstinline[basicstyle=\ttfamily\normalsize, prebreak=]}
 
@egreg Didn't seem to do the trick. Shame.
 
3:38 PM
@1010011010 But why are you doing that inside \IfFileExists?
 
@1010011010 this is tex there is always a workaround, but you need to make sure that you don't read the verbatim stuff before you execute the verbatim (or listings or minted) command. Hmm I may have mentioned this before but the details depend on the details and are much easier given a real example on site.
 
@DavidCarlisle @egreg I want to offer a "compatibility mode" for when whoever compiles the document elsewhere can instead use the listings package when the minted package isn't available or not fully functional. For that I have defined the Rcode environment (which is called Rcode in both true and false values for IfFileExists) and the idea is to do the same for inline code listings using a \code macro, also defined for both true and false.
 
@1010011010 as I say a real question with a real example would be easier but you can do \iffileexists{..}{\let\myverb\verb}{\def\myverb{\lltinline....}` then use \myverb| verb stuff| without pre-scanning the verbatim argument
 
@1010011010 It's better to use options, in my opinion; \IfFileExists could result in a false positive: minted can exist, but the user can't use it because Python is not available.
 
@egreg That's precisely how I intend to use it. Partial functionality should allow the user to compile a "backup" version with similar aesthetics and/or visual appearance.
 
3:46 PM
@egreg or minted and python can both be available but the user has never managed to get them to work together :(
 
4:01 PM
If there's any way I can get this to work inside the IfFileExists then I'll gladly hear it, I'm willing to go far for this one because it's such an essential part of the visual appearance.
Especially if there's a way to include the \lstset
 
Hello there
How do I use a metapost file
I have compiled it
I have got 0 errors
it generates another set of files
how to embed them to my tex document
 
@subhamsoni they should be postscript files probably with extensions like .1 but it is simple postscript tex can understand so with pdflatex at least \includegraphics shoudl work
 
hmmm nothing like .1 here
only .mps
.eps
 
@subhamsoni ah probably in the 20 years since I last tried they've given it a more sensible default extension:-) just use \includegraphics{file} without an extension and it'll probably just work, picking up whatever it needs
 
So now what Should I include??? :(
 
4:10 PM
@subhamsoni What's the name of the produced .mps file?
 
sorry mps is not produced, thats the name I gave. the files generated were drawing.eps , drawing.mpx, drawing.log and my file drawing.mps
 
Gotta love this turtle.
 
its just so funny @PauloCereda
:)
 
@PauloCereda I think its owner needs a biology lesson: that is not where a turtle's teeth are.
 
4:16 PM
anybody with my metapost question
:(
 
@subhamsoni you presumably told it to use .eps as the output extension?
 
but includegraphics generates an error
as file not found
 
@subhamsoni The file with metapost code should be named with extension .mp. Depending on the contents, running mpost on it will produce an output file named with extension .1 (or other number), or, if you set the extension in the code, with extension .mps.
 
ok will do it right now
 
@subhamsoni yes because pdflatex can't read arbitrary eps, that's why I asked where you had set eps as the output extension, if you use a default mps extension latex will know it's metapost, or of course you could treat it as arbitrary eps and use eps2pdf to make a pdf file and then include that but its more efficient to let latex read it directly
 
4:23 PM
made the file as .mp
the files i get are
drawing1.eps
 
@DavidCarlisle LOL
 
@subhamsoni Is the file long? If not, you can show it here
 
@DavidCarlisle en.wikipedia.org/wiki/Turtle: "Instead of having teeth, the upper and lower jaws of the turtle are covered by horny ridges."
 
@HeikoOberdiek still not on its back though:-)
 
5:03 PM
Great! The documentation on TeX Live in English disappeared from TL2014. :)
 
@egreg all of it, or just some? I'm sure I may have seen some doc??
 
@DavidCarlisle Try texdoc texlive; only the HTML “cover” appears, but when you click for the doc in English…
And texdoc texlive-en shows the "Building TeX Live” document.
> tlmgr info texlive-en --list
package:     texlive-en
category:    TLCore
shortdesc:   TeX Live manual (English)
installed:   Yes
revision:    34163
sizes:       doc: 1073k
relocatable: No
collection:  collection-basic
Included files, by type:
doc files:
  texmf-dist/doc/info/tlbuild.info
  texmf-dist/doc/texlive/tlbuild/tlbuild.html
  texmf-dist/doc/texlive/tlbuild/tlbuild.pdf
 
@egreg u there
there was a poweroff
so couldnt reply
 
@subhamsoni Don't worry. I was updating my OS.
 
u use linux???
 
5:17 PM
@subhamsoni No, but also Mac OS X some times needs updates.
This one required restarting.
 
 
2 hours later…
7:33 PM
@egreg texdoc texlive gives me a pdf "the tex live guide 2014" ? (texlive-en.pdf)
 
@DavidCarlisle Did you update today?
 
@egreg yesterday (should I try again:(
 
@DavidCarlisle So it's today's update.
 
@egreg trying now...
 
@DavidCarlisle You're bold.
 
7:35 PM
@egreg well worse that happens is I disable latex (and I don't really need that:-)
 
@DavidCarlisle I restored revision 33927; lo and behold, the documentation is there!
 
@egreg I have no idea what you're on about with your tikz \& replacement comment:-) (feel free to edit the answer if you think it's important)
@egreg oh quite a big update this close to the cut-off 2/75,...
 
@DavidCarlisle When a \matrix is processed in the argument of a command, & as a column separator is not allowed any more, because TikZ does some catcode trickery. So a replacement must be defined. For a one column matrix it's unimportant, but…
 
@egreg cor you've been learning tikz in your spare time
 
@DavidCarlisle I saw it somewhere and I had to use it a couple of times.
Mainly with tikz-cd
 
7:48 PM
@egreg seems like you're the only one who knows about it:-) they all by you
 
@DavidCarlisle Ehm, didn't you see user:4427?
@DavidCarlisle Anyway, I'm a big expert in , having 145 answers with 873 votes. Compared to 97 answers with 753 votes of someone else…
 
@egreg oh yes;-)
 
@DavidCarlisle 754 votes now. ;-)
 
@egreg but I expect that someone else has better quality answers (like one today about tikz matrix)
 
8:06 PM
@egreg: Interesting... here I thought I would dupehammer the question you just commented on...
...but it didn't work.
 
@Werner original tag was just matrices
 
Hi guys! Do you know how to obtain this i.imgur.com/aIHFxaE.png?
 
@DavidCarlisle Of course... thanks tag-janitor. ;)
 
@DavidCarlisle Yes, I changed into
 
@DavidCarlisle: I also just re-read:
121
Q: Increase close vote weight for gold tag badge holders

Travis JIn order to increase the efficiency with which poor quality questions are closed, it could make sense to have weighted close votes for a small subset of qualified users. A very good way to measure the qualified users would be to leverage the tag badges. However, it needs to be rationally limite...

 
8:09 PM
@Werner I too can't hammer close it. Which is reasonable, as I changed the tag.
 
@egreg @DavidCarlisle nailed the reason though... ;)
 
@Werner I guessed that as it happened to me in reverse someone had asked a question about some editor I'd never used and I put on a close as dup vote as it looked a bit like another q and it turned out the OP had tagged the question as tex-core originally for no good reason, so I had powers I didn't know:)
 
@egreg You don't have a gold badge?
 
@Werner Not ATM.
 
@Werner he doesn't know enough about matrices (I on the other hand even know about tikz matrix)
 
8:11 PM
@egreg Oooo, in the words of Barney Stinson... "Challenge Accepted!" :)
 
@egreg I don't think it matters who edits, it's just the original tags that count
 
@Werner This is what shows my front page: ten gold tag badges. ;-)
 
@DavidCarlisle I know. We all know.
@egreg Yes, but no ... yet.
 
@Werner I rely on egreg to fix up my answer though:-)
 
@DavidCarlisle He should also change them to CW then. I need something to keep you off my tail.
 
8:13 PM
@DavidCarlisle Yes, tomorrow I'll be explaining the determinant. I'm asking myself how I can do it, knowing nothing about
 
@egreg 10 gold badges hanging on a wall, and if one gold badge should accidentally fall there'd be 9 gol ...
@egreg you're a university lecturer, make it up as you go along
 
@DavidCarlisle That's what I usually do.
 
The current pdfTeX of TL2014 seems to be broken, details:
http://tug.org/pipermail/pdftex/2014-May/008900.html
2
\documentclass{article}
\begin{document}
\noindent
\rlap{\textbf{Hello}}World
\end{document}
 
@HeikoOberdiek Eeek!
@HeikoOberdiek Karl is going to be so pleased
 
@HeikoOberdiek oops and Karl said we were brave making inputenc not trash luatex:(
 
8:21 PM
@HeikoOberdiek Who uses pdftex anyway nowadays...
 
@HeikoOberdiek WOW! This is nice!
 
@JosephWright I have put him into the CC field, thus he should get notified.
 
@StephanLehmke most tex users:-)
 
@HeikoOberdiek How did you stumble over this?
So who's taking
 
@DavidCarlisle Font encodings! Bleah!!
 
8:26 PM
2
Q: Arguments to function not losing braces

A.EllettThis MWE is boiled down so much the usefulness of the code is lost. However, the code nevertheless still exhibits the problem I'm encountering. \documentclass{article} \usepackage{pgffor} \usepackage{etoolbox} \makeatletter \newcommand\aetest[1]{%% \foreach \myl in {#1} {%% \expan...

@StephanLehmke So not dependent on OS => some people very keen on this
 
@JosephWright I decided to leave it for egreg
 
@HeikoOberdiek This suffices
\noindent\rlap{\bf Hello}World
\bye
 
@JosephWright I experimented with different methods for
1
Q: No hyperlink for server name part of multi email text

M0HS3NI am trying to have something like {name1, name2}@server.com where name1 and name2 have hyperlinks of name1@server.com and name2@server.com. This is my code: \usepackage{hyperref} \{\href{mailto:name1@server.com}{name1}, \href{mailto:name2@server.com}{name2}\}@server.com but in the out put, }@...

 
@HeikoOberdiek hmm your test file is OK in a tl2014 cygwin updated a few minutes ago
 
@JosephWright Yea but can you imagine what it means to manage arbitrary Unicode->Glyph mappings this way, for customers using up to 30 languages :-(
 
8:28 PM
@StephanLehmke so what do you use xe or lua ?
 
@DavidCarlisle Well we are using pdftex of course, I was jesting.
 
@DavidCarlisle You have used PDF mode? I have tested with x86_64-linux and i386-linux.
 
But font installation is a real adventure.
 
@StephanLehmke Oh yes, quite true
 
@HeikoOberdiek ignore previous comment apparently I can't remember what \rlap is supposed to do :(
2
 
8:30 PM
@StephanLehmke But for many users that might not be a bad thing: can't do 'fun' font things :-)
 
@DavidCarlisle \relax, \rlap is not a genuine LaTeX command ;-)
4
 
@JosephWright Our situation is maybe a little uncommon as customers buy arbitrary fonts and then of course expect to get each and every glyph in there mapped from unicode.
 
@StephanLehmke As so often, 'standard' use depends on your point of view
 
@StephanLehmke tell them they wasted their money and they are better off using computer modern
 
@StephanLehmke For example, where I sit, 'most users' are writing a thesis using TeX, whereas for someone like Barbara Beeton I guess 'most' users are typesetting articles written by others, ...
 
8:33 PM
@DavidCarlisle Well after reading up on font licensing conditions, I'm tempted to tell all customers to use free fonts.
 
@JosephWright and here we use 3b2
 
@StephanLehmke Sounds about right: font licenses are a nightmare
@DavidCarlisle Yes, I'd guess that
@DavidCarlisle Expensive
 
The toughest part so far was creating a font encoding for greek which allowed mixed greek and latin text with correct hyphenation.
 
@JosephWright not as expensive as the man hours it would take to recreate the formatting in something else
 
@DavidCarlisle In the short term, certainly
 
8:40 PM
Hi people, I have a very strange problem with \includegraphics.
I can't prepare an MWE now, but I'm asking here in case somebody might know what's going on.
So I'm doing this:
\setbox\@tempboxa=\vbox{\includegraphics[height=42pt]{file.pdf}}
\show\ht\@tempboxa
and I get 41.9993pt. Anybody knows what might be happening?
 
@mbork Rounding :-)
 
(To prepare an MWE, I'd have to somehow include the pdf file, and I'm not sure how to do it properly.)
Hm. If I use [demo] with graphicx, everything seems fine...
 
@mbork PDFs use 'big points', so the drivers work out the size in bp then 'convert back' to pt to make sure things line up
 
@mbork do you know how big .0007pt is ? :-)
 
Aha. So if I convert my pdf to, say, png, would it be better?
Yes. But what I want to do is to make sure that the box's height is a multiple of`\baselineskip`
 
8:44 PM
@mbork no it will make it a bitmap so spoil it (if it is a scalable image)
 
so I wanted to compare things in sp
@David: not really, the pdf I'm going to produce will be printed at 300 dpi (and I just know that),
so scalability is not a huge deal for me.
 
@mbork well round it to the nearest pt before comparing (or 100sp or whatever works) or use \setbox\@tempboxa=\vbox to 42pt{\includegraphics{..}\vss}
 
@mbork Use everything in bp units.
 
@DavidCarlisle: the latter idea wont work - I don't know the height in advance (in general).
 
Thats the "postscript pt" anyway.
 
8:47 PM
@mbork ok well round then (just divide by some integer and multiply by the same one
 
@StephanLehmke: but inside (La)TeX everything is in sp anyway, so I don't see how it would help
 
@JosephWright Indeed, Karl is not happy, direct quote: "This is terrible news."
 
@HeikoOberdiek Yes, I saw
 
@DavidCarlisle: I guess that using \numexpr would be better than \divide, since what I want is rounding and not truncating, right?
 
@HeikoOberdiek What's the 'fake space' comment about?
 
8:49 PM
@mbork possibly
@mbork if you don't know the size what options are you passing to includegraphics in your real case?
 
@mbork If you define \baselineskip in bp and then give the height of the image as 5\baselineskip or whatever, then everything will be fine also on sp level.
 
OK, so I'll try this idea. Thanks a lot! (BTW: the rationale is that I'm writing a replacement for the figure environment, using \vadjust, so that it does not float (I want manual positioning), and ayt the same time maintain the grid...)
@StephanLehmke: Ah, nice idea. I'll think about it. I hope this wouldn't break anything elsewhere in my class.
@DavidCarlisle: [width=0.8\textwidth], for instance. And - in the real case - I have a caption in a smaller font, so even if I do [height=10\baselineskip], the caption will break the multiplicity.
 
@StephanLehmke well not necessarily, height= is defined as a scale so it still does tex arithmetic to work out the scale fator, although using bp would avoid conversion from pt to bp in addition
 
@DavidCarlisle Ok so the safe solution would be to put the image in a \vbox to the correct height with some \vss thrown in.
 
@StephanLehmke er like:
9 mins ago, by David Carlisle
@mbork well round it to the nearest pt before comparing (or 100sp or whatever works) or use \setbox\@tempboxa=\vbox to 42pt{\includegraphics{..}\vss}
 
8:56 PM
@DavidCarlisle Nobody ever listens to you.
2
 
@JosephWright AFAIK there were some efforts to improve compatibility to PDF/A. TeX's space handling by using a skip instead of a space characters is not very friendly for PDF/A.
 
@StephanLehmke I know:(
 
@HeikoOberdiek Ah, right
 
@DavidCarlisle, @StephanLehmke: that's what I want to do - however, I want to compute the right height...
 
From the NEWS:
- new primitives for fake interword spaces:
\pdfinterwordspaceon, \pdfinterwordspaceoff, \pdffakespace
 
8:59 PM
@HeikoOberdiek I think I vaguely remember this now
 
@mbork take the height divide by baselineskip then force it to be that exact multiple of baselineskip,
 
@DavidCarlisle: that's what I'm doing. I just want to round it up. \divide rounds down, \numexpr rounds t the nearest integer...
 
@HeikoOberdiek Not sure they actually work
@HeikoOberdiek I tried
\documentclass{article}
\begin{document}
ABC \pdffakespace EFG
\end{document}
and got
kpathsea: Running mktexpk --mfmode / --bdpi 600 --mag 0+61/600 --dpi 61 dummy-space
mktexpk: don't know how to create bitmap font for dummy-space.
mktexpk: perhaps dummy-space is missing from the map file.
kpathsea: Appending font creation commands to missfont.log.
 )
!pdfTeX error: /usr/texbin/pdflatex (file dummy-space): Font dummy-space at 61
not found
 ==> Fatal error occurred, no output PDF file produced!
 
And I don't want to just truncate and add one, since the height might just be exactly a multiple of \baselineskip
(in case someone said [height=10\baselineskip] and there were no caption)
 
@mbork Subtract 1sp, round down, add 1.
 
9:02 PM
@mbork add half a baselineskip before dividing ?
 
@JosephWright Isn't this a sign you're creating dvi?
 
Well, I tried adding 0.5\baselineskip, it didn't work (with \numexpr) - I didn't investigate further. And because of these rounding issues with pt <-> bp conversion, I'll probably stick with rounding to the nearest, say, 2048 sp. Thanks for your help!!!
 
@StephanLehmke No, I think a map problem
 
@JosephWright I am trying, too. Next bug-reports ...:-(
 
@mbork This works for me:
\baselineskip12pt
\@tempdima72pt
\@tempcnta\@tempdima
\advance\@tempcnta\m@ne
\divide\@tempcnta\baselineskip
\advance\@tempcnta\@ne
\showthe\@tempcnta
 
9:12 PM
@mbork \numexpr divisions round.
 
@egreg: I know that.
 
@egreg ... which is an abomination.
 
@StephanLehmke Yes.
@mbork Here's something that should work
\documentclass{article}
\usepackage{graphicx,expl3}
\newsavebox{\mborkbox}

\ExplSyntaxOn
\cs_new:Npn \normalizetobaselineskip
 {
  \fp_eval:n
   {
    ceil ( \dim_to_fp:n { \ht\mborkbox } / \dim_to_fp:n { \baselineskip } )
   }
  \baselineskip
 }
\ExplSyntaxOff

\begin{document}

\setbox\mborkbox=\hbox{\includegraphics[height=42pt]{example-image-a.pdf}}
\setbox\mborkbox=\vbox to\normalizetobaselineskip{
  \vss
  \box\mborkbox
}

\showthe\ht\mborkbox

\setbox\mborkbox=\hbox{\includegraphics[height=3\baselineskip]{example-image-a.pdf}}
 
@egreg: thanks a lot, though I don't use expl3;). But maybe it is a good idea to post my problem as a question (even though I have it solved now), so that people might find the answer in the future (in case someone wants to do something similar)? If yes, I can do it tomorrow (now it's a bit too late - it's 23:21 here and I need some sleep;)).
 
@mbork That puts the excess space at the top, so the bottom of the image will be on the baseline
 
9:23 PM
@JosephWright The "fun" continues: ntg.nl/pipermail/ntg-pdftex/2014-May/004049.html
 
@HeikoOberdiek One list I'm not one
 
9:37 PM
How can I stretch the unicode symbol ↡ to a decent size in latex?
 
@Alizter depends: you could use \large but if you mean really stretch you want one of the commutative diagram packages with arbitrary long arrows and customisable heads
 
@DavidCarlisle Thank you, I tried huge and it worked for what I needed. I do not think I need a whole package. Thank you very much! Good day!
 
How can I use \pmb for all formulas in my document? I suppose I can do that somehow.
 
9:55 PM
@Cortizol you should feel guilty for each symbol that you inflict pmb on, to inflict it on a whole document is probably against international treaties...
@Cortizol if you want all maths to be bold use \boldmath
 
@DavidCarlisle I don't think that pmb is great idea generally, but I just want to see how it looks.
 
@Cortizol It will look horrible:-)
@Cortizol are you using a font family that has no bold?
 
@DavidCarlisle I cannot use packages for this arrow as it is being rendered with MathJaX. Do you know of anyway to have a double headed down arrow in latex that can work in MathJaX? Also \Huge wasn't enough.
 
@DavidCarlisle uhm....I suppose not. I can use \mathbf{} if you mean that. And for \boldmath is there a way to use that on every formula in document? Of course, I can do that on each formula separately, but there must be some more elegant way.
 
@Alizter there is mathjax support for amscd although I can't remember if that has double arrows (t's the simplest of the cd packages) You shoul dbe able to specify an arbitrary size (as you can in mathml) I'll check in a bit
@Cortizol \boldmath makes all math bold (until the end of the group) so just stick it after \begin docuemnt
 
10:05 PM
@DavidCarlisle Hey, it looks nice. Nothing so horrible, except that some symbols now are looking somehow childish. And I am rigorous person when we write maths :)
@DavidCarlisle is there similar command for \pmb like \boldmath, um? Just to try
 
@Alizter try this using your 21a1 character (I havent tried just from teh mathjax doc) \unicode[.55,0.05]{x22D6} % less-than with dot, with height .55em and depth 0.05em
@Cortizol no why would you even want to do that \boldmath uses whatever bold fonts you have available, so it makes sense to use as a document setting, \pmb makes a rigid box and sets it three times one over the other, offset slightly it's a method of (very) last resort for use on very occasional symbols.
 
@DavidCarlisle I literally just found that :P
Thank you very much!
 
@DavidCarlisle Okey, you convince me :)
 
@DavidCarlisle However it does not enlarge the symbol itself.
 
@DavidCarlisle Again, I must ask, maybe you know, how to obtain this symbol i.imgur.com/R68qs7z.png?
 
10:13 PM
@Alizter it doesn't? as I say it must be possible as it is in mathml (<mo mathsize="10">&#x2a1a;</mo> but it's late here and I can't be bothered to try:-) ask on the mathjax help forum mostly they are in the states so it's a more sensible time:-)
 
@DavidCarlisle We are both in the UK :p
 
@Alizter well get the Americans do do the work:-)
@Cortizol if you can't find a font with it you could use \stackrel with \rightarrow and \sim, probably
 
@DavidCarlisle I tried before, but it not looks so good. \rightarrow is longer than \sim, and \sim need to be closer to \rightarrow. Never mind, I will ask separate question here (not chat) if I couldn't solve it.
 
10:53 PM
A pity Alan Munn is not here. I'd have liked to tell him my impressions on the last piece of music on the radio, Barber's “Adagio for strings”. I find it superfluous. ;-)
 
02:00 - 23:0023:00 - 00:00

« first day (1298 days earlier)      last day (3633 days later) »