TeX, LaTeX and Friends - 2015-01-14

« first day (1536 days earlier) ← previous day next day → last day (3380 days later) »

Christian Hupfer

12:02 AM

Have a nice time ... I am off (to bed)

@cfr Do you have any news on the beta-testing of forest? (I ask you because I didn't login for some weeks).

@Jubobs This many? :)

Hi, I have a bit of a problem with lstlisting captions in xepersian (tex.stackexchange.com/questions/222590/…). cfr suggested a solution using titles instead of captions, but now I cannot have a listoflistings. Can anyone help with that?

The code I posted in the PPS suggest that there mustn't be any problems to use lstlisting captions with xepersian. But we cannot get it right.

1 hour later…

1:22 AM

@GonzaloMedina No, sorry. I was wondering about that earlier because I've not heard anything further.

@Furihr See my edit. I think I got it! You have to look at the documentation for bidi to find the solution, though. (Or what I think is the solution.) ;)

user image

@Furihr The reason for all the boxes is because the only font I've got on this machine appears to lack all punctuation characters!

I saw it.

thanks. There is just an issue with the numbering. The caption numbers in the listoflistings are LTR oriented.

1 hour later…

3:02 AM

@JosephWright Maybe, but I've decided on a non-ideal voting pattern for this one ;).

3 hours later…

6:03 AM

@PaulGessler Just south of Madison, actually. Not too far away, though :)

@ChristianHupfer <3

3 hours later…

8:46 AM

It's multiple choice time...

	What is FORTRAN?

	a: Between thre and fiv tran.
	b: What two computers engage in before they interface.
	c: Ridiculous.

[paulo@cambridge ~] $

@DavidCarlisle: ^^ :(

@PauloCereda lol

9:16 AM

Tonight I have to go to the doctor and see what can be done about my left hearing. :)

@egreg I've added as much of the info on primitives as I can for that 'engine variation' question. I wonder where it's meant to be going: this is the sort of thing Heiko or I worry about, not most end users

@JosephWright Not much further; as I commented, just dealing with primitives is not enough.

@egreg I know

@egreg Not really sure what is wanted

9:40 AM

Master of puppets, I'm pulling your strings!

@JosephWright Too localized? tex.stackexchange.com/questions/209165/…

@Johannes_B Well 'it worked' isn't going to help others, so either off-topic or (perhaps) unclear?

@JosephWright Dunno, your decision ;-) -- unclear

@Johannes_B I've duped as there is at least some link to another question

@JosephWright Ok, Thank you.

2 hours later…

11:38 AM

@PauloCereda 1984 all over again

books.google.co.uk/…

@SeanAllred cool. Congratulations on the new career! I bet I could guess the company if you wanted me to. ;-)

@DavidCarlisle Oh my!

@PauloCereda "the old jokes are the best" :-)

@DavidCarlisle did you read the PC ad that follows :-)

@DavidCarlisle I particularly like #7

12:02 PM

Guys, is there any command (probably GS or something) that I could run against a PDF in order to "correct" it?

@JosephWright ohh 2 cpu system

@PauloCereda rm

@DavidCarlisle LOL GS did the trick. :)

@PauloCereda gs -sDEVICE=pdfwrite -dNOPAUSE -dQUIET -dBATCH -sOutputFile=output.pdf input.pdf

@JosephWright Thanks! :) Why can't @DavidCarlisle be helpful? :P

@PauloCereda sniff. I thought you meant a different type of fix (to the text not the internal structure) and was writing some suggestions (eg pdfxchange under wine) when you said you'd fixed it :-)

12:10 PM

@DavidCarlisle <3

@David: I wub you. :) But you are still in the naughty list, let's see if things change during this year. :)

@PauloCereda I shall be reformed, I shall use vim, I shall use Word, I shall swap my Android phone for an iPhone

@DavidCarlisle Don't do that. :)

12:46 PM

@DavidCarlisle yes, yes, no, no :)

Christian Hupfer

Good maen again ...

@ChristianHupfer hi

Christian Hupfer

1:02 PM

@yo': yo! ;-)

@ChristianHupfer I'm here, but I'm seriously busy today, and will be so tomorrow

Christian Hupfer

@yo': Vice versa... I had no time for TeX.SX yesterday... today and tomorrow more 'spare' time ;-)

1:33 PM

Anyone around with experience creating new arrow tips in the new-and-improved arrows.meta paradigm?

2:09 PM

@PaulGessler I bet you could – fancy a guess? XD (I'm only a little bit excited.)

@SeanAllred I'm guessing Epic, but I might be wrong

2:38 PM

@PaulGessler Correct :) It's actually a very nice fit fir me, it seems. I've been involved in healthcare texh since before I graduated high school -- my mother and sisters all work in healthcare (administrative arm), so I already have quite a bit of domain knowledge

It beats the hell out of working on weapons systems for the government :/

@SeanAllred excellent! I hope you enjoy it there. Madison is a great city too; I don't get over there nearly as often as I should. Well, as I said, congratulations! If you're ever in Milwaukee (or if I'm in Madison) we can have our own little TeX User's Meeting if you'd like. :-)

Hi, any pgfplots experts here? I want to read data from a file and apply ln() on one column, but I can't find any info how to apply the ln.. I tried
\addplot+ table [x expr=\thisrowno{0}, y expr=\ln{\thisrowno{1}}] {data.txt};
But this doesn't work.

@PaulGessler Thanks, and that would be awesome! I'll probably be in Milwaukee on a pretty regular basis (once every month or two) to fly out and visit family, so the opportunity may very well arise :)

@Shepherd ln() as in natural log?

yes

apparently this part: y expr=\ln{\thisrowno{1}} is probably incorrect, any pointers?

PGFPlots uses the natural log by default for logarithms

2:47 PM

@PaulGessler Willy Wonka!

\addplot+ table [x expr=\thisrowno{0}, y expr=\log{\thisrowno{1}}] {data.txt};

doesn't work either

I wanna work in a chocolate factory. :)

This is related tex.stackexchange.com/q/105766/17423

5

Q: Using mathematical function in PGF/TikZ

I want to use ProbXpos for example to position tic marks and to position line ends. It seems to calculate as I intend, but how can I use the calculated value? The example below prints a sample result value. If I un-comment the commented lines it won't even compile. Also -- minor point -- some i...

tikz-pgf pgfmath

@Sean: and congrats! :)

@Shepherd It might just be log(\thisrowno{1})

@PauloCereda Thanks! :)

You could use \pgfmathlog{\thisrowno{1}}\pgfmathresult, but that seems too wordy

2:50 PM

@PauloCereda ?? /missed a joke somewhere xD

@PaulGessler @PauloCereda Now I'm lost, too XD

Where is @PaulMcCartney?

3

addplot+ table [x expr=\thisrowno{0}, y expr=\pgfmathlog{\thisrowno{1}}\pgfmathresult] {data.txt};

made my texmaker crash :(

but I guess i'll open a question. thanks for the help

Try using \fp_eval:n from expl3

@Shepherd That would be best at this point :)

user image

@PauloCereda ^^ that's not "nothing to see"! That's a duck! :-)

3:02 PM

@PauloCereda FWIW, you'd start to hate chocolate really soon, and that's something yo do no wo :D

@PaulGessler <3

@yo' Nope. :)

@Paul: you were trying to guess the company, and I tried Willy Wonka. :)

@PauloCereda How are you?

@PauloCereda I wish XD

@yo' Feeling better, thanks. :) I have a doctor appointment today.

@PauloCereda fingers crossed

3:12 PM

@SeanAllred Oh. :)

@yo' <3

@PauloCereda just don't forget, in the doctor's office, you are neither a duck nor a bee :)

latex-community.org/forum/…

@yo' woof. :)

Could be done using minipages. Any alternatives?

@Johannes_B \intertext?

3:21 PM

@yo' If i understood the op correctly, he wants to have a pseudo-twocolumn layout and center the first equation in the first column and the second equation in the second column. I think every line is independant.

@Johannes_B yeah, I think I get it. However, using minpages directly sounds wrong. A new align-like environment would work the best I think. Where is @egreg when you need him? :D

@yo' minipage would be extremely buggy.

@Johannes_B yep, think of the baseline preservation, for instance

@yo' Maybe @david has an idea?

@Johannes_B more reliable than @egreg anyway

3:34 PM

@Johannes_B See my reply to that. (vermiculus)

\intertext from amsmath seems to be the best bet.

@egreg: there's a rumour Mercedes Benz will sponsor Palmeiras in this season. :)

@yo' ...Which you already found out :)

(I just got that 'Oh, oh! I know this one!' rush :)

@SeanAllred Thanks. Was my first guess as well. But on second reading -> chat.stackexchange.com/transcript/message/19532774#19532774

@Johannes_B But isn't that exactly what is happening now?

Oh, whoops. I'm blind.

Some sort of \hfill <math> \hfill <math> \hfill

3:38 PM

Yeah that was my immediate thought, but that's really hacky.

No, quatsch, that wouldn't work as well.

@yo' Tutoring students. ;-)

@yo' Can you explain what's the topic?

30 mins ago, by Johannes_B

http://latex-community.org/forum/viewtopic.php?f=46&t=25496&view=unread#unread

4:22 PM

egreg to the rescue :)

Of the people who use emacs here, how many of you would like a way to print questions (to PDF with TeX, of course) from within emacs itself?

@Johannes_B -- you really want something more like \hfill <math> \hfill\hfill <math> \hfill if the idea is to have the equivalent of two distinct columns.

(Or perhaps this may be better as a poll or call-to-arms (for templates) on meta?)

@barbarabeeton and even better with \mathclap, but that's just bad...

@Johannes_B This can cause overlaps, but seems what the OP wants

\documentclass{article}
\usepackage{amsmath,environ}

\makeatletter
\NewEnviron{dblequation}
 {\expandafter\dbl@equation\BODY\@nil}
\def\dbl@equation#1&#2\@nil{%
  \begin{equation}
  \makebox[\dimexpr\displaywidth-6em]{%
    \makebox[.5\displaywidth]{$\displaystyle#1$}%
    \makebox[.5\displaywidth]{$\displaystyle#2$}%
  }
  \end{equation}%
}

\begin{document}

\noindent\smash{\vrule depth 8cm width 1pt}\hfill
\smash{\vrule depth 8cm width 1pt}\hfill
\smash{\vrule depth 8cm width 1pt}\hfill
\smash{\vrule depth 8cm width 1pt}\hfill

user image

@yo' oh, i wasn't suggesting it was a good idea, just that you need twice as much space in the middle as on the two ends.

4:35 PM

@barbarabeeton I got the purpose of your post, I just wanted to make sure that everybody knows it's not a good idea.

@egreg and now good luck with \begin{dblequation}\begin{cases} 1 & \text{if X} \\ 0 & \text{otherwise} & wh = \text{atever here}\end{dblequation} :)

@yo' Of course. But it's such a bad idea that I don't want to bother any more.

@egreg :)

@egreg @barbarabeeton @yo' I guess the op will see in the end that this is going to be bad.

@egreg @barbarabeeton @yo' If i remember correctly, he is typesetting exam papers.

@egreg Big thanks for taking a look at this. The screenshot looks great :-)

@Johannes_B So just centering the pair, with a double quad in between.

well, actually, something like this could work (untested)

\begin{equation}
\text{\clap{\hss\begin{tabularx}{\linewidth}{%
  @{}%
  >{\centering\arraybackslash$\displaystyle}X<{$}%
  @{\quad}%
  >{\centering\arraybackslash$\displaystyle}X<{$}%
  @{}%
  }
a = b & c = d
\end{tabularx}}}\end{equation}

@Joh ^^

Christian Hupfer

4:42 PM

Back....

oh no, it has of course the same bug as the code by @egreg.

@yo' Interesting.

Christian Hupfer

@yo' A paradox? ;-)

@yo' A different delimiter must be used, of course.

@yo' I thought of typesetting in a \halign and then retrieving the boxes, which would remove the bug, but it's a waste of time, considering the ugliness of the result.

@egreg yeah, that would be it. Or just \newcommnand\twoeq[2]{...}

4:57 PM

Operation international credit card succeeded. :)

Christian Hupfer

@PauloCereda: Please post the number ;-)

@yo' boo

@ChristianHupfer I need to wait for teh card. :)

Christian Hupfer

@PauloCereda: Please post the number ;-)

@ChristianHupfer 1234 5678 9098 7654

There are 3 numbers in the back, do you want them too? :)

5:01 PM

@DavidCarlisle ?

:19535079 ugly red box on that page, how did that get there?

@PauloCereda it would be really funny if this worked :D

please stop stalking

user image

Christian Hupfer

@PauloCereda: I am member of the NSA ... Number Storage Agency ... I want to have any number ;'-)

@yo' I'd be broken. :) TUG memberships for everybody! :)

5:02 PM

@ChristianHupfer lol

@ChristianHupfer LOL

@PauloCereda well, not for me, probably :-/

@yo' using a table has same problem you objected to originally about the use of two minipages, that it's hard to get top and bottom baseline skips right (you want a coffin not a minipage:-) that and I always cry when I see a tabularx with {XX} as I know how much work and iterations TeX goes through to work out half the text width if done that way.

@DavidCarlisle well, p{\dimexpr0.5\linewidth-0.5em} would do it I know. I just wanted to show the idea. The alignment shouldn't be a problem since both X and p align the baseline of the 1st line, don't they? The problem is that a & inside {cases} or similar is not "hidden"

@David so a trick similar to this one would have to be used:

3

A: Modify eqnarray to match that of amsmath's align

This is a robust solution based only on making & active. The code is well commented I hope: \documentclass{article} \usepackage{amsmath}% http://ctan.org/pkg/amsmath \begingroup % we group everything so that our \catcode changes are local, and we use \gdef everywhere \makeatletter \gdef\eqna@or...

@yo' a tabular (same as a minipage) is vertically centred by default . You can use [t] to align on top but then for tabular or for minipage it is hard to get the space right after the environment as you have no information about the bottom baseline.

5:09 PM

@David: new credit card. :)

@DavidCarlisle but what about \begin{tabular}{p{5cm}p{5cm}} $\displaystyle \frac12$ & $\displaystyle x$ \end{tabular} -- this should properly align the two equations, shouldn't it?

@yo' why is there a problem with & ? all ams alignments should nest correctly and use ifnum0={ groups to hide inner & from the outer alignment

@DavidCarlisle ah true

@yo' yes but you could do same more simply with two minipages. the problem is getting the space before and after the display correct as the boxes hide the inner structure (but you may not care about that and decide just always doing \belowdisplayskip is good enough

@DavidCarlisle yep, that's it

5:13 PM

@yo' which is why I said boo. I hadn't been following the chat as was doing work or something but at the start minipages were dismissed and when I came back I saw tabularx which gives same layout but more slowly:-)

@DavidCarlisle well, the point is that with tabular(x) you have the & meaning the right thingy :)

@cgnieder One for you ;-) latex-community.org/forum/…

@yo' why do you need & to separate the halves rather than \foo{left eqn}{right eqn}

@DavidCarlisle so that people have their favourite syntax? :D :P

@JosephWright: 50 GBP, any discount for a duck? :)

5:27 PM

@ChristianHupfer Can you take a look at dnb.de/DE/Wir/Projekte/Abgeschlossen/dissonlineTutor.html and tell me what you think?

Christian Hupfer

@Johannes_B: Right now? I am a little bit busy

@ChristianHupfer No, keine Eile ;-)

@yo' This solves the problem and also allows a label

\documentclass{article}
\usepackage{amsmath}

\makeatletter
\newenvironment{dblequation}
 {\setbox\z@=\vbox\bgroup
    \global\let\dbleq@label\@empty
    \def\label##1{\gdef\dbleq@label{##1}}%
    \halign\bgroup&$\displaystyle##$\cr}
 {\crcr\egroup\setbox0=\lastbox
    \dimen@=\displaywidth
    \advance\dimen@-\wd\z@
    \ifdim\dimen@<3em
      \@tempswatrue
    \else
      \@tempswafalse
    \fi
    \setbox\z@=\hbox{%
      \unhbox\z@
      \unskip\global\setbox\thr@@=\lastbox
      \unskip\global\setbox\@ne=\lastbox

Christian Hupfer

@Johannes_B: Thanks, I'll try in two hours

@egreg Wow, i hope the OP can appreciate your effort. Thanks in my name.

5:38 PM

@egreg say "wow"

@Johannes_B I leave as an exercise to add checks and remedies against overlapping.

Ayo

Could someone try to decode a rather cryptic error message?

The message is.... ! \OML/cmm/m/it/9 H[]\OT1/cmr/m/n/9 (\OML/cmm/m/it/9 X; \U/rsfs/m/n/9 H\OT1/cmr

@Anthony One on't cross beams gone owt askew on treadle. :)

@Joseph, @Christian: ^^

@Paulo.... what?

@Anthony of course it is rather more cryptic if you omit most of the text:-)

5:42 PM

Oh, sorry I didn't know what to look at I guess.

@Anthony I would guess it was preceded by "overfull box" or some such?

The document itself compiles- it just gives me that thing after compiling.

@Anthony It's a warm Monty Python reference. :) On a more serious note, it looks like some font info, but we need more than that. :)

@Anthony It's a warning then, not an error?

Oh, I see.

Christian Hupfer

@ThomasF.Sturm: Are you willing to place your name on the TUG membership competition? ;-)

5:43 PM

Are warnings bad?

user image

Christian Hupfer

@PauloCereda: It's a MP reference I don't recognize ;-)

That's what I'm looking at.

@ChristianHupfer The Spanish Inquisition sketch. :)

@Anthony if tex needs to show you a box in its internal form it shows you it one letter at a time (and one line at a time) so harder to read as you have it so that's an italic H followed by some other stuff, with teh font used for each character shown

Christian Hupfer

@PauloCereda: really?

5:45 PM

@Anthony \OML/cmm/m/it/9 H is an H in a font that is in OML (math letter) encoding cmm (computer modern math) family medium italic 9pt.

@DavidCarlisle I'm afraid I'm not very familiar with tex yet- boxes are part of the language right? In any case, is there a better way to view the warning that makes more sense?

@ChristianHupfer Yup, in the beginning. :) It's what Graham says to the lady. :)

Oh.

That's cool.

Christian Hupfer

@PauloCereda: Oh yes, that hardly understandable speech of him ;-) RIP Graham :-(

@Anthony ignore the internal form, just look at the output (you have some text sticking into the right margin) and the line number eg 530-532

5:49 PM

@Anthony try \documentclass[draft,.... -- the overfull boxes should get a black mark at the end of the line

@Anthony don't do \DeclareMathOperator\Mod{\ensuremath{Mod}} the \ensuremath is doing nothing useful there at all

Whoops just got back, let me read the responses.

@DavidCarlisle This wasn't my code- someone else made this and I was just trying to understand it.

(But why is it doing nothing useful?)

@Anthony it doesn't often do much useful, but there it does nothing at all as if you are not in math \Mod will generate an error because it is defined by \DeclareMathOperator so there is no way \ensuremath{Mod} will do anything different to Mod other than be slower

@yo' I'm trying that now,
@DavidCarlisle I also don't see the output you're talking about, you just mean the continuation of the message? For some reason maketex won't let me view the middle of it- I'll copy paste it to read it.

Also does ensuremath just take the name of the command, with out the backslash then?

@Anthony No I mean the document pdf has stuff that does not fit into its specified space (most likely the page width) so wil be sticking in to the margin. The log file (the bottom pane) is more useful than the middle "filtered" view eg that red error is about something on page 53

@Anthony no \ensuremath{mod} is the same as mod if you are in math and the same as $mod$ if you are not, but in that context inside \DeclareMathOperator you are always in math so there is no point in testing if you are in math or not

6:05 PM

I see.

Thank you all!

6:49 PM

@Johannes_B I'll see if I can take a look but actually I don't have much time... maybe on Friday

7:08 PM

@ChristianHupfer Saved my day: texwelt.de/wissen/fragen/11020/bin-ich-doch-schwanger

@cgnieder Wow, deleting that question was fast. (schwanger)

@DavidCarlisle goes directly to jail. — Gaussler 8 mins ago

2

7:36 PM

@Johannes_B Indeed! :) I had closed it almost immediately... seems like Stefan is online right now :)

@JosephWright @egreg response re mu I see:-)

7:54 PM

This seems like a bug, but I'm not sure whether it should be fixed in datatool or hyperref. I doubt whether tabularx is at fault here, and @DavidCarlisle would agree. ;) — Werner 16 secs ago

@DavidCarlisle: ^^^^

Hi all, can someone shed some light on expansion of default argument values for me please? tex.stackexchange.com/questions/223182/…

8:18 PM

@Werner quite right, although to be honest if you generate table rows via a loop inside the table then getting an empty 1 cell row at the end is the expected behaviour unless you are really careful, you would get the same from \relax so your loop code needs to clean up without generating a single non expandable token

8

Q: What is the mindset to take when using ConTeXt vs LaTeX?

Having recently gotten back into writing with LaTeX, I explored XeTeX and found advances in using modern fonts, explored LuaTeX and found a next-gen development underway, and explored ConTeXt and found ... a focus on formatting? To me, LaTeX and its variations have always been about separating ...

Anyone know what a 'good' answer here would look like?

@JosephWright there's an argument for closing as opinion based, although possibly it's answerable, I don't know.

Christian Hupfer

Good maen again

@DavidCarlisle Crystal clear, I'd say. ;-)

@egreg Indeed

@DavidCarlisle Did you read tug.org/TUGboat/tb35-3/tb111reviews-partha.pdf (in response to the team list mail)?

8:43 PM

34

Q: What are the incompatibilities of pdftex, xetex and luatex?

[Please correct me if my statements are wrong] TL;DR: pdftex, xetex and luatex exhibit different behaviour. What are these differences and why are they present. I know of three engines, which implement TeX and support native PDF output: pdftex, xetex and luatex. Unfortunately they are not com...

xetex pdftex luatex tex-core

@JosephWright yes:(

Erm, anyone know what the 500 bounty is really for? What more is there that can be added in the abstract?

@DavidCarlisle Sounds like an 'interesting' book

@JosephWright Ideal for people who follow the instructions on the TL list to install latex2.09

@DavidCarlisle Indeed

@DavidCarlisle That said, LaTeX2.09 does make interesting reading

@DavidCarlisle A sense of how Leslie thinks, for a start

@JosephWright I'm done with format changes I think, just needs package interface finalising

8:47 PM

@DavidCarlisle Cool

@DavidCarlisle Did you and Frank finish working out the details?

@JosephWright I spent a lot of time reading those comments and keeping or losing them to make first version of source2e..

@JosephWright Frank's a bit out of the loop this week, so we'll try again at weekend I suspect.

@DavidCarlisle Ah

@JosephWright I just had an interesting error I've not had in a while:

! Output loop---100 consecutive dead cycles.
\end@float ... \@tempdima \penalty \@floatpenalty
                                                  \else \vadjust {\penalty -...
l.17 \xxx\xxx\xxx

? h
I've concluded that your \output is awry; it never does a
\shipout, so I'm shipping \box255 out myself. Next time
increase \maxdeadcycles if you want me to be more patient!

@DavidCarlisle Oops

It's actually easy to hit now you can go \extrafloats{300} and have a stack of 300 unprocessed floats, if you dont set \maxdeadcycles to 305 you die..

@JosephWright I'm half wondering if \extrafloats ought to increase \maxdeadcycles

8:52 PM

@DavidCarlisle You know a lot more about this than me!

@DavidCarlisle Current morefloats packages doesn't do that, does it?

@JosephWright No but it allocates from the classic pool so can't usually get much more than 100 floats anyway and to get the error you have to have a run of floats where nothing gets output so they all stack up. and the default maxdeadcycles is 100

@DavidCarlisle Ah yes, makes sense

downvote for this, someone obviously doesn't think latex being complicated is a matter of opinion:-)

39

A: Why is LaTeX so complicated?

'complicated' - surely everyone can agree LaTeX deserves that adjective. No, I don't really agree. I think it is fairly simple to use and personally I have always found that a WYSIWYG style editing interface (on the few times I have tried to use one) is much harder to use and less intuitiv...

9:30 PM

@DavidCarlisle -- oh, don't feel so "special". every answer posted before jan 11 has at least one downvote. (the only time i think latex is complicated -- more often than i'd like -- is when i have to make it do something that the class i have to use wasn't designed to accommodate, and the only packages i know about either don't fill the bill, or aren't compatible. but authors are strange beasts, and have their own ideas of how something should be expressed, or what it should look like.)

@barbarabeeton Nah it had one of those but that got overturned this was a new one today:-)

@topskip Did you see the comment aimed at you in the bounty for tex.stackexchange.com/questions/222286/…?

@JosephWright I think you got the wrong user ping ;)

Good evening (or every other time of day you are experiencing now) everybody!
I got a question about positioning of nodes in tikz pictures. There is the syntax [above left=of someothernode], and I can specify a distance. However, the manual tells I may specify two different distances for each variable, but I cant figure out the syntax. I already tried [above left=(1:5) of someothernode], but it didn´t work.

Anybody got an idea?

@yo' That should be right

@yo' I've no idea what 'more detail' can be given, so I'm hoping someone else has!

9:44 PM

@JosephWright at this minute I'm running the latex test suite with xetex, If I post the resulting diff files, will I get the bounty?

@DavidCarlisle Perhaps :-)

@DavidCarlisle Then again, we already have lots of those for LuaTeX, so you/me/Frank/Will/Bruno can probably split it 100 each :-)

@DavidCarlisle Have to get Morten here too to register (as he seems to have reappeared somewhat)

Ah ok, I misunderstood something, sorry for that...

@JosephWright do you think we should do something about ^^ab v utf8 in log files, eg normalise all utf8 characters to use ^^^ ?

@DavidCarlisle Well we can't do all chars!

@DavidCarlisle We do have

for i = 1, 31, 1 do
  line = string.gsub (line, string.char (i), "^^" .. string.char (64 + i))
end

@JosephWright why not?

9:48 PM

@DavidCarlisle Rather 'what is the point': pdfTeX can only cover the 8-bit range so anything out of that must be XeTeX/LuaTeX and so should be fine 'left alone'

@JosephWright well OK to put it another way we could normalize the chars in the test suite which as you say means the 8bit ones test suite running so don't know how many tests it is yet, I suppose if it's only a dozen or so simpler just to have the separate tlg file but if its hundreds normalising away the diff woul dhelp

  ....\T1/cmr/m/n/10 ^^9f
  ....\T1/cmr/m/n/10 .
  ....\glue 4.44336 plus 4.99878 minus 0.37027
! ....\T1/cmr/m/n/10 ^^a0
  ....\T1/cmr/m/n/10 .
  ....\glue 4.44336 plus 4.99878 minus 0.37027
! ....\T1/cmr/m/n/10 ^^a1
  ....\T1/cmr/m/n/10 .
  ....\glue 4.44336 plus 4.99878 minus 0.37027
! ....\T1/cmr/m/n/10 ^^a2
  ....\T1/cmr/m/n/10 .
  ....\glue(\rightskip) 0.0
  ...\glue(\baselineskip) 1.49226
  ...\hbox(8.56377+1.94397)x345.0, glue set 0.1535
! ....\T1/cmr/m/n/10 ^^a3
  ....\T1/cmr/m/n/10 .
  ....\glue 4.44336 plus 4.99878 minus 0.37027

@DavidCarlisle I'm fine with the idea of normalising the 8-bit range ones

@JosephWright ^^^ :-)

@JosephWright although actually we current force pdftex to use ^^ form with that ascii.tcx enc file thing, so we could force it to use utf8 output which might make more sense

I am back, the doctor said I'm fine! :)

@JosephWright hey it's not bad, currently there are 18 failures with etex and 50 with xetex, mostly due to ^^ stuff

9:55 PM

@PauloCereda Could you hear him saying it or did he write a note?

@JosephWright will checkengines = {"luatex"} work as well?

@PauloCereda never believe what they tell you

okay, I found the answer. I have to link them with and

@egreg I did hear him but he asked an audiometric test just to be sure. :)

@DavidCarlisle Oh no! :)

@DavidCarlisle Should do: I generalised the format building

@JosephWright OK, lets see...

9:57 PM

@Joseph: I'm very close to a membership. :)

@DavidCarlisle If it is workable, yes

@DavidCarlisle On the other hand, if we build it into the script that means we don't have to worry about it again in other cases (L3, ...)

@PauloCereda yay, party time! :D yo!

@yo' yo! :)

@JosephWright true, anyway luatex seems to be running....

@DavidCarlisle Cool

10:01 PM

@JosephWright ah ho what's that character doing?

--- ../build/test/tlb0376.luatex.log	2015-01-14 21:59:31.660386100 +0000
***************
*** 17,23 ****
  %%   Right brace   \}     Tilde         \~}
  %%
  LaTeX Font Info:    External font `ecrm1000 at10.0pt' loaded as
! (Font)              T1/cmr/m/n/10  on input line 29.
  LaTeX Font Info:    Switching to \T1/cmr/m/n/10  on input line 29.
  LaTeX Font Info:    Switching to \OT1/cmr/m/n/10  on input line 31.
  LaTeX Font Info:    Switching to \OT1/cmr/m/n/10  on input line 31.
--- 17,23 ----

@PauloCereda I too had an audiometric test a while ago; for some minutes I was asking myself when they would start. They had already started. ;-)

@JosephWright U+00ff in front of the font name?

@DavidCarlisle Probably one of those out-by-one things LuaTeX seems to specialise in!

@egreg uh-oh! :)

@JosephWright more emails to luatex list? :-)

10:03 PM

@DavidCarlisle I guess so :-(

@JosephWright you are probably used to seeing it in the l3 tests, but seems lots of gratuitous differences eg:

! l.27 ...how\csname S@\strip@pt\@tempdima\endcsname

! l.27 ...\show\csname S@\strip@pt\@tempdima\endcsname

@DavidCarlisle lost vs lots, lol

@DavidCarlisle Yes: this is a 'known issue' I've raised with them

@yo' ??

@DavidCarlisle Some of this is fixed in their SVN

10:06 PM

@DavidCarlisle nothing now, you have corrected it

@yo' me?

@PauloCereda The flu had affected my left vestibulocochlear nerve, so I could hear nothing from the left ear and couldn't walk straight nor standing for more than a few seconds.

@DavidCarlisle Isn't that the \csname bug?

@DavidCarlisle chat.stackexchange.com/messages/19541999/history

@JosephWright possibly: I'm just watching the diffs go by, as it runs at the moment

@yo' cheat

10:08 PM

@DavidCarlisle LOL

@egreg Oh my, I'm sorry. I was about to faint on saturday if I hadn't look for a hospital.

@JosephWright test failures:
50 xetex
18 etex
127 luatex

@DavidCarlisle Sounds about right pending LuaTeX fixes

@PauloCereda you scare me, pal!

@PauloCereda Ear problems have that effect. But I recovered from the problem.

10:14 PM

@yo' I'm fine now. :)

@egreg Indeed!

@JosephWright yes it's a lot better than I feared, certainly a manageable number of engine differences.

@PauloCereda Fine like this? :D dailymotion.com/video/…

@yo' poor Tom! :)

@PauloCereda you're mentioned at ~03:15 of that vid ;)

« first day (1536 days earlier) ← previous day next day → last day (3380 days later) »

Transcript for

Jan '1514

$TeX - LaTeX$ TeX, LaTeX and Friends

\begin{chat}

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

$TeX - LaTeX$

site design / logo © 2024 Stack Exchange Inc; legal

mobile