« first day (1292 days earlier)      last day (3644 days later) » 

@JosephWright can I look too:-)
 
@DavidCarlisle Of course: I'm just fixing a few issues on Windows
@DavidCarlisle Mail to team list in a second
 
not a dup:
This is not a duplicate of the linked question that shows how to get 1.1, 1.2, 1.3, 2.1, 2.2 or 1,2,3,4,5 this question asks for 1,2,3,1,2 — David Carlisle 21 secs ago
 
8:13 AM
@DavidCarlisle See team e-mail with notes!
 
@JosephWright got it thanks
 
8:49 AM
@JosephWright in l3packages xtemplate004 was the one expected to fail make.lua check? and make.lua ctan seemed to run without error resulting in total bytes=5597008, compressed=5342917 -> 5% savings
 
@DavidCarlisle Yes, xtempalte004 should fail: unlike the 2e tests we can't make the simple 'all tests must pass before a CTAN release' assumption
@DavidCarlisle That compression is for the final stage?
@DavidCarlisle I've gone with a two-stage zip process to keep Frank happy about line ends and PDFs
 
@JosephWright yes I just meant it got to the final stage
 
@DavidCarlisle Cool
@DavidCarlisle So the script works with Cygwin :-)
 
@JosephWright so only tested ctan and check, in l3packages, but basically it seems to work on cygwin
 
@DavidCarlisle Those are the two most complex targets, so if they work all should be fine
@DavidCarlisle I've not set up outside l3packages just yet
@DavidCarlisle Need Frank's feedback (and @PauloCereda's on the Lua!)
 
8:53 AM
@JosephWright we could check in the tlg so that it passes, if it passes well enough to be released, why have a check that fails?
 
@DavidCarlisle It's a reminder there are things about the templates that are not right
@DavidCarlisle I suspect some of this really needs the LDB to get properly addressed: perhaps a little later this year :-)
 
@JosephWright yes but we know that and it's better to check in a base result so if we do ever get round to changing xtemplate we see the resulting diffs against the current base
 
@DavidCarlisle Failing test file written by Frank, not me!
@DavidCarlisle Take it up with FMi!
 
@JosephWright it probably worked when he wrote the test so we'll blame you for breaking the code:-)
 
@DavidCarlisle All I have to know, for CTAN releases, is which tests fail deliberately
@DavidCarlisle Probably true
@DavidCarlisle One for you:
2
Q: Are there external graphics files with natural depth?

AlexGThe \includegraphics command from the graphicx package provides options height and totalheight for resizing embedded graphics. As long as the graphics file doesn't possess a natural depth, it doesn't make a difference whether the one or the other option is used. Therefore I wonder why totalheig...

 
9:03 AM
@JosephWright I was just answering it!
 
9:35 AM
@JosephWright speaking of which, seeing that in PDF would ease diagram pre-compilation so much!
 
9:46 AM
@tohecz seeing what?
 
@DavidCarlisle PDFs having depth
 
Hello
 
@tohecz would require a change to pdf or "inventive" interpretation of fields such as MediaBox though wouldn't it? although perhaps you could set ArtBox to be the part above the baseline, If the info is in the PDF, pdftex could probably use it... prepressure.com/pdf/basics/page-boxes
 
Hi @DavidCarlisle , yesterday you were right about text in cercle
 
@DavidCarlisle but I would use it incorportated into tikz-standalone in order to be able to use TikZ figures easily in places like arXiv or like many journals.
 
10:02 AM
@Alji I know:-)
 
hhhh
@DavidCarlisle don't be so sure :D
 
@Alji I think I was probably the last person to edit the source of those commands in the latex format
 
seriously ?
 
@Alji yes
 
@DavidCarlisle Pre-dates the SVN or can we check?
 
10:04 AM
@DavidCarlisle it is an honor to talk to you @DavidCarlisle
 
@JosephWright you can check (and I was wrong:Frank's been there since:-) I was thinking of
1995-10-27 David Carlisle <latex-bugs@latex-project.org>

	* ltpictur.dtx: Initialise \@wholewidth in autoload kernel.
But Frank has
2003-12-30 Frank Mittelbach <latex-bugs@latex-project.org>

	* ltpictur.dtx: issue warning if circle size can't be met
@Alji :-)
 
I have a question @DavidCarlisle it is related to what you said yesterday
I have this snippet of latex : `\setlength{\unitlength}{1mm}
\begin{picture}(10,120)
\thicklines
\put(0,0){\includegraphics{word07}}
\put(120,120){\circle{10}{A}}
\put(130,112){\circle{10}{B}}
\put(120,95){\circle{10}{C}}
\put(170,50){\circle{10}{D}}
\put(40,50){\circle{10}{E}}
\end{picture}`
 
@JosephWright I think 1995 predates svn as a system not just our use of it
 
I would like to correct it, by putting A,B,C ... in circles correctly
 
@DavidCarlisle Is/was there a CVS before it for the kernel? If so, did the data get lost in the move from CVS to SVN?
 
10:09 AM
@Alji as I said It's easier to use \makebox(5,5){A} then the A is centred at the specified coordinate rather than its baseline being at that coordinate
 
@DavidCarlisle you mean I put it like this \put(120,120){\circle{10}} \makebox(5,5){1} ?
 
@JosephWright there was CVS although that predates CVS and it was RCS. I think Rainer gave up doing the cvs to svn conversion and simply checked in the top level files when creating latex2e-public, I'm sure he has the logs though. the changes file is though a better indication of who did what as we were using rcs synchronised via emailed pactches so each of us had our own master checkin and all activity would be logged as the local user.
 
@DavidCarlisle I am testing right now ...
 
@Alji \put(120,120){\circle{10}} \put(120,120){\makebox(5,5){1}}
 
@DavidCarlisle it worked with makebox(0,0)
 
10:13 AM
@DavidCarlisle :-)
 
@dais there another way to
 
@DavidCarlisle We have basically a full house of version control systems if we count the GitHub mirror
 
@DavidCarlisle is there another way to make cercles lines more thicker ?
 
@Alji yes the size doesn't really matter in this context the important thing is that it makes the origin in the centre not at the baseline
 
@DavidCarlisle How to do that ?
 
10:16 AM
@Alji if you use pict2e package the lines and circles are drawn using graphics primitives (pdf or postscript) and so if you go \linethickness{5pt} they will be very thick by default it uses fonts so limited to two widths. Or of course you could do what the youngsters around here will tell you to do and load half a million lines of tikz code which implements an entire graphics subsystem that would allow you to draw all kinds of fancy shapes with gradient colour fills etc
@Alji I mean that's what \makebox(,) {A} does
 
@DavidCarlisle So Email -> RCS -> CVS -> SVN at the moment :-)
 
@DavidCarlisle I don't want to use Tikz I am beginner with LaTeX
 
@JosephWright Rainer had implemented quite a sophisticated mail server hooked to RCS so al checkins automatically mailed patches to everyone and then we (got emacs to) apply the patch to our local copies and checked then in to local RCS. so basically DVCS before its time (and before UK was on internet:-)
 
@DavidCarlisle Ah right: TeX ahead of time as usual
 
@DavidCarlisle until now Picture package work fine for me , I just need to make lines thicker,
 
10:21 AM
@Alji It doesn't really matter how complicated it is (the code to picture mode is pretty horrendous:-) if you asked on site how do I circle a label with tikz you would get an answer in seconds saying you use \usepackage{tikz} ... \node[type=circle]{A} (I just made that up, don't try it:-) and then you just use it,
 
@DavidCarlisle \linethickness{} didn't work with cercles, I have just checked
 
@Alji No it won't only with horizontal and vertical lines, unless you add \usepackage{pict2e} as by default circles and sloping lines are lots of characters from fonts patched together
 
@DavidCarlisle another alternative, is there a way to make A,B,C ... more thicker :)
 
@Alji \bfseries that is just normal text so any latex font commands work
 
@DavidCarlisle it is like you are talking chinese :D I don't understand a word from what you are saying :D :D
@DavidCarlisle what is bfseries ?
 
10:25 AM
@Alji which is why it's best to ask questions requiring code as real questions on the main site, formatting code here is a pain and it's better for chatting
@Alji It's the standard latex command to get bold text
 
@DavidCarlisle you are right :)
@DavidCarlisle thank you any way for your precious help
 
@Alji ` hello {\bfseries this is bold} this is not bold`
 
@DavidCarlisle I use \textbf{this is bold } :D
 
@Alji yes same thing more or less you could use that form in the picture as well
 
excellent
 
10:45 AM
good bye
 
Does LaTex have an established name yet for the "long division symbol," or sometimes the division bracket?
 
@skullpatrol you mean the layout with a | or ) at the left and an overline?
 
Yes.
 
@skullpatrol no:-)
 
@DavidCarlisle OK, thanks for responding :-)
 
10:56 AM
@skullpatrol (I ask as the long division notation is highly dependent on location and culture) but in the standard and ams packages don't really support it other than using a array with \cline everywhere. There are I think contrib packages for long division (of numbers and polynomials)
 
@JosephWright: working on it. :)
 
@PauloCereda Cool
 
11:19 AM
@DavidCarlisle Is there an accepted name for the "traditional display" notation for division of a horizontal fraction bar?
$\frac{a}{b}$
I mean the horizontal bar itself?
 
11:31 AM
vinculum
 
@skullpatrol not as far as I know, we got into all kinds of cultural issues trying to name these things in mathml, look at the names we ended up with:-) w3.org/Math/Group/draft-spec/chapter3.html#presm.mlongdiv.ex
@skullpatrol yes but as far as I know that's just latin for overline not specifically the long div notation
 
@DavidCarlisle Thanks :D
 
11:49 AM
I do like taking ticks off @egreg, even if it takes three years and a new latex release to do it
 
12:04 PM
@DavidCarlisle But does the computation really work?
 
@egreg You might notice an odd edit on one of your answers!
 
12:20 PM
@JosephWright I noticed; good job, thanks.
 
@egreg Not really me: altering the history requires staff powers!
@egreg, @DavidCarlisle I see we all went for the plain TeX question :-)
 
@egreg well it works as well as can be expected I think, it arbitrarily chops some lengths to maxdimen but that's no worse that the existing arbitrary limits and the end result is a typeset tab;e that's probably too big but at least you can see it rather than getting a primitive arithmetic error
@JosephWright ohh egreg as well now:-)
 
@DavidCarlisle So the change in the tick is completely unmerited. ;-)
@DavidCarlisle With the simplest solution. ;-)
 
@egreg but I got a Necromancer badge for it as well so it must be good
@egreg only simpler because you thought you might get away with "use eplain" instead of the logically similar (and practically more sensible) "use latex"
 
12:37 PM
@DavidCarlisle Or as we had yesterday 'use LamsTeX'
@egreg I was more interested in the general point: I'd expect heavy plain users to be 'up' on this
 
@JosephWright I said (perhaps not explicitly) “don't use LamsTeX”. :)
 
@egreg Yes, I noticed that too
 
@JosephWright I expect Plain TeX users not to know about eplain.
 
@egreg Possible in the early 1990s it would have been different
@egreg :-)
 
@JosephWright I've never met any LamsTeX user. I think they sold just a few copies of the manual. But I did buy MathTime from them.
 
12:42 PM
@egreg I meant more the idea of integrating the AMS stuff into LaTeX: probably impossible in 1994 given memory situation
 
@JosephWright I threw away my LamsTeX book about two years ago. I didn't know it was such a rare item :-(
 
@JosephWright AMSTeX was much lighter than LaTeX, as it was just oriented to mathematics and didn't want to do font selections. LamsTeX didn't add much and it had loading on demand. The big error was making it commercial.
 
@egreg I'd agree on that last point
 
@JosephWright AMSTeX's manual was also nonfree, but AMS made the macros available long before Spivak put LamsTeX on CTAN.
 
@egreg Understood
@egreg I was thinking more that looked at from today's POV, adding the additional maths stuff from the AMS bundle to LaTeX would be reasonable
One for LaTeX3, I guess
 
12:52 PM
@JosephWright The core of amsmath must be in LaTeX3.
5
@JosephWright By the way, environ should find its way into xparse.
 
@egreg Oh yes: Frank's position is that we need at least all of The LaTeX Companion in terms of concepts
@egreg Bit more complex, but I understand what you mean
 
@egreg I don't think this should be a separate package at all.
 
@StephanLehmke That's what he means, I think
@StephanLehmke 'Functionality equivalent to ...'
 
@StephanLehmke Yes: there should be no need of \usepackage{amsmath}.
 
@egreg Nor lots of other things (which is why I ended up on the team in the first place)
 
12:55 PM
@JosephWright did we? missed that one (I think I installed that once)
 
@egreg Will there be licensing issues (with fonts, or whatever)?
 
2
Q: How to install LamsTeX on TeX Live (Linux)?

MatteoLamsTeX macro package is on CTAN but is not installed in my TeX Live. How can I install it? Where should I put the files? Note that, in addition, LamsTeX uses the index-processor and dvipaste programs. So they should also be installed.

@StephanLehmke Fonts might be a slightly different issue: I was thinking macro code
@StephanLehmke They are all in TL
 
@JosephWright Well but "absorbing" the stuff completely so that the user doesn't see the letters "AMS" when using it might be a different issue.
 
@JosephWright funny: I just noticed the other day while looking round my local svn for a place to sore some longtable experiments that I had a latex interface to dvipaste, I had forgotten all about it (the program and that I'd written a latex package)
 
@StephanLehmke Work is LPPL and in any case it's the functionality we want: a lot of the code might well be different
@DavidCarlisle :-)
 
1:03 PM
What do you think about tex.stackexchange.com/q/177362/13304 ?
 
@JosephWright which work amsmath or lamstex?
 
@DavidCarlisle amstext was the one I checked
@DavidCarlisle I'm pretty sure all of the AMS macros
 
@JosephWright oh ok that's not only LPPL it's also mostly by Frank and Rainer the only reason it didn't go into 2e was space reasons, that's why it was made part of latex/required
 
@DavidCarlisle CTAN list LamsTeX as Public Domain
@DavidCarlisle main amsmath file is the same: LPPL1.3c or later, 'maintained' status
 
@JosephWright yep
 
1:17 PM
@ClaudioFiandrino You're thinking bad. ;-)
@ClaudioFiandrino And yes, I definitely don't like that kind of usage for \ensuremath. ;-)
Did anybody notice that pdflatex in the 2014 version always throws a warning when the PNG example images in the mwe suite are used?
libpng warning: iCCP: profile 'icc': 0h: PCS illuminant is not D50
libpng warning: iCCP: profile 'icc': 'RGB ': RGB color space not permitted on grayscale PNG
 
@egreg Hadn't checked but you are right
 
1:32 PM
@JosephWright It doesn't do it for every PNG, though.
 
@egreg Issue with the images then?
 
@JosephWright I guess problems when the conversion was made.
 
@egreg Yes: get @martinscharrer to take a look?
 
@JosephWright I get warnings also when I add PNG created with screen shots (with Command-Shift-4).
libpng warning: iCCP: Not recognizing known sRGB profile that has been edited
 
@egreg OK, so perhaps a libpng issue?
 
1:36 PM
@JosephWright Possibly.
Maybe libpng 1.6.10 has more checks. As far as I know, it's the most recent version.
 
@egreg :) As I told you, I was expecting a comment from you. Telling truth, I agree: coming back to the source code later on, it is far way better not having such macros. However, everybody has right to decide how to suicide, that's why my answer ;)
 
@ClaudioFiandrino Tutti possono darsi una o più martellate sui … ;-)
@ClaudioFiandrino But I think I have to advise them not to. ;-)
 
@egreg Absolutely :)
but it's better if they realize by their own why that's wrong... I've learned a lot in that way ;)
 
2:31 PM
Is "place a sheet of paper over the projector's lens and slide it from left to right slowly uncovering the lens" a valid answer to tex.stackexchange.com/q/177383/3954?
3
 
2:46 PM
@GonzaloMedina This brings back fond memories of the animation tricks people used to do with tiny transparency snippets on overhead projectors :-)
 
 
1 hour later…
3:48 PM
@egreg, @DavidCarlisle Could one (or both) of you make comments into answers for tex.stackexchange.com/questions/176625/…?
OP has asked via a flag that this happens
 
@JosephWright me me :-)
@JosephWright done
 
@cgnieder Is \iupac in the chemmacros manual: I'm failing to find it
 
4:16 PM
0
Q: Compile database inputs

user51789I want to write a Latex-API on a Server! It is possible to transmit all the data of document (picture, .tex and so on) through an string to the latexcompiler (pdflatex or lualatex). Or shall the data represent by real file-objects?

Anyone understand this?
 
@JosephWright I think he wants to send a single stream with images and text to latex, so no in general
 
@JosephWright it is, in section 8 Nomenclature, ...
 
@cgnieder Ah, I see the issue: texdoc chemmacros opens the chemformula PDF by default with TL2014, so I have to texdoc -l chemmacros
 
@JosephWright Same in TL2013, for that matter.
 
@JosephWright also in TL13
 
4:27 PM
@TorbjørnT. I suspected so but didn't test :-)
 
@JosephWright He could encode stuff in, say, Base-64 and reconstruct it in the other side. But how is he supposed to write a webservice if he doesn't know the inners of the stuff itself? :)
 
@JosephWright I always wanted to contact someone who can change this but just as often forgot again to do it ...
 
@cgnieder E-mail the TL list
 
@PauloCereda some people write web services that just push everything to wolfram alpha and hope something good happens without really inspecting the content of the request
 
@DavidCarlisle So true.
 
4:36 PM
@cgnieder code freeze for tl2014 is today with package updates from ctan allowed for a few more days so now would be a good time to mail:-) (<tex-live@tug.org>)
 
@JosephWright Is “The question is stupid” a valid answer? ;-)
 
@DavidCarlisle Yup
@PauloCereda How does my Lua look?
 
@JosephWright It looks good from my side. :) I'm monitoring the code.
 
@DavidCarlisle (@JosephWright) Good to know, thanks :) email sent
 
4:57 PM
@cgnieder and arrived:-)
 
5:48 PM
Happy birthday Maria Gaetana Agnesi
 
6:22 PM
Does anybody know where can I download a free test .avi video?
 
@MoneyOrientedProgrammer Thanks.
 
@JosephWright expl3: You are defining l3kernel.strcmp in \directlue. But this definition is lost, if expl3 is put into a format. Therefore the definition needs to be repeated in \everyjob=\expandafter{\the\everyjob\directlua{...}}
\newcommand*{\FixExplThreeLua}{%
  \directlua{%
    if not l3kernel or not l3kernel.strcmp then %
      l3kernel = { }%
      function l3kernel.strcmp (A, B)%
        if A == B then %
          tex.write ("0")%
        elseif A < B then %
          tex.write ("-1")%
        else %
          tex.write ("1")%
        end %
      end %
    end%
  }%
}
\everyjob=\expandafter{\the\everyjob\FixExplThreeLua}%
 
@HeikoOberdiek In the source
%<*initex>
      \directlua{\tempa}
      \global\everyjob\expandafter
        {%
          \expandafter\luatex_directlua:D\expandafter{\tempa}%
        }
%</initex>
 
I am using the previous code for the temporary formats used by latex-tds.
 
6:36 PM
@HeikoOberdiek Ah, are you talking about people doing silly things like making their own format?
@HeikoOberdiek We've never supported that, that I know of
 
@JosephWright It's a simple \RequirePackage{expl} that goes into the format (with other packages).
@JosephWright Own formats are useful, if many documents with the same base set of packages are to be build.
 
@HeikoOberdiek I guess I have a slightly different view: the LaTeX2e code was never intended to support this
@HeikoOberdiek I guess we can add something for this: while it's not really 'supported' there is not actually a cost to allowing for it
 
@JosephWright welcome to my world, users have never used latex code in the way it was intended to be supported:-)
 
@JosephWright Of course, you can put LaTeX packages into a format, if they do not depend on a document class. The only problem is \directlua, because the Lua code does not get dumped into the format.
 
@HeikoOberdiek I've raised the latter point on the LuaTeX list before: no change yet :-(
 
6:44 PM
@JosephWright You can even put a whole preamble of a LaTeX2e document into a format. There are even some tools that support this (mylatexformat, ...). I think, expl3 should survive this, because it is also loaded by other packages. Then these packages would inherit the problems of expl3.
 
@HeikoOberdiek yes mylatexformat is an update on mylatex which does this and is written by a team member:-)
 
@HeikoOberdiek Like I say, we can add something for this
@DavidCarlisle Doesn't mean that it's supported :-)
 
@JosephWright Thanks.
 
@JosephWright always use everyjob or test for initex and just do it in that case?
@JosephWright we rely on you to support anything written by any team member.
 
@HeikoOberdiek I guess you want it to make the TL2014 DVD :-)
 
6:49 PM
@DavidCarlisle In pdftexcmds I am testing the lua variable status.ini_version.
 
@DavidCarlisle IniTeX is already covered as that would be format mode, which at the moment I think only I ever use :-)
 
@JosephWright Too late :-) I need it now, thus I have written a patch for latex-tds (it is only internally used).
 
@HeikoOberdiek What do you use it for there?
 
@JosephWright Creating a format that is used to accelerate the generation of the documentation.
 
@HeikoOberdiek I meant 'I assume you want me to make sure that we upload to CTAN asap so the DVD version has this fix in it, so other people can use it'
@HeikoOberdiek Risky
 
6:54 PM
@JosephWright If you want, that other people can put expl in a format file, then yes. For me personally it is not urgent, because I am using a workaround.
 
@JosephWright no I mean Heiko's use where an end user loads the package into initex
Is it just me or do other people think the spacing in the output here is hmm sub-optimal?
\documentclass{article}
\usepackage{longtable}
\begin{document}
\begin{longtable}{|ll|}
zzz& zz\\zzz& zz\\\caption{zz}\\aaa&aa\end{longtable}
\end{document}
 
@DavidCarlisle Don't follow: if we add an \everyjob repeat in generic mode then IniTeX or not doesn't matter, does it?
 
@JosephWright if it's just this then probably not, but if there were a lot of code you might want to not do it if it isn't initex (as \everyjob is dead code in that case)
 
You want the someone quotes the documentation of package `caption`? Here, the start of description of option `position`: "The `\caption` command offered by LaTeX has a design flaw: The command does not
know if it stands on the beginning of the figure or table, or at the end. Therefore it does
not know where to put the space separating the caption from the content of the figure
or table. While the standard implementation always puts the space above the caption
in floating environments (and inconsistently below the caption in longtables), the
 
@HeikoOberdiek yes but the LT version seems particularly bad, I was just experimenting with some possible extensions and thought I'd broken the caption code, but then saw it was like that already:-) (It has had hooks for caption package since forever though ) oh well.
 
7:08 PM
@HeikoOberdiek, @DavidCarlisle Proposed new version of code in l3bootstrap:
  \expandafter\ifx\csname directlua\endcsname\relax
  \else
    \ifnum\luatexversion<36 %
    \else
      \catcode`\_=11 %
      \catcode`\:=11 %
      \def\tempa
        {%
          l3kernel = { }
          function l3kernel.strcmp (A, B)
            if A == B then
              tex.write ("0")
            elseif A < B then
              tex.write ("-1")
            else
              tex.write ("1")
            end
          end
        }
      \directlua{\tempa}
      \ifnum 0%
        \directlua
Obviously hijacking @HeikoOberdiek's test for IniTeX
 
@JosephWright There is an \expandafter missing between \the and \everyjob.
 
@HeikoOberdiek So there is
 
@Joseph: hey, there's is a missing \expandthingy (I have no idea of what I'm talking right now). :)
 
@PauloCereda You stick to the Lua :-)
 
@JosephWright sir yes sir. :)
 
7:15 PM
How to make this symbol i.imgur.com/R68qs7z.png?
 
@HeikoOberdiek no other issues?
 
@JosephWright If you want to be defensive, then the table l3kernel in the code for \everyjob should only be cleared, if it does not yet exists.
 
@HeikoOberdiek That one I suspect we can live with :-)
@PauloCereda A Lua question: I'm trying to do something like foo = "text " if bar then .. "more text" end, but Lua doesn't like it. Is there an easy way to do this other than if bar then morefoo = "more text" else bar = "" end; foo = "text " .. morefoo?
@HeikoOberdiek On the other hand, it costs nothing so I'll add it
@HeikoOberdiek Committed to SVN
 
7:32 PM
@JosephWright Untested: foo = "text " if bar then foo .. "more text"
 
@PauloCereda Doesn't help in my real case :-(
 
@JosephWright Ah wait then. :)
 
@JosephWright: Quick question: Any sense in addressing (possible) robo-reviews?
 
@PauloCereda Real case is not an assignment but in the arg of a function
@Werner In what sense?
 
@JosephWright If we see something that seems robo-like, is there anything we should/could do about it?
 
7:34 PM
@Werner The review system is meant to be checks-and-balances, but you can always ping the mod team :-)
 
Longer/clearer form:

foo = "text"
if bar then
foo = foo .. "more text"
end
 
@JosephWright I know really that there's not much value in addressing it, since it could be very subjective - the reviewing process in general.
 
@HeikoOberdiek Yes, that's my conclusion but it's a bit of a pain as you can't then inline into construction of an argument to a function
 
@JosephWright Short form: foo = "text" .. (bar and "more text" or "")
 
This is a suspicious one to me when compared to the previous one in terms of the time-line.
 
7:35 PM
@HeikoOberdiek Ah
 
@HeikoOberdiek Damn, you beat me to it. :)
 
Question, anyone know how to center `\multicolumn headers properly?
At this point I will take "how to shift a cell in a table left by a defined number of cm"
 
...I rarely get to review anything in 4 seconds. But then again, this could just be me.
 
@PauloCereda :-)
 
@HeikoOberdiek Great, that solves my issues
@PauloCereda, @HeikoOberdiek Next issue: how do I get Lua to return an errorlevel? We have a 'chain' of scripts, and they call each other via os.execute so I want to be able to set the errorlevel on exit.
 
7:41 PM
 
@JosephWright os.exit(<int value>)
 
@JosephWright The return value of os.execute should be the error level.
 
@PauloCereda Ah, right
@HeikoOberdiek I'd found that part: it was the 'throw' that I was missing
 
bar = true
foo = "foo"
foo = (function(x) if bar then return foo .. "more text" else return x end end)(foo)
print(foo)
@Heiko: ^^ I'm definitely crazy. :)
 
I tried to use \stackrel{\sim}{\rightarrow} to obtain i.imgur.com/R68qs7z.png, but it's not looking good.
 
8:05 PM
2
Q: How to make a document with light colors printer friendly?

Antoine LecailleI have a document with some text in light colors. It's looking good on screen but I need to print it. Unfortunately, once printed on a black and white laser printer, the light blue is no longer legible due to dithering. How can I render a document in which the text is going to be correct once pr...

Is "Don't use colours at all" a valid answer? :)
 
@crasic ?
@crasic oh you mean the third header that's wider that the column it spans? you just need to make the spanned columsn wider (if you are using tabular* you can use shrinkable tabskip glue to automate it
@crasic vvvvv
4
A: Table column widths disproportionate due to multicolumn cell being too long

David CarlisleIt's a feature of the \halign primitive: \documentclass{article} \usepackage{tabularx} \begin{document} \centering \parskip2\baselineskip \begin{tabular}{*5{c}} \multicolumn{5}{c}{This is a piece of text. The longer it is, the longer the last column.} \\ 1 & 2 & 3 & 4 & 5 \\ \hline \end{tab...

 
@david
thanks, I will try that
Also the cmidrules are extending too far on one side
cmidline rather
 
@crasic can't tell from the picture but aren't they just stretched by the heading?
 
yes that is true
there is some alternate tabular package that lets you move around the whitespace
but I can't remember the name
 
@crasic so fix the headers and the rules will fix themsleves
 
8:14 PM
Right, I wasn't using a single multicolumn header, but rather individual ones for each header
 
@crasic yes so i assumed but the last one is wide r than the columns it is spanning isn't it? (really this sort of question is much easier if you make a small example and post it on site)
 
only because C{whatever-cm} is set just wide enough to keep it to 3 lines
 
@crasic if you make a multicolumn wider than the columns it spans all the extra width goes into the last column, as seen in your image: don't do that:-)
 
Heh
Its not my thesis, I'm just the middleman here ;)
So switching to a single \multicolumn and then using the fix from that question to get the last row fine is the way to go?
or last column rather
 
@crasic so if you want the heading that wide force the columns under it to be that wide (by adding \hspace{anything} if necessary so all the columns get wider not just the last
 
8:17 PM
Ah
 
@crasic no you want one multicolumn per heading or the headings won't sit over the right columns
 
@JosephWright: Do you find this weird?
 
@Werner Yes, I guess so
 
There are plenty more examples here:
14
Q: How is this user able to review posts so long after they were completed originally?

WernerThis is a revisit of Are certain users favored the opportunity to review posts? that seems to have grown stale without any solid answer. Just recently I was baffled by a user's ability to review a "First Post" almost 6 hours after it had been reviewed previously: In my experience, certain rev...

Still unsolved.
 
@david hspace goes where, in every cell below or in the same cell as header?
it seems to just move stuff around
 
8:24 PM
It's one of those things were I'd post a bounty for a question (to get more attention to it), but I don't know what a bounty would do in terms of a bug in the software...
...they'll get to it when they get to it I guess; and it doesn't seem important.
 
@crasic just once per column so for example in the last case the min^{-1} column is too narrow but \makebox[half the width you used for the multicol header]{$min^{-1}$} would make that columns wider
 
8:39 PM
worked beautifully
\hspace was the command I couldn't describe
 
@crasic :-)
 
 
1 hour later…
10:06 PM
Quick check: I get a "This item is not reviewable" message. What do you get:
 
According to the OP¿s last comment, the problem here tex.stackexchange.com/q/177451/3954 was just an editor issue, so I guess it can be safely closed? If so, what's the standard reason in these cases?
@Werner The same as you.
 
@GonzaloMedina Weird.
 
@Werner I've also noticed some "oddities" with the reviewing process.
 
10:29 PM
@Werner I reviewed it first. There are two “delete” votes and four “recommend deletion”.
 
@egreg According to what I see in the review history, you reviewed it last.
 
@GonzaloMedina Yes...
 
@GonzaloMedina Oh, right.
 
@egreg ...my question revolves around the fact that I was unable to review it 20 minutes ago, although the history does show that reviewing was going on unperturbed.
 
@Werner I can't say more. If I click on the link, I get a normal review history.
 
10:34 PM
@Werner I couldn't either, and I tried before egreg did his review (according to the time stamps).
 
@egreg That's also what I get now. 20 minutes ago I got an empty review history, yet I (and Gonzalo) was unable to review it. Just strange...
 
@GonzaloMedina Off topic (reason: editor quirk)
 
@egreg Thanks. Closing process started.
 
 
1 hour later…
11:39 PM
I don't really know whether I like this new "closing-power" for users with gold medals on a tag.
 
@GonzaloMedina Dangerous.
@GonzaloMedina And you have only five. ;-)
 
@egreg Indeed! I just realized I had it a few seconds ago when I voted to close a question as duplicate and it got automatically closed.
@egreg Do you know if we also have the inverse power? I mean, is just one of our votes to reopen enough?
 
@GonzaloMedina Yes.
 
@egreg Twice the danger, then :)
 

« first day (1292 days earlier)      last day (3644 days later) »