@marmot I very like how you think those things, I could not be able to understand you, but I like it :):):)
Hey, I have a bad box message: the famous \underfull hbox (badness 1000) in this piece of code:
\hypersetup{pdfborder={0 0 0}}% % list of OCGs, space separated \def\ocglist{}% \foreach \X in {0,...,\totalnodes} {\xdef\ocglist{\ocglist\space myArrow\X}} \fbox{% \parbox{50cm}{% % on/off All \showocg{\ocglist}{\fboxsep=1pt\fbox{\rule{0pt}{1.7ex}SI}}/% \hideocg{\ocglist}{\fboxsep=1pt\fbox{\rule{0pt}{1.7ex}NO}} Seleccionar todo\\% % checkboxes \foreach \X in {0,...,\totalnodes} {% \layerCheckBox{}{myArrow\X}{on}\begin{tikzpicture}[baseline=-.82ex]% \pgfmathsetmacro{\huenum}{\X*0.75*(1/\totalnodes)}%
I think the problem is the \quad in \end{tikzpicture}\quad% because in another examples I delete \quad and the bad box message disappears, but in this case when I delete it the message does not disappear :(((
Sorry, when I delete \quad the message dissapears!
But now the checkboxes + arrows are very close to each other, @marmot @AlexG
@marmot no messages, thanks! Yeah, I think it is due to the insertion of a new line. That causes the warning
@marmot regarding to the orange-rounded-rectangle shapes (do you remember the questions?), now I want to fit the page to the limits of those shapes. Actually I am adjusting paperwidth parameter from geometry to what I want, but I would like to make it automatically. Do you know how?
Maybe changing article by standalone? (Not a good idea, I have 4 errors)
@marmot I get a very small page. I also tried \documentclass[varwidth]{standalone} but this time commenting \usepackage[margin=0in,footskip=0in,paperwidth=52cm,paperheight=29.5cm]{geometry}, also tried \documentclass[varwidth,varheight]{standalone} with \usepackage[margin=0in,footskip=0in,paperwidth=52cm,paperheight=29.5cm]{geometry} commented and uncommented, and the same results: a page that is reduced (I have 6 orange shapes where I see only 2). No problem, I will use article
@marmot when I replace \parbox{12cm} by \parbox{\paperwidth} I do not get a vertical line at the end (I think that it is there but it is outside the dimensions of the paper size):
Any idea? I tried \parbox{\paperwidth-<something><px/pt/cm/em/...>} but it is not correct
@manooooh Well, it was not my idea to use \fbox when you are loading tikz... you cannot expect that a \parbox parses things in the way tikz does. I think that it is best to go in small and logical steps. So far all the answers on this you got by @AlexG and me were under the premise that you are using article but now it is standalone, which changes a lot. Do you have made your choice on this? I will be really online in an hour or so.
@manooooh As I said it should be feasible to get collect the size of the thingy in the end, write it to the aux file, and then feed geometry with that if you really want article. (But I do not necessarily see why that would be better.)
@manooooh \fbox adds \some space (\fboxsep+\fboxrule) around its content, Therefore \paperwidth is too much. You could try using \frame{...} instead of \fbox{....}. Or \dimexpr\paperwidth-2\fboxrule-2\fboxsep\relax as the parbox width.
You can have as many \caption as you need in a figure
\documentclass{article}
\usepackage{graphicx}
\begin{document}
\listoffigures
\begin{figure}[thp]
\begin{minipage}{.5\textwidth}
\centering
\includegraphics{man}
\caption{man}
\end{minipage}%
\begin{minipage}{.5\textwidth}
\centering
\in...
@DavidCarlisle some hope that font stuff don't break to fast: "font related code is also supposed to work with plain which in turn means work with luatex"
@UlrikeFischer The picturemodalism image was was on the bottom left of the previous version of the IFT poster. I just noticed that it got replaced in the version published on the Dante website, so I'm now wondering if its still secret or can it be shared?
@marmot Don't worry, now that the tikz/pgf code is easily accessible on github, maybe one can write an interface to create picture mode drawings in the same hight quality as @DavidCarlisle airplane!
@samcarter ah. You mean the marmot. The designer of the poster decided in agreement with the artist @DavidCarlisle to use the newest outstanding picture "duck or grouse". But the marmot isn't secret, it has been shown previously somewhere (chat?) and was also in the exhibition. .
According to the almighty DPC, picture-mode is also a recursive acronym: Picture-mode Is Currently The User Recommended Enterprise Method Of Drawing Easily!
Hi. I was thinking of asking to move tex.stackexchange.com/q/481892/3406 to maybe the Emacs SE. Would that be possible? And if it's possible, should I flag it, or something else?
@daleif Maybe adding the decorative element to the titlepage definition would be easier? Then you would not need to remember to issue a \frametitle on the title frame?
@samcarter I have to disable the title frame and force users to use \maketitle. Title frames need a special background, and we cannot change that from within the frame. So messing with the style of the frametitle is not an issue here.
@samcarter plus. since the contents of the title page contents is vertically centered it is hard to place something in a specific location. I could add it to the background, but, the decoration moves position if there is a subtitle....
@JosephWright @DavidCarlisle @UlrikeFischer did you see Herberts issue with babel+lualatex on the texlive list?
@daleif As for the positioning issue: for easy absolute positioning I either use picture mode in the background canvas or tikz with overlay, remember picture options.
@samcarter that was what I was using earlier, but it requires at least two compilations before it settles, and that may confuse users. Right now I just take over the background canvas.
@samcarter why, it works just fine. I have the frame numbers in the footline (is goind to disable it, as I'm currently doing a special cutdown version, and will make a general one later)
I even add the background for the footline from the background canvas (using a tikz picture in the canvas)
@daleif True, on the other hand a user will less likely overlook a malformed image than some tiny numbers in the bottom corner and will hopefully learn to compile often enough
@naphaneal No publisher will want dvi. If they want ps then you compile to ps, I see no benefits from "cross compiling" from within LaTeX it self. It will just be seperate runs anyway. Might even be impossible as a subsequent LaTeX run from within say pdflatex would want to access the aux file while is helt open my the parent pdflatex process. Plus, most documents needs multiple passes, so you are better of doing an external script that compiles for you.
@naphaneal cross compiling normally means using one back end other than the default, not that you can compile all at once doesn't it?
@naphaneal most latex sources work unchanged for latex (for dvi) and pdflatex (for pdf) main things are to avoid using explicit driver options like [pdftex] and to avoid using explicit graphics extensions.
@daleif that particular tracker only been there for a couple of days but babel was covered by the core latex bug reporting tracker up until now.
@DavidCarlisle ahh, that makes sense. But in general I think it is a good idea to mention the tracker and contact information at the start of sty files etc.
Exactly which files can \IfFileExists look at in a LaTeX context? Font files, map files, fd files? (trying to determine if a specific font is available (it is a custom font and does not reliably have a package associated with it).
@DavidCarlisle no, it means one form of source code (e.g. C# code) get's compiled by different types of compilers, e.g. by Mono for Android and by MSIL for Windows.
@naphaneal yes that's what I mean, but it doesn't mean you can do all them at once. so taking the same source and running with pdflatex then latex is as close as you are going to get.
@daleif IfFileExists is \openin followed by \ifeof so it can test any file that is in tex's input path, the file type not important but it has to be in TEXINPUTS
@naphaneal or simply get the pdf and then convert to ps at the end if you need both, You don't have to generate the ps via dvi
@daleif for fonts in pdftex load the font in \nonstopmode then test if it is \nullfont
\scrollmode
\font\zzz=cmr10 at 12pt
\errorstopmode
\ifx\zzz\nullfont
\show N
\else
\show Y
\fi
\scrollmode
\font\zzz=wibble 12pt
\errorstopmode
\ifx\zzz\nullfont
\show N
\else
\show Y
\fi
\bye
@daleif ^^ unfortunately it triggers maketfm to try to make the font but does in the end say Y for cmr10 and N for wibble
@Skillmon uhh, could someone please make the pgf (and tikz) manual indexes more readable, for example with letter group headers and perhaps page headers. It is very difficult to navigate when you are down one of the very long entries.
@daleif there are different Makefiles for different engines contained in pgf/doc/generic/pgf, e.g., the version for pdftex is in: pgf/doc/generic/pgf/version-for-pdftex/en
which is just misusing etoolbox lists and the fact that the header on a page runs after the page is finished
@Skillmon When I look at the tikz or pgf index I'm often annoyed that it is so hard to navigate. We do not have to do it now, but IMO it should be on the list of nice to have
Was looking at it once just for myself, but could not figure out how to compile it.
@PauloCereda and @DavidCarlisle -- An even worse story here. Dave Walden, Karl Berry and I had a two-part paper about TeX accepted to the IEEE Annals of the History of Computing. The first part, prepared with LaTeX, was published last year, and the second scheduled for this summer. But in the interim, control of that journal was transferred to a different group, and we have learned that the journal is now to be prepared entirely in Word. We may withdraw part 2; status still unknown.
in ../../text-en/pgfmanual-en-pgffor.tex: ! Package pgfkeys Error: I do not know the key '/pgf/foreach/parse', to which you passed 'true', and I am going to ignore it. Perhaps you misspelled it.
@AlanMunn -- Nice idea, but the IEEE production group is reprocessing it entirely. You don't want to know what they did to the various TeX logos, which they did "paste" in from a faulty preprocessing attempt. (Computer modern, in the wrong size, mixed in with sans serif text. With all the kerning gone. Reverted to the text form.) If they don't permit including an explanation about the change from LaTeX to Word, I have asked that my name be removed, which is equivalent to pulling the article.
@barbarabeeton Wow that's pretty awful. Do you think you could do a better job of the conversion with pandoc and send them a Word version you could live with (perhaps with the TeX logos as images?)
@JosephWright -- Unknown. All I know for sure is that the previous production crew for this particular journal were actively trying to make a usable LaTaX environment for authors, but this was derailed by a change in the overall IEEE publication strategy.
@AlanMunn -- That might be done for the text, but forget the logos. There were at least three different font families involved, in as many different sizes, and getting an attractive XeLaTeX logo isn't trivial under any circumstances. Better as the alternate "text" style, per DEK's stated preference.
In other good news: generating an index like the one used for the memoir manual (check the letter bookmarks), that is doable in aprox 10 lines of code and 10 lines of makeindex configuration
@Skillmon I'm not a git user, plus I never got the entire manual to work
@barbarabeeton not really, most is just ad hoc programming. Then coming back to a macro years later and you are embarrassed you implemented it what way.
@Skillmon since most of may work is not collaborative, subversion is fine for me. Plus, the manuscripts we work on are easier in subversion, git is too confusing'
@barbarabeeton It's a horror story, but somehow I am not surprised. Back in 2011 I helped tweak IEEEtran.cls in order to make its output more like their standard Word template. That, right there, is a big red flag. Looking at the file, I really wanted to change everything, but we weren't paid for a major overhaul of the file.
@daleif I never use the external library. I tend to do things properly in a Makefile and all the tikzpictures are in an img folder and build by the Makefile with standalone.
@daleif if you want to, I can send you my setup :)
@daleif though there is an answer of mine that shows it (it is a copy of an actual Makefile I used for a project of mine)
Hmm, my index code is some jacky s***, but I somehow got broken indexitems (aka lists of subitems that are broken across a page break) to start with xxx continued.... Now how did I do that
@daleif if your code is s***, why not make it accessible so that others can take a look at and propose something better/cleaner. That's the whole point of open source
@Skillmon I was considering opening the source for my book. But decided against it after someone from another university asked for a copy so they could make their own version of it. No thanks.
@daleif understandable. Sitting on a couch looking football and trying to get things done while watching with one eye:)
@daleif but if they cite properly they can use almost anything from your book (assuming academic purposes here). I don't see why you couldn't open up the sources once it's published.
@daleif I will not (for now) as my intention is to write my code, describe it and then send a pull request. I don't want to include stuff that is not really related to that.
I'm just telling you if you wanted to test it. I'm nut running it against the pgfmanual, just stole the idx file (did not want to wait 30min for it to compile)
@Skillmon trick, all those files are via \include add \includeonly{} and you can get it to ignore most of it, and it will still build the full index ;-)
@Sebastiano not all the names are known 5 people starred the comment of @CarLaTeX that you are questioning that is 5 (unknown) people who agreed with the comment for example. You can search this chat or the main site and you will find dozens of requests that you improve your questions by supplying proper examples. I find it surprising that you seem to be surprised or object to the comment.
@DavidCarlisle I don't have to look for anything. I've already been reprimanded enough by you and I'm not here to argue. I'm trying to do my best. You have to appreciate this as well, and you also have to be tolerant.
@Sebastiano well stop arguing then chat.stackexchange.com/transcript/message/49785813#49785813. I don't think I have ever reprimanded you, I have just commented that your questions are hard to answer (and may get downvotes from others) if you provide no example
@DavidCarlisle I felt reproached by you last week, I seem to remember. Can you at least accept this? I have my own sensitivity and I have always respected everyone. But being persecuted, I don't accept it at all.
@JosephWright I'm sorry, Joseph. I'm just reading now. I have finished. Fullstop.
@Sebastiano Unfortunately, that's not how you actually behave. You showed a lot of disrespect, for example: You edited many questions and changed their meaning completely --- how is this respectful towards the person who asked the question? I personally can tolerate a lot, but I cannot tolerate someone blatantly lying.
@Skillmon It always depends who the readership is and what the scope is. If you think as the manual as a reference that describes some commands, then yes. However, the pgfmanual also has a crash course for newcomers, and explanation of what this is good for and so on, and examples. If I was Till Tantau and got to see that people complain that the required libraries are not explicit enough, but at the same time manuals with no explanation whatsoever get praised, I might have mixed feelings.
@marmot (sorry to interrupt your discussion) The problem I have with pdfmanual is that you have no other place where to get to know what tikzlibraries are needed for a specific example to work. No matter how you try, you are only left with trial and error. And of course, this is just a small nitpick in otherwise a great an well documented package.
@yo' Well, yes. (I was setting out to cure this but terribly failed, not because I cannot find out what the libraries are but because I cannot deal with repositories.) But IMHO still better than no examples at all.
@Skillmon git is not very beginner-friendly. For people who don't need the whole rebase thingy of git, you are better off using Mercurial. It has a much nicer command-line interface and works natively on Windows.
@HenriMenke Git is fine on Windows too nowadays, though I totally agree that the basic commands are easier in Mercurial. But almost anyone at least as a beginner will be using a GUI ...
@yo' Yes, but this was what my conversation with @Skillmon is about. I personally do not work with xparseand expl3 because even if I understood what they are doing, my collaborators or users I might write answer for, may not. (But I am very exciting to just realize that the l3draw manual has grown a lot and looks really great!)
@JosephWright I am a beginner who refuses to use GUI.
@marmot But the point of expl3 and too some extent xparse is they are developer tools: developers write packages, document them on their own terms, don't expect users to start hacking into the internals
@marmot Fine with me: I do use a GUI, but also quite a bit of command line stuff (entirely for SVN, quite a lot of Git)
@marmot well, the problem is, in some sense, that expl3 has a very large scope, and especially at the beginning, the learning curve is steep. (Well, the same is true for tikz, after all.) And once you get higher up, the reference manual in interface3.pdf is just fine
@egreg are you sure....? Today I asked my teacher about why the complex numbers are not a vectorial space and he answered this (I do not remember his exact words): "The complex numbers ARE NOT vectors because for example: 1) vector · vector = scalar, meanwhile complex · complex = complex; 2) complex^complex = complex and vector^vector = not defined; 3) complex/complex = complex meanwhile vector/vector = not defined"!! How could you then argue that they form a vector space if
there are complex operations that do not inherit the vectors?
@JosephWright Yes, that is great, and I agree with all that. The question is, though, if one should extensively use in documents for others, which they may want to customize. That is, even though this might very well be the much cleaner approach, they may end up getting a code that they cannot understand.
@marmot you don't really need to use git at all, you could simply download the zip file of the files which github makes available, work on the manual and then (someone) would get it posted back upstream
@manooooh Oh that's a stupid teacher (sorry to say that). Mathematics doesn't work like your teacher expects, obviously, because set X is a vector space over field F if they satisfy the axioms.
So you can also easily say that the complex numbers are an additive group (and they indeed are!) or that non-zero complex number are a multiplicative group (and again, surprisingly enough, they are!)
@JosephWright Did you write \draw_transform_matrix:nnnn? if so, do you think it is feasible to extend it to 3 dimensions, including the inversion (which will obviously require more than just multiplying a rotation angle by -1)?
@egreg I do not understand you. I completely understand that vector/vector is undefined, butcomplex/complexis defined. So, why would you say that the complex numbers are vectors?
You can have as many \caption as you need in a figure
\documentclass{article}
\usepackage{graphicx}
\begin{document}
\listoffigures
\begin{figure}[thp]
\begin{minipage}{.5\textwidth}
\centering
\includegraphics{man}
\caption{man}
\end{minipage}%
\begin{minipage}{.5\textwidth}
\centering
\in...
