This is probably not worthy of a real question, so I will try here first: Why is \draw (2,2) circle (3); not valid TikZ(that is without unit for the radius)? The default grid in TikZ is 1 cm rectangular. Why can't the default radius unit not also just be cm?
@marmot: Thanks - something told me that it was a stupid question. But now I have another question: Why is the radius in the circle command in manuals and/or tutorials given with units while the position is given unitless.
@marmot: Ok - not even that is true but often. I have now found that the default unit is taken from the x coordinate - so "(if the x-unit is set to, say, 2cm, then x radius=3 will have the same effect as x radius=6cm)"
@hpekristiansen Yes. I also get confused with the units from time to time. Especially since internally TikZ is using pt units. And those get, of course, stripped off when one uses \pgfmathsetmacro (which is, however, a pretty decent tool).
@hpekristiansen And I see your question. Without further checking, \r and expands to 1and which the parser cannot make sense of. @egreg would love this space, I guess ;-)
@CarLaTeX Yes. But at least people don't down vote the answers to those questions. I've have answers of mine downvoted on another site because people decided the question should have been closed and therefore I shouldn't have answered it!
@CarLaTeX Needless to say I don't do anything more on that site.
@CarLaTeX It's just stupid. Some people like to be the rule enforcers. And the site is one in which I will never ask a question, only answer. So it's not good to alienate.
@AlanMunn Well, I see the logic behind that: by answering a question which the community does not want on the site, you're rewarding the posting of that question and thus harming the site. People see your answering as a type of behavior which should be discouraged, and that's why they downvote it. Sure, it may be out of line with the norms on tex.se, but I wouldn't call it "mean".
@DavidZ Well I didn't call it mean. But to the extent that it made me stop using the site, I guess it worked. :) And the question was more a duplicate than off topic.
@LoopSpace You only need a single counter if all of the on-the-fly functions have the same general name format (I think they were \__prg_map_<int>:w). Once you decide to separate out internals, that's no longer the case as they are dedicated, e.g.\__prop_map_<int>:nw. The bit you do need for nesting is the break-point stuff: that's now public.
@DavidCarlisle If you get chance, I'd appreciate a look over github.com/josephwright/texlive-source/tree/xetex-tidy. The idea there it purely to re-arrange xetex.web to get it into line with pdftex.web to help with the merges. I think I've just tided up ..
@JosephWright See texlive list and github.com/latex3/latex2e/issues/43. classicthesis does \PassOptionsToPackage{latest}{latexrelease} \RequirePackage{latexrelease} and this then breaks newer packages. Perhaps the documentation should make clearer that packages should only load latexrelease (and so with the default option current). I hate to think what would happen if a number of packages uses conflicting options here ...
@JosephWright you upped the date on expl3 and triggered the error:-)
@UlrikeFischer current (or no option) would be as bad as latest here, both are equivalent to [2018-04-01] classicthesis shouldn't load latexrelease at all and (I think, but there's still some discussion) that the "suspicious rollback" error should be downgraded to a warning.
@UlrikeFischer well, that's a good question. Post 2015 it's in the format so you can just use it, if you want to support releases older than that then .... put a comment in issue 43...
@DavidCarlisle The OP in the issue actually wrote the more or less the same: the description of the "current" option says "does not change the effective date of the format.". I'm interpretating this as "does no harm but ensures that \IncludeInRelease is defined".
@UlrikeFischer that was the case until it became false this release, there is some discussion whether we should change the documentation or change the code or both.
@DavidCarlisle at least the documentation ;-). Btw: I think classicthesis is trying to replace "\RequirePackage{fixltx2e)". But the method is a bit curious. If it is in an environment with a new latexrelease the format will probably be new too anyway.
@UlrikeFischer that's what its comment says yes. Originally (when we introduced the package) we did say it could be used to update old releases as the end user just needed to get latexrelease.sty from a more recent release rather than update the format, but the idea is that's a choice of the end user. Making the package unconditionally load latexrelease means by default it gives errors in old distributions
@DavidCarlisle ok to recap. As packages which uses \IncludeInRelease are perhaps used on pre-2015 system they need a way to define the commands. What about a separate latexrelease-base only with the definitions which they could distribuate? Then the documentation could say clearly that latexrelease itself should only be used by documents. And if the kernel doesn't use latexrelease-base itself it wouldn't do to much harm if they are more than one version in the wild.
@UlrikeFischer perhaps but before looking at new packages I think there are still issues with using latexrelease (as in the example in the github issue) we'll see...
Quick question: The code below produces a misplaced \noalign error when it encounters the \toprule. Appending \\ to the \caption line fixes it, possibly at the expense of an empty table row. But it feels wrong. Is it?
/bin/bash ../../../texk/web2c/web2c/convert xetex
syntax error: Last token = 290 ("), error buffer = `XeTeXint',
last id = `XeTeXint' (not in symbol table).
../../../texk/web2c/web2c/convert: conversion of xetex.p failed, moving dregs:
@DavidCarlisle Next job is to try to 'collect up' the various internal utility functions from pdfTeX that are already in XeTeX: it's the order these are declared that keeps causing problems with file functions, etc. They seem to be all over the place :(
@JosephWright I think the force push probably confused git here, it started moaning I was 7 commits ahead of the upstream xetex-tidy I force deleted (-D) my branch and checked it out again to track yours, see what happens this time
@DavidCarlisle What I'm on now is trying to sort out the order of various 'pdfTeX support additions'. These are the bit that fail badly when I try to add file utils/timer stuff, so I think are crucial to getting everything right.
@DavidCarlisle Probably we'll need a load of extra tests ... but I really do need to get the code into the same order as pdfTeX: it's nigh-on impossible to follow what is going on as it is
@JosephWright I have some test files for pdf file utils and the rng in the right format but I was going to wait until branches stable before adding, as adding to the makefile is a pain (and the additions would conflict if the branches were merged)
@DavidCarlisle I think the best plan is for you to hold off until I can get everything added and at least passing on Travis-CI for 'does it build at all'
@DavidCarlisle If I can get everything at least added in some way into xetex.web and not entirely killing the build then I suspect we can sort the rest
@DavidCarlisle I'm wondering about all of the utility stuff (for XeTeX and upTeX), e.g. the hex escapes ... but I'm not sure I want to do all of that!
@JosephWright yes that's why I used the expanded branch for getting it working as that seemed fairly stable. It's easy enough to add the tests now i know all the places you need to mention a new test, and the expanded branch is a template for that, so that can be left until the end (or even after its merged upstream)
@BAYMAX subsection 4.2.2 of the tikzsymbols manual contains 3d emoticons. also, since it's a tikz package, coloring shouldn't be a problem. you could go as far as coloring each one of the teeth a different color 😄
@BAYMAX You were asking earlier about why we need TikZ: everything beamer does is done through Tikz. Also, e.g. in my field (linguistics) it's invaluable for drawing trees, something that a GUI program is terrible for. But there are many other uses of it too.
@BAYMAX In support of @AlanMunn I'd like to add that if you use TikZ (or pstricks) for your pictures, then the fonts match those of the ambient document. (And the file is sooo much smaller than some imported graphics.)
@thymaro If this is a question to me, what is "here"? Otherwise please ignore this. ;-)
@UlrikeFischer it does. maybe that's what's so scary about it. and its smoothness. maybe a papery texture would make it more friendly. not that I know how to do that...
@DavidCarlisle err... wasn't the request for the result to be halloween-like? not the ways of making the picture. oh, well, people need a good challenge, I guess.
@DavidCarlisle I must have rebased github.com/josephwright/texlive-source/commit/… at least 10 times trying to get the **** thing to compile: the order (or something) of these functions is the big issue adding the other primitives
@DavidCarlisle My current issue is more basic: why is it that moving a few functions around, when they seem to be defined correctly, breaks everything, and why are the errors so unhelpful
@JosephWright as I say I can look later, and if you are just changing files in xtexdir I can run make in web2c which runs in a few seconds rather than waiting for travis (or waiting even longer for my machine) to do a full clean build
@DavidCarlisle Just changing xetexdir: back on the 'no functionality change' branch, trying to get stuff into the same order as pdfTeX as somewhere there's a function that must get messed up
»The fraction slash ⟨⁄⟩, is used between two numbers to indicate a fraction or ratio. Fractions, unlike inline division, are often given using smaller numbers, superscript, and subscript (e.g., ²³⁄₄₂).«
»The division slash ⟨∕⟩, is used between two numbers to indicate division (e.g., 23÷43 can also be written as 23∕43).«
@JosephWright when I was looking at something similar I think I was failing on similar issues in that you make a "cosmetic2 change in one place but then you find a change file no longer matches,
@DavidCarlisle Trying to use the upTeX .ch file simply fails, and my tests to date show it's something to do with how you create tokens_to_string, flush_str or some such
@JohnDorian well that's what it might say but the division slash in latin modern is at a very gentle angle for 1/2 unless you are going to raise the 1 over it
@JohnDorian actually I think the advice might have been wrong and that it's a bug in latin modern, this is / U+2215 U+2044
@JohnDorian note U+2215 has the gentle slope but U+2044 for fractions looks more or less same slope as /. end result is stick with / but I do wonder if that latin modern output is correct
@JohnDorian as I say, i's probably unused in math other than some magic property in Murray's linear markup for word. unicode-math follows mathtext-14 and assigns it binary infix class, as you say.
@DavidCarlisle if you ever manage to reconstruct the list, you should ask a Linux/Ubuntu geek to write a package that contains the list as dependencies. That way you would only have to install one thing.
@JosephWright I don't know enough about travis. If it was a clean well done script, I would assume all the dependencies would be in the apt addons, but I have seen stuff stuck in all sorts of places and I don't understand the assumptions about what packages are included in the base installation.
For your disappointment, none. They are the same.
If you run
\documentclass{standalone}
\begin{document}
\(\show\lnot a \, \show\neg a\)
\end{document}
you'll get:
> \lnot=\mathchar"23A.
l.3 \(\show\lnot
a \, \show\neg a\)
?
> \neg=\mathchar"23A.
l.3 \(\show\lnot a \, \show...
I fount them in plain.tex, but couldn't find in latex.ltx...
@JosephWright ah yes I knew I missed one libfontconfig-dec (from travis.yml) the other ones that sets up are all xll but I uses --without-x instead to skip the X parts as that's not well supported in the subsystem anyway
@JosephWright So the proper fix would be to define my own int, say \g__spath_prg_map_int and use that to keep track ... of whatever it is that it keeps track of (did someone say "Cargo Cult Programmer"?)
@JosephWright Fair does. I really ought to sign up to the right mailing list, or watch the github repository so that I get notified of stuff I ought to know. Still, it's an ego boost when someone asks a question on the main site about one of my packages.
@JosephWright I've been listening to too much Radio 1 (blame the kids ...) where they frequently talk of artists "dropping" a track meaning that it's been released.
@JosephWright just lines with dates coming back? I can eyeball them but if you'd rather I can leave it for now and just update the ones with the suspicious rollback error
Hey guys, I have a quick question. I'm recycling a Thesis.cls from sharelatex and turning it into something useful for my university's particular requirements. One thing I have an issue with is that I need to use `\protect\` in the title, but I don't want to subject future users of this template to that
I've found a command \pdfstringdefDisableCommands{\let\\\space}... could I stick a \protect in there? Or is this not a good diea
I also have another question about some weird spacing issue with an underline
@Ronen probably not a good idea but it's a bit hard to know what to suggest, it depends why \protect is needed (it shouldn't be needed as \\ is a robust command by default)
It is a bit messy, but I don't know what's in book.cls to differentiate from the few commands I'll need in a custom .cls. Specifically, I want the .tex to have no \newcommands or \usepackage or anything that a non-coding user would ever need to think about
Then again, I have no idea what's in this Thesis.cls that I actually need, which isn't in book.cls
@Ronen it's so much better to start with a basic class and add what you need than start with a class that loads hundreds of packages with undocumented and unknown conflicts that's impossible to configure. A major problem with the templates available on the web is it leads people to do the latter.
@Ronen well it's loading book.cls but allowing you to change that to memoir or whatever by changing baseclass and then (slightly inelegantly) changing the default size option from 10pt to 11pt
@Ronen but then all uses of \justifying and \newlength and \small{} are confused. \uppercase it should be \MakeUppercase and only applied to #1 . vmargin package for setting page size is a bit deprecated these days, better to use geometry you probably don't need to load epstopdf Although actually (unlike some classes you see) it could be a lot worse it doesn't load hundreds of packages, it's only 128 lines altogether
@HenriMenke I just came across this question of yours and feel that perhaps this answer of mine might be related. Please have a look. If not, I'll be happy to remove the link.
@JosephWright it's sort of understandable but is this as intended?
+ (foocmtt.fd(foocmtt.fdFile: foocmtt.fd YYYY-MM-DD Font defs for `foo' fonts.(foocmtt.fd(foocmtt.fdFile: foocmtt.fd YYYY-MM-DD Font defs for `foo' fonts.LaTeX Font Info.....
@marmot To be honest, I haven't used circuitikz in a while. Back then I was doing electronics lab at uni and that was the only time I ever needed this package.
tikz?, pstricks? everyone knows the underworld used picture mode.
\documentclass{article}
\usepackage{color}
\usepackage{pict2e}
\pagecolor[RGB]{255,177,69}
\begin{document}
\begin{picture}(120,200)
\thicklines
\put(100,100){\circle{150}}
\put(70,130){\line(1,-1){10}}
\put(70,130){\line(-1,-...
For your disappointment, none. They are the same.
If you run
\documentclass{standalone}
\begin{document}
\(\show\lnot a \, \show\neg a\)
\end{document}
you'll get:
> \lnot=\mathchar"23A.
l.3 \(\show\lnot
a \, \show\neg a\)
?
> \neg=\mathchar"23A.
l.3 \(\show\lnot a \, \show...
