I forced a line break at the beginning of my proof, so that my itemize can always be flush left (so the first item isn't pushed over by the "Proof." text). But there seems to be too much space between the lines. What could cause this?
Anyone has a different opinion? Should I cast a close-vote and flag it for migration?
Welcome to TeX.SX! You can have a look at our starter guide to familiarize yourself further with our format. I'm afraid that while you stumbles across this problem when using LaTeX, the question is not about "how to make the figure 0.618\textwidth in LaTeX", rather about the design customs and tips. This is off-topic on this site. However, it can be on-topic on a sister site GraphicDesign.SE. I'll see if moderators can move the question there :) — tohecz1 min ago
@JosephWright There comes a point where you seriously consider dropping classic tex support. If we advertise a new lowercasing function I think it should (this century) work with utf8 and writing an expandable lowercaser decoding utf8 as it goes sounds rather dull.
@JosephWright The zeroing of the \lccode (huge) array would be performed at compile time. It would be different if LuaTeX had a concept of "lccode status" similar to what it has for catcodes.
@egreg Yes, I follow. What I mean is the current discussion in the issue comments covers what one could do if you built a format, where you could reasonably zero everything as part of the format-building step.
Clearly that is not on for a package, so we have to compromise/rethink interfaces/...
@DavidCarlisle I don't see Frank going for dropping pdfTeX support any time soon, and indeed that would have a big knock-on effect on packages using expl3
@JosephWright I know, but sometimes you can't help but wonder....
@JosephWright I don't see how you can lowercase without decoding the utf8 bytes giving the unicode slot, lowercasing to another slot and then re-encoding as utf8, although I suppose if you have already expandably got to an 8bit LICR representation \^O or whatever you can get away with leaving the \^ and lowercasing the O, I suppose the question then is how much of Unicode has a lowercase/uppercase pairing that can't be expressed as the lowercasing of an ascii letter together with a diacritic
@DavidCarlisle As you might know, I favour saying 'pdfTeX is essentially 8-bit: if you want/need Unicode use a suitable engine'. On the other hand, LICR does work for lots of people, so there is some realism
Is there a way to redefine \mathnormal? I have the problem that mathnormal still uses Latin Modern, while e.g. \mathit uses already the Minion Pro (tex.stackexchange.com/questions/152714/…)
@DavidCarlisle I'm currently imagining that an expandable \tl_to_(lower|upper)case:n will 'expect' to receive LICR in the pdfTeX case, and of course 'proper' UTF-8 otherwise. So as you say \^o => \^O for upper casing, etc.
@DavidCarlisle I guess at present I'm mainly thinking about the subset of cases where \(upper|lower)case does work but is not expandable. As far as I know, the primitve fails for Greek, etc. with pdfTeX anyway
@JosephWright A fair analysis wouldn't fit in the line length of the comment block, and anyway if you are poking the elephant in the room, it's best to be brutal rather than fair:-)
@DavidCarlisle I have in the past suggested that we say 'OK, LaTeX3 will assume only engine-supported input, so pdfTeX strictly 8-bit and UTF-8 with XeTeX or LuaTeX', so I'm not necessarily disagreeing entirely. However, lots of people do manage perfectly happily with pdfTeX, at least for western European languages.
For end users, moving to XeTeX/LuaTeX is easy enough, but that's not so simple for publishers
@JosephWright Speaking of a (relatively small) publisher: I'm not sure which will be bigger pain-in-the-keyboard: whether switching to L3 or switching to Xe/Lua*.
@tohecz Thinking is for a larger publisher using something customised rather than just TeX Live, adding a new format is easy but altering binaries is hard
Of course, in reality it's quite possible that it might be an 'all of nothing' switch anyway
One more practical problem is that as @DavidCarlisle says at present you'd want the XeTeX font loader but other aspects of LuaTeX, so there isn't exactly one idea engine
I'm trying to plot Weierstrass function using only basic TikZ picture functionality (no gnuplot or whatnot). How do I use sum in a \draw? Do I have to make a new command? Use a loop?
An alternative (ugly) solution with manual summation:
\begin{tikzpicture}[xscale=2.2,yscale=2.7]
\draw[thick, co...
Thanks. I manage to change the color of error bars in this plot: http://tex.stackexchange.com/questions/156224/how-to-align-two-pgfplotstable-objects-with-graphs by \addplot+[color=black,only marks][error bars/.cd,x dir=both, x explicit], but cannot find out how to change the color of dots. Any idea how to do that?
it does not seem to have \addplot slot for the dots (based on my limited understanding), so I am confused
@JosephWright I like Bruno's analysis. In \tl_transform:nnn one could set codes to 0 for the ASCII characters and warn about the danger of using other tokens in it.
@JosephWright why don't we use \savinghyphcodes in latex (is it an etex feature that doesn't work, or did we just never do it)?. If the lcodes for hyphenation were language specific (by setting \savinghyphcodes non zero) then you could switch to an all 0 lccode regime as a no-op base for transforms by switching to a null language.
@DavidCarlisle I think that \savinghyphcodes is used only in the hyphenation mechanism, to make pattern lookup independent of the current lccodes; switching to all 0 lccode would make \lowercase unusable in a document.
@egreg oh yes so it is, well you could use it as a "clean" transform primitive for switching tokens specified in each call, and for lowercasing strings you could use a slow expandable version, perhaps, but at least you wouldn't mess up hyphenation while doing so,
@egreg I had a dim recollection of some ancient discussions along those lines but the details escape me (I must have been in some discussion for me to remember that there was such a thing:-)
@egreg, @DavidCarlisle Reading the e-TeX manual, it seems clear that the codes only get saved w.r.t hyphenation, so that doesn't help in the case of 'weird tokens'.
Also, we are talking something at format-building time: I wonder what LuaTeX does in this regard (as hyphenation patterns are not saved in the same way anyway)
Hmm:
> The motivation behind the e-TeX extension `\savinghyphcodes` was that hyphenation heavily depended on font encodings. This is no longer true in LuaTeX, and the corresponding primitive is ignored pending complete removal. The future semantics of `\uppercase` and `\lowercase` are still under consideration, no changes have taken place yet.
In the TeX FAQ it says that TeX wasn't designed as a programming language. Is there a definitive source for that or is it just generally recognised as true?
@jtbandes oh my, I'm lost. Well, maybe I'm not, since in Czech, some valences have alternating cases and you can decline the phrase in two different ways :)
@AndrewZabavnikov I think it must be that if it only happens at a certain zoom level. I tend to not worry about that sort of stuff since it depends on how the displayed page is aligned relative to the pixel grid of your display.
@tohecz LOL :-) If you've got "looking into" you've got an adjacent "ing" and "in" which can clash a bit and with "to look into" you've got too many "oo"s although there's some variation as to the reader's pronunciation of "look", which has regional variations. I'm a writer so I get pulled up on such minor details as repetition and word/syllable etc clashes ;-) Of the two "started to look into" sounds too clumsy but both are grammatically correct.
@tohecz I don't know much formal linguistics, but you can do the same with "started investigating" or "started to investigate", so it's not entirely surprising that it works with "look into"
@tohecz :-) Yes, it's great, there's a completely different set of "rules" for children's writing. In fact, it's almost the opposite. (For example, adult fiction should minimise the use of adjectives or adverbs, but with children's fiction the more the better!)
@AndrewZabavnikov I think you're correct in that it can't be enabled. From what I've read it is a compile-time switch but those who have tried it ran into errors.
As far as other viewers, I'm not sure. I've used Sumatra for some time and I have no issues with it.
@AndrewZabavnikov Nothing to do with your LaTeX sources. Open the pdf in Adobe Reader and it will display fine. What I meant when I said "no issues" is that I see this aliasing sometimes, but I don't worry about it.
@JosephWright yes I was thinking that because hyphenation was using saved lccodes you could zero the main lccode table, but as egreg said that would mean \lowercase was pretty useless for lowercasing
I have \DTLforeach{mycsvstuff}{\listofguys=Names}{...} where \listofguys gets a comma separated list as well. Apparently, when I go with \docsvlist{\listofguys}, the list is not parsed. I probably need to expand something, I guess. :)
@JosephWright with a cameo of @egreg's code, of course. :)
@JosephWright I wish. :) Actually, I edited the .csv file manually (just one entry was bad).
(sorry, notebook is big, but apparently not big enough -- stupid 15.6"). :)
Did Apple stop those insane 17" MacBook Pros?
@JosephWright: about the case case (ooh look what I did there): the L3 team could distribute some wearable devices that produce a 220V shock every time a user tries to (lower|upper)case some sentence, and probably some red text on a yellow background appears on screen shouting "YOUR TEXT, YOUR pROBLEM! (oh teh ironY!)" :)
@PauloCereda With arara 7.0 it will be possible to think compiling a file to have the preview open on the screen. And with error correction on the fly.
@Speravir @JosephWright Heiko's answer should be moved to the other question. There's a difference though, because the newer one asks specifically for LuaTeX.
@egreg Yes, I noticed this too late, as I started the process. But I added a link to Heiko’s answer in my one (before he made the edit for Windows in his answer). (@JosephWright) When Heiko’s answer would be moved, this would be good, but then both our answers had to be tidied up(?).
@percusse this answer: tex.stackexchange.com/a/158792/3954 I flagged it after quickly reading it in the review section without seeing the actual question first (something I had never done before), but then when I read the question, I decided it shouldn't be flagged.
I'm wondering (since it just happened to me), is it possible to "cancel" ("undo", "delete", whatever) a flag I just raised if I misclicked it, so the mods won't waste time?
I looked everywhere, but I can't find such a feature.
@Speravir Thanks. I guess I just had a little "panic attack" being the first time I consciously mis-flagged an answer; after all, I guess it's not such a big deal.
I'm wondering (since it just happened to me), is it possible to "cancel" ("undo", "delete", whatever) a flag I just raised if I misclicked it, so the mods won't waste time?
I looked everywhere, but I can't find such a feature.
