@PedroTamaroff They work, but not for multiline entries. For that, you need to add four spaces in the beginning of each line, or click a button named Fixed font which appears near the common buttons. :)
@PedroTamaroff I will see what I can do, I'm not very good with math stuff, but maybe I can help. :)
@PedroTamaroff The syntax is wrong, I guess. \usepackage usually has one optional argument and a mandatory one, usually in the form of \usepackage[<opt>]{<required>}.
I am trying to compile a file using TL 15, which uses animate package, but I get this error
/usr/local/texlive/2015/texmf-dist/tex/latex/animate/animate.sty:84: Package an
imate Error: pdfTeX, version >= 1.20, required.
My file just does
\usepackage{animate}
I would like to know your opinion about what is the correct approach in the case of the fontspec commands. Look at the comments between Mico and I here.
Originally it was \setmainfont[<options>]{<font>}. Now it's \setmainfont{<font>}[<options>], although, for compatibility the old behaviour works (and will continue to do so).
What's the correct approach if I see someone with the old syntax? (I will tell a little bit more about my opinion after I see your opinion.)
@Nasser luatex (now) gives luatex version not pdftexversion for \pdfversion (next release it will be undefined) just \def\pdftexversion some big enough number
@PedroTamaroff if you use miktex or texlive (which is almost every tex user) you should not have to download and install that "by hand" the package manager of your tex system should do it. Also, for font packages like amsfonts or euler it's not enough to have the package, you also need the fonts and font metrics
@JosephWright I guess we need to ask heiko if he can update or if he wants the kernel updated a bit to make it easier to spot the internal use of stepcounter (perhaps we should check for loops too)
@JosephWright yes it fixed a real reported bug but we could perhaps change the coding a bit (not looping would be good;-)
@JosephWright I don't think we discussed it! Do you disagree with the syntax in the first place, or the change? I have to admit I didn't fully think through the ramifications of picking up a "[" accidentally with the optional argument last, but I stand by the rationale that logically it makes more sense to put the font name first :) Also, people shouldn't be using \fontspec anyway—perhaps in a future release I'll deprecate it (but never remove it entirely).
@Manuel You've got me on this one—I don't know what I'd suggest either! Would it be better if the manual said "both are fine"?? :)
@WillRobertson certainly options first is more normal (and behaves better:-) I suppose which is more logical depends whether you think of pre-specifying the options before you load a font or whether you load a font and then apply some options, (preferably without knowing the implementation details:-)
@JosephWright @egreg do you have recent texlive installations to hand? I'm trying to add release date markers to changes.txt along the lines of the following but without fetching from remote archives I only have 2015,2014,2011,2005, would be intersting to see head */texmf-dist/doc/latex/base/changes.txt (and matching \fmtversion) if you have it easily (otherwise don't bother:-)
######################
# 2014/05/01 Release
######################
2014-09-29 Frank Mittelbach <[email protected]>
* nfssfont.dtx Typo in definition of \encoding (no change to processing)
@WillRobertson What I mean is I saw your change and thought 'Hmm, I would not have done that'. I understand the idea of logic but really the LaTeX2e standard is \foo[options]{thing}
@WillRobertson We can talk about this in the context of xparse and wider LaTeX3 design in a couple of weeks :-)
@WillRobertson My main objection is that \fontspec is not defined with the argument list momO{}, but rather with mom and leaves for later a check that a [ follows.
@HeikoOberdiek a lot more code than my suggestion (which I'll add below so others see it:-) is it really worth raising a warning rather than just silently accepting as before?
@HeikoOberdiek ooh the meta-meta-meta-meta-compiler evolved into a syntactic macro expander, so it's crazier than ever. :) I need to stop writing these things and finish the arara user manual (as the code is ready for production). But I keep procrastinating! :)
@HeikoOberdiek yes but it isn't you or me that sees it (and we could use \tracingall:-) it's the document author, and it's not clear that the message tells them how to unwind the mess they are in:-)
@HeikoOberdiek but mainly I wanted to check that if you were planning to update zref to detect this, we don't update stepcounter again and break it again, so wanted to keep you in any loop
@DavidCarlisle I have added a todo to remove the counter used in \zmakeperpage from the clear lists of the other counters. It's not perfect, but should hit the common use cases.
@DavidCarlisle I think your patch is ok for the normal user. I agree, it's an too internal issue for him. Thus a package would be nice, which visualizes the relationships between the counters.
@JosephWright BTW, is it possible to purge date from the database including old revisions? For example, if a user exposes email addresses, see question https://tex.stackexchange.com/q/252745/16967
Anyone know why xcolor provides RGB but not CMYK? I'm defining some colors and they are both listed in RGB and cmyk as either triples or quadruples within {1,...,255}^4
@DavidCarlisle I see, "your" agency wants to read our chats. If we encrypt it, it becomes far more interesting for them and they learn about LaTeX and may be wants to use it ...
@JosephWright Okey, then we could see it as a way of creating peer pressure. I do prefer \setmainfont{Font}[key=value] (although with an optional space before the [, but that's an xparse).
@WillRobertson If you are in doubt, well, the question doesn't matter anymore. In any case, as it stands now in the manual, what would be a corrrect behaviour?
@PauloCereda by the way the country is about to come to a standstill, trains running slow due to fear of lines buckling... forecast to get well over 30 this week....
@egreg It was briefly defined as momo or similar, but there were a couple of edge cases that needed the obfuscation. Erm, you're totally right that mucks up what signature should be.
@DavidCarlisle Indeed: I am a great believer in allowing the units to fit the situation. At work I weigh in grams but at home for cooking it's pounds and ounces.
>The UK-TUG Committee are pleased to give a first notice that the 2015 Speaker Meeting will take place on the 31st of October in Oxford. Attendance will as usual be free for UK-TUG members. Full details including the venue and a call for talks will be circulated nearer the date of the meeting.
@WillRobertson I personally like it. But since you don't have a clear answer, that discussion is over. Now I'm a but nervous about if I acted properly, given what is said in the manual. And also I'm a bit worried, because in the past I had an argument about this exactly, and it was the reason I stopped editing things on this page (I decided not to edit, nor to review edits), so it bugs me.
@JosephWright I generally can’t around make any date around 1 November (family commitments), but if @PauloCereda is thinking of coming I’ll see what I can do :-)
@JosephWright I know you picked the date very early to enable people to free that weekend, but it didn’t seem my input was going to change anything so I didn’t say anything at the time :-)
deluxe option for luatexja-preset crashes. For example, when I compile the following code:
\documentclass[12pt,a4paper]{ltjsarticle}
\usepackage[hiragino-pro,deluxe]{luatexja-preset}
\newcommand{\jugem}{寿限無 寿限無 五劫の摺り切れ 海砂利水魚の 水行末 雲来末 風来末 食う寝る所に住む所, Supercalifragilisticexpialidocious! }
\begin{do...
@WillRobertson Okey then, I won't recommend changing it whenever I see the old syntax, although that would be well to make peer pressure. I do prefer the ConTeXt way, with logic behind, \setmainfont{<font>}[<tons of options>] :) In my opinion is an old costume to put the optional argument before, and, in that particular place, the preamble configuration, it's definitely unnecessary.
@egreg Oops me. Sorry I didn't see this earlier. I'd complain that people didn't email me about this… but I can't be certain I would have had the time to switch things back :(
So speaking of interfaces… we've been having some lively discussion about what unicode-math should do with things like \mathbf{a} vs \mathbf{Hom} vs (maybe) \mathbf{M_x}. And similar for \mathrm, \mathsf, etc.
The next version of unicode-math will define commands \symbf for accessing the plane 1 bold math chars that are only suitable (in current unicode maths fonts) for single letter variables. This is in contrast to \mathbf, which uses (ascii) bold letters with kerning etc suitable for multi-character strings of text. Let's say that I freeze the current definition of \mathbf as \mathbf* —…
@WillRobertson probably:-) distinguishing the first two is easy enough of course, but the third gets you into \bm territory and that might be too hard to make predictable results, could be made to work though, perhaps. But would it be better to just force better markup \mathbf{X_y} is already a bit dubious and would be better marked up as \mathbf{X}_{\mathbf{y}}
@DavidCarlisle I'd be happy with that for now :) At one time Morten and I talked about the usefulness of a state machine that could parse TeX mathematics to do things breqn does without needing active chars. In addition, you could invent new syntax like "x^abc y" == "x^{abc} y", and have hooks for these sorts of things. But as far as I know it was only just talk.
Another nice piece on Hermann Zapf, as described by those who collaborated with him, in this month's MyFonts newsletter: myfonts.com/newsletters/cc/201506.html
@WillRobertson I can't remember do you support remapping for (U)mathchardef defined tokens or do you always def a csname to a character (so you only need to change mathcodes)?
@DavidCarlisle At the moment all (I think—maybe some exceptions with large operators etc) maths symbols are accessed via either a char with the appropriate mathcode, or a macro that expands to that char. That's a very old decision that was simply made to keep things simple, but I'm happy to adopt changes—why do you ask?
@WillRobertson oh because if you want to allow chardef tokens it complicates even the simple switching between \mathbf/\symbf (as you have to recognise that's the kind of token it is, bm spends most of its time dissecting \meaning of things:-) If you can get away with not doing that it simplifies things greatly
@DavidCarlisle Ah, the first exception I've noticed is for math active chars, such as primes. They're defined with Umathchardef with the symbol expanding to the token. The reverse :) None of this is set in stone; IF I end up making everything mathactive to support breqn I guess it'll all have to be that way
@DavidCarlisle I've never quite figured out whether bm should ever work with unicode-math or not. Because bold unicode math fonts are still hard to come by, I haven't done much testing yet with what's currently working. (I do have the Lucida maths fonts, though, so I really need to.)
@DavidCarlisle Well, I guess the pmb aspect of bm is useful no matter what. But really, accessing bold symbols should (but might not yet) be a thing that just works.
@WillRobertson I got asked and started to then I realised what unicode-math was doing with math alphabets so I gave up:-) well it shouldn't really be needed I guess so long as you have a bold math font available (which doesn't seem to be the case in general) if there isn't a separate bold font available, bm can't really do much (other than what unicode-math already does for the bold alphabets, of pmb faking)
@DavidCarlisle I've heard a rumour. I know Karl expressed reservations about the Lucida one (it is a LOT of extra design work and addition complexity in design decisions), but IMO having bold maths as a standard option is something that's kinda essential in the long run.
@WillRobertson of course if you have a math font, a bold math font a bold math alphabet in the first math font and a bold text font loaded as a math \fam you have even more options for how to define \mathbf :-)
@Johannes_B, I have successfully installed now, which of them do I start to use? Is it PS_View or Text work editor, dviout that will substitute MSword, I am not seeing myself getting anything yet, the whole thing seems a bit complex even with books I have before me. Please kindly explain to me as if you are explaining to a 5 year old boy, how I can get started and running in short time. Thank you. — Gideon Ebelebe1 min ago
@JosephWright ^^ Can i just refer him LaTeX for complete Novices by @nicola?
@PauloCereda Talking about ducks, when I stopped at “lago di Misurina” for a coffee and the bathroom, I stayed for a short time looking at a mother duck with her five chicks, who were playing with a piece of plastic ribbon, while mother duck was looking over them.
@PauloCereda But it can also be freezing, especially in the mountains. Trento is very hot during summer, because the valley it's in is exactly North-South; it can be very cold in winter, for the same reason, if North winds arrive. October is just unpredictable.
@egreg thanks for your input on tex.stackexchange.com/questions/252937/… I've deleted it since you said it does not relate to tex live. I do not use the pythontex package anyway now. I just tried it once to learn about it.
@egreg: I checked the dictionary and saw gelato is also an adjective (molto freddo). We have almost the same word, gelado. But ice cream is sorvete for us. :)
@egreg -- but gelato (if i've ever actually consumed the real thing) is quite different from u.s. ice cream, and also from sherbet. ice cream can be very heavy, but the gelato i've eaten has been much lighter and finer-grained. the recipes i have for sherbet (always fruit, usually just juice, no milk) call for the mixture to be "frozen to mush", then beaten smooth, then allowed to finish freezing. if the middle step is omitted, the frozen mixture is too crystalline to be really pleasant.
@egreg When I am in USP, I go to a restaurant named Sweden to get an Italian gelato from a Brazilian attendant using an American brand machine made in China. :)
@egreg -- oh, i wasn't trying to "compare" it in the sense of saying that they're somehow alike -- they definitely are not. (i was, however, questioning whether i've ever enjoyed the "real thing". i was first introduced to something called "gelato" in palo alto, and it was definitely not ice cream! or sherbet. but i'm not sure that one can assume that what is served up in palo alto is echt gelato.)
@PauloCereda -- no, not the same. there's a rhode island delicacy known as "frozen lemonade", which i grew up knowing as "slush"; it's only partially frozen. another variation is a fruit (or other flavored) syrup poured over shaved ice (or nice clean snow). real sherbet doesn't have any liquid with it when it's first served; that only happens when it's left too long in a warm place.
@egreg -- that was a long time ago (early 1980s?). at that time, the gelato stand was near the train stop.
@PauloCereda -- "cool" is the general idea. (lemon sherbet was a "dress-up" dessert for company dinners in the summer when i was growing up. homemade sherbet served in stemmed glasses, something like martini glasses, but more delicate, with a sprig of mint, fresh from the garden, on top for decoration and additional flavor.)
@PauloCereda You should be proud of me, just written some java, it even works, clearly it's been a while as first attempt failed as I only had the jre, no compiler:-)
Is anybody behind an academic IP? I'm trying to access an online resource through my institutional login, I get in, but I keep being redirected to the login page.
It seems that University of São Paulo is a partner, so I can't understand why it's failing for me.
