« first day (2298 days earlier)      last day (2628 days later) » 

12:00 AM
@egreg yes that occurred to me later, that it probably worked, are you still up? care to try some xetex?
anyone else with a newish xetex care to try
 
Thanks guys :)
 
12:19 AM
@DavidCarlisle Not on the bleeding edge.
 
can you try
\font\m= "[latinmodern-math.otf]/OT:script=math;language=DFLT;" at 10pt
\font\sm= "[latinmodern-math.otf]/OT:script=math;language=DFLT;" at 7pt
\font\ssm= "[latinmodern-math.otf]/OT:script=math;language=DFLT;" at 5pt

\textfont0=\m
\scriptfont0=\sm
\scriptscriptfont0=\ssm


\Umathcode`.= 0 0 `.

\hrule
$$
\hbox{$\left)\vbox to15pt{}\right.$}_{\smash.}
\hbox{$\left)\vbox to15pt{}\right.$}_{\smash.}
\hbox{$\left)\vbox to15pt{}\right.$}_{\smash.}
$$
\hrule
$$
\hbox{$\left)\vbox to15pt{}\right.$}_{\smash.}
 
@DavidCarlisle Same for me
 
@egreg thanks. I'll take to xetex list
 
@DavidCarlisle For once LuaTeX behaves.
 
@egreg you get interesting effects by changing the size of the scriptscript font so i suspect it's a failing in \mathchoice somehow
@egreg yes (as you may have seen on unicode-math github)
@egreg with \font\ssm= "[latinmodern-math.otf]/OT:script=math;language=DFLT;" at 25pt
@barbarabeeton ^^^^
 
1:19 AM
@DavidCarlisle -- ick. does it do this only on .otf fonts, or are there other examples?
 
 
8 hours later…
8:51 AM
@DavidCarlisle Works ok in miktex. Imho it is part of a number of questions and bug reports (e.g. sourceforge.net/p/xetex/bugs/75/#b012) which should be fixed in the development version.
 
@barbarabeeton otf I think
@UlrikeFischer ah I suppose I could try building that again, not built xetex for a while
 
 
1 hour later…
10:11 AM
Not a TeX question, but maybe someone here can answer
Is there any other font than Cambria Math that works with Microsoft's new equation editor in PowerPoint for Mac? I am hoping for a Computer Modern lookalike.
Latin Modern Math does not work.
 
@Szabolcs no idea about powerpoint but I'd have expected any font with a math table to work. so TeX Gyre math fonts or xits-math or stix2 or asana math for example. What happens with lm-math?
 
@DavidCarlisle \langle and \rangle look like a bold P and Q. ( and ) look like narrow roman F and G.
I got the font from here: gust.org.pl/projects/e-foundry/lm-math
@DavidCarlisle Does Latin Modern Math have those math tables? If yes, I will just give up, as other fonts from the same sources are likely to be similar. If not, I guess I just need other fonts.
 
@Szabolcs yes, but how are you entering \langle into powerpint?
 
@DavidCarlisle literally as \langle. I like Microsoft's new equation editor a lot because it understands a lot of TeX.
I do not have to learn a new way to enter math.
Nor do I have to use the inconvenient GUI buttons.
Plus, it makes it easy to inline a lot of math into normal running text.
 
@Szabolcs I'd trust it to understand the font more than I trust it to understand tex what happens if you just use ⟨(X)⟩
 
10:23 AM
@DavidCarlisle This time it refuses to change fonts (it claims to have changed it but the look doesn't change). I think I'll just be happy that I can enter equations at all and I'll forget about changing the font. Thanks for the help!
 
@UlrikeFischer thanks for following up on xetex list
 
@DavidCarlisle I also added a reply to github -- no need that people spent to much time on an hopefully already resolved problem.
 
@UlrikeFischer yes there as well. I'm trying to build from the texlive head at present but cygwin is a bit delicate the build doesn't always work (They always get it working in the end by the full release) so I'll see what happens...
 
@Szabolcs I can get the font to load but not all of the glyphs are right: what's hard to say of course if if the issue is at the font end ...
 
11:08 AM
Quack from SP!
 
yo'
@PauloCereda oh! :-)
 
@PauloCereda A good shake makes for a nice active day!
 
@egreg :)
 
11:39 AM
I voted for @DavidCarlisle in the TUG board election!
2
 
12:02 PM
@PauloCereda ducks are, as we know, very untrustworthy.
 
 
1 hour later…
1:16 PM
hi *
is there a way to get a list of all declared Unicode codepoints?
i.e. ASCII, anything I define with \newunicodechar myself, anything packages define, etc.
background: I need to “print” arbitrary text as PDF, I will do that via LaTeX, and for that, I need to know which chars to strip (and which to encode with \% and \textbackslash and so on, but those are few)
 
1:47 PM
@mirabilos simpler to use xetex or luatex where any codepoint in the font will work without needing to be declared
 
@DavidCarlisle no, we use pdflatex throughout everything (for various reasons)
also, then I’d need a list of “any codepoint in the font”…
 
@mirabilos well then the answer is no, really you can look at utf8enc.dfu to see all the ones defined by default bu [utf8][inputenc} but if any extra definitions made later there is no list
 
hmmh
can I figure them out somehow?
hm maybe redefine \newunicodechar but I don’t know if everything uses that
I also wonder which ones the font (inconsolata) defines
 
@mirabilos if you are using pdflatex then it's just as defined by T1 encoding, whatever the font is
 
no, it loads three encodings
\RequirePackage[LY1,TS1,T1]{fontenc}
 
1:59 PM
@mirabilos yes I was going to say or whatever other encoding you are using. but point is it doesn't depend on the font. (no point in using LY1 there. I suspect)
 
hm, not there probably (we do use some LY1 chars in other documents, so best to keep everything synchronised though)
hm
can I tell pdflatex to output the entire source, like gcc with “preprocess only”? then I could easily grep for newunicodechar, DeclareUnicodeChar, etc.
 
@mirabilos I wrote LY1 but are there any characters in LY1 not in T1+TS1 ?
 
unsure
I know we currently use LY1 for € and …
 
@mirabilos no, you can redefine each of those commands to write to the log if they do not already.
 
(in lstlistings, probably not applicable to the new usecase)
@DavidCarlisle yeah… are there other commands than these two I need to know about?
 
2:03 PM
@mirabilos probably not
@mirabilos that is in TS1
@mirabilos newunicodechar package has a verbose option to write to the log
 
hm ok
 
@mirabilos is it not easier just to process the document once catch any "this character not defined" errors, then define those characters and re-process?
 
-recorder seems to be what I want
@DavidCarlisle no because there is no 'this document'
use case is a java web application that can output a PDF catalogue generated from information the user previously put into a database
latex happens to be the easiest way to do that
(also, won't be able to define those characters, need to strip them out… or replace by making a nice box with the hex code in them)
 
@mirabilos well that's what I'd do just redefine the utf8 undefined character error message not to make an error but to print a hex box.
 
2:20 PM
ah
now that’s a plan
(and a warning, and then we can just use that in general)
I’ll experiment with that
hm, since I have you here:
utf8enc.dfu has \DeclareUnicodeCharacter{2190}{\textleftarrow}
our site .sty file has \newunicodechar{←}{$\leftarrow$}%
when I remove it, the arrows don’t show up… why is that, font doesn’t have them, or doesn’t declare them correctly?
font, in this case, is a .ttf font with appropriate .enc .tfm and .fd written without a real clue ;)
 
@mirabilos in UTF-8 is 0xE28690; if you use OT1 encoding, you get nothing, because there are no glyphs in slots "E2,"86 and "90. If you add \usepackage[T1]{fontenc} you'll see some strange glyphs popping out.
 
@mirabilos All the codepoints declared with \DeclareUnicodeCharacter are recorded in the log-file as defining Unicode char U+2026 (decimal 8230). newunicodechar doesn't record new definitions but it would be easy to add this. Ask @egreg. Beside this you naturally can get a list of all unicode code points somewhere and then process it to find out which one are undeclared.
\documentclass{article}
\usepackage[utf8]{inputenc}

\begin{document}
\makeatletter
\gdef\printUnicodeNum#1:#2\relax{%
 (U+\expandafter\UTFviii@hexnumber\expandafter{%
                     \the\numexpr\decode@UTFviii#2\relax})
}
\makeatother

\ifcsname u8:\detokenize{ä}\endcsname YES! ä \printUnicodeNum u8:ä\relax \else NO!\fi

\ifcsname u8:\detokenize{Ѡ}\endcsname YES! \else NO! \printUnicodeNum u8:Ѡ\relax\fi

\ifcsname u8:\detokenize{Ⰴ}\endcsname YES! \printUnicodeNum u8:Ⰴ\relax\else NO! \printUnicodeNum u8:Ⰴ\relax\fi
 
@UlrikeFischer newunicodechar doesn't go through the step UTF-8 to Unicode point; it just defines \u8:<bytes>
 
@egreg no, for some reason, nothing is shown, as if it expands to empty space
@DavidCarlisle seems like I “just” need to copy the \UTFviii@defined definition and change it appropriately, for that
 
@egreg Yes I know. But it does \@ifundefined{u8:\nuc@tempa}{} and could easily add a message to the log there.
 
2:32 PM
and do something like \UTFviii@splitcsname for decoding
 
@mirabilos If you load \usepackage[utf8]{inputenc}, then would give an error.
 
@egreg well it doesn’t ;)
but it’s not a real problem, I was just wondering
I’ve had other problems with mapping from .ttf to Teχ
 
2:48 PM
@mirabilos utf8enc.dfu isn't loaded, you need to use textcomp package (and so load ts1enc.dfu) if you want a definition for 2190.
@DavidCarlisle What is the purpose of utf8enc.dfu in latex/base?
 
@mirabilos the error comes from
\def\UTFviii@defined#1{%
  \ifx#1\relax
      \PackageError{inputenc}{Unicode\space char\space\expandafter
                              \UTFviii@splitcsname\string#1\relax
                              \MessageBreak
                              not\space set\space up\space
                              for\space use\space with\space LaTeX}\@eha
  \else\expandafter
    #1%
  \fi
}
 
@DavidCarlisle yeah, already found it
 
so you can just define that to do whatever you want instead of Package Error (the human readable hex number is \UTFviii@splitcsname\string#1\relax
 
\makeatletter%
\def\UTFviii@hack@hex#1{%
 \ifnum#1>65535%
  U-000%
 \else\ifnum#1>4095%
  U+%
 \else
  U+0%
 \fi\fi%
 \expandafter\UTFviii@hexnumber\expandafter{#1}%
}%
\def\UTFviii@hack@decode#1:#2\relax{%
 \expandafter\UTFviii@hack@hex\expandafter{%
  \the\numexpr\decode@UTFviii#2\relax%
 }%
}%
\def\UTFviii@defined#1{%
 \ifx#1\relax
  \PackageWarning{inputenc}{%
   Unicode\space char\space\expandafter\UTFviii@splitcsname\string#1\relax%
   \space not\space set\space up\space for\space use\MessageBreak with\space%
 
@mirabilos was working so hadn't checked the log here:-)
 
2:57 PM
this is what I already have, now I need to make that into a real package, add the box stuff, etc.
 
@mirabilos I may make something like that a standard option at some point.
 
ah nice
 
@UlrikeFischer a list for Frank and Chris to see what the coverage was:-)
 
@DavidCarlisle ;-)
 
3:12 PM
ok, making a box from the ¿? stuff (with 2x2 or 2x3 nybbles in it) will have to wait, as I’ll have a workshop later, but thanks @DavidCarlisle for the inspiration, I’m now confident I can actually do it! ;)
quick question: FrutigerLTStd-Cn (found in a template PDF the customer gave me) is likely a commercial font, right?
that is, no pdflatex integration currently exists
 
@mirabilos font will be commercial there are some latex setup for commercial fonts (without the actual fonts) on ctan eg ctan.org/tex-archive/fonts/truetypemetrics/frutiger?lang=en (not sure if that is the right one but..)
 
@DavidCarlisle hm I found tex.stackexchange.com/q/254996/43807 which is close
seems to be '57 Condensed'
argh I really gtg, but good to know
 
4:01 PM
@JosephWright will we be getting the new header do you think (as on SO)
 
yo'
@DavidCarlisle that's so f****** huge! Zero added value, a lot of added space.
 
@yo' I was reading their blog about it, they are worried that people google straight to an answer and miss the added stackexchange value so they thought they stick a massive great header in the way:-)
 
4:26 PM
@DavidCarlisle, @egreg -- you've got mail; an amsthm/hyperref-specific question on which i need advice for an author. (thanks.)
 
@barbarabeeton Why not using cleveref? Can you add some example code?
 
@egreg -- it's a specific request from an author to use \autoref. i will try cleveref to see if it gives the result he wants. he wasn't nice enough to provide an example, so i'm having to invent one myself. (this is typical of my "day job".) thanks for suggestion.
 
4:42 PM
@yo' Don't worry, one can simply disable it
 
@barbarabeeton aliascnt can do the job. Much more clumsily than cleveref. Mailed both solutions.
 
@egreg -- thanks for suggestion. i've gotten an example using cleveref that works, and it's confirmed by your mail.
 
@samcarter Does nothing for me
 
4:59 PM
Voted for TUG Board...
 
@egreg Just tested a bit more and on list like stackoverflow.com/questions it works, but not on individual questions. Strange.
 
@samcarter Oh, now I saw the difference; with the “disable” on, you just see the question list; if off, you see the usual tabs “newest”, “featured” and so on.
 
@egreg I hope the "disable" feature will be improved. But the creation of custom tabs sounds nice.
 
yo'
5:15 PM
@samcarter good to hear, thanks
 
5:57 PM
@barbarabeeton my thought as well that autoref might not be first choice these days, will look later
 
6:33 PM
@egreg thanks:-)
 
6:51 PM
@samcarter: He(!) is in chat sometimes....
 
@ChristianHupfer Just never when I'm here - at least I don't get a tab completion for the name.
 
@samcarter Ah, I see. So he wasn't here for some days apparently.
 
 
2 hours later…
9:14 PM
There is no one in chat tonight... Have they finally followed the advice of karlkoeller, and they are celebrating St. Valentine's Day with their wives/husbands/girlfriends/boyfriends?
 
yo'
@CarLaTeX And then there's people like me who work still, at 10:15pm
 
@yo' oooh, poor!
 
yo'
@CarLaTeX I live this way
 
@CarLaTeX -- some of us are still at work, although not as late as @yo'. (but leaving in about an hour for exactly the kind of celebration indicated.)
 
@barbarabeeton Brava Barbara!
 
yo'
9:18 PM
@barbarabeeton well, here, St. Valentine has zero tradition. Whence, it's all purely about commerce. You can buy plushy hearts, chocolate hearts, whatever hearts, but not love
@CarLaTeX I'm afraid it's neither AMS nor TUG who's out for celebration, so you better speak about barbara
 
@yo' Yes, it's the holiday for florists and pastry shops!
 
@yo' -- oh, there's a lot of commerce involved here too, but our local french restaurant always has a splendid menu on special occasions, so how could one not try to take advantage of the opportunity?
 
yo'
@barbarabeeton right, if you have someone to be accompanied by...
 
@yo' -- quite true. and now that he's out of the hospital, who better?
 
@barbarabeeton Oh yes, you have to celebrate!
 
yo'
9:25 PM
@barbarabeeton oh, well, I'm sure the question in your case is not who, but rather, being happy he's with you at home!
 
@CarLaTeX Wondering about writing some siunitx code: does that count?
@CarLaTeX Seriously: my mum's birthday, and this year is a 'big' one
 
@yo' -- yes, that's quite true! teddy bears are nice and cuddly, but don't really generate much warmth.
 
@JosephWright giving your mum some siunitx code for her birthday? how kind.
 
@JosephWright Happy birthday to her!
 
@barbarabeeton they burn well
 
9:27 PM
@CarLaTeX Indeed :)
 
@JosephWright like my new xparse implementation?
 
@DavidCarlisle -- not safe indoors, and i wouldn't send a teddy bear out in the snow.
 
@DavidCarlisle Oh indeed
 
@DavidCarlisle Do you hate teddy bears like ducks?
 
@JosephWright not an entirely frivolous suggestion for once.
 
yo'
9:29 PM
@CarLaTeX he doesn't hate them, that's a huge misunderstanding
 
@yo' Hahahaha
 
@CarLaTeX no just replying to barbara's comment. Ducks taste better than teddy bears though,
 
@DavidCarlisle Indeed, I have never heard about Chinese teddy bears...
 
@CarLaTeX ^^^^^ roast teddy...
 
@DavidCarlisle OMG!!!!!
 
9:34 PM
Jun 9 '16 at 19:38, by David Carlisle
@JosephWright sounds very technical, I just thought whatever stupid phrase you type in, google will always find a suitable image
 
@DavidCarlisle That's why they invented it!
 
10:08 PM
@JosephWright: I think, I can finally 'kill' the text in my nomination answer for T.F. Sturm -- he did not answer apparently.
 
10:19 PM
@DavidCarlisle Another hidden feature in longtable. ;-)
 
@ChristianHupfer 'Fraid not, no
 
@JosephWright Well, I could keep the deleted 'answer' -- it does no harm anyway. But it is sad ....
 
@egreg not that hidden, you and I both found it easily enough
 
@DavidCarlisle This is a scaring image. The teddy bear already looks like laying on his death bed ... :-(
 
@ChristianHupfer That's exactly what I thought!
 
10:29 PM
@CarLaTeX @DavidCarlisle is a fan of animals;-) Ask him about the family cat in the House Of Carlisle ;-)
 
@ChristianHupfer I hope @DavidCarlisle don't want to cook the cat, too!
 
@CarLaTeX No, I don't think so...
 
@ChristianHupfer Gott sei Dank!
 
@CarLaTeX :D
 
@CarLaTeX ^^
 
yo'
10:38 PM
@DavidCarlisle that's T&J inverted ....
 
@DavidCarlisle Apparently, Carlisle is some medieval English for 'Cat Cooker' ;-)
2
 
@DavidCarlisle hahaha My cats also used to sleep in unthinkable places!
 
@CarLaTeX some of the other images google turned up were less likely to be misinterpreted as sleeping, but I thought them a bit gruesome for a chat room full of cat owners:)
 
@DavidCarlisle I'm surprised of your sensitivity!
 
10:44 PM
@CarLaTeX: My late mother had a cat once that loved to sleep on wet laundry, coming fresh out of the washing machine and put into the basket before the clothes could be hung on the rotary clothes dryer. When none of us was quick enough the cat was in the basket :D
 
@ChristianHupfer OH NOOOOOO! Not in the dryer!
 
@CarLaTeX There is some misunderstanding
 
@ChristianHupfer Fiuuu, only in the basket...
 
@CarLaTeX: I meant this thingy:
They are quite popular in Germany .... and no, the cat was not hung on it ;-)
 
@ChristianHupfer Sorry, I thought of a dryer machine! Hahahah
 
10:50 PM
@CarLaTeX No, no...
 
@DavidCarlisle Do you see any possible side effect?
 
@ChristianHupfer However, a dog of mine liked to pull down the clothes from the drying rack!
 
@CarLaTeX Naughty dog! I don't assume that the dog used the iron and put the clothes back to the closet? ;-)
 
@ChristianHupfer Hahahahah, unfortunately not!
I'm going to sleep, good night to everybody!
 
@CarLaTeX Good night, sleep well
 
11:09 PM
@egreg I was wondering if doing \everypar{} would be better or different, most likely if it is true then it is set from
\def\@afterheading{%
  \@nobreaktrue
  \everypar{%
    \if@nobreak
      \@nobreakfalse
      \clubpenalty \@M
      \if@afterindent \else
        {\setbox\z@\lastbox}%
      \fi
@egreg and just setting the flag back leaves clubpenalty etc not reset
 
11:19 PM
@DavidCarlisle Probably so; \everypar is the same also in all alignment cells, by the way, if longtable immediately follows a section title. Maybe you could add \everypar{} also in \longtable, although in a parbox it doesn't do much harm (except if somebody wants to indent their X cells).
@DavidCarlisle Anyway, it's not sufficient to do \everypar{}. \@nobreakfalse is needed nonetheless.
 
@egreg tabbing is probably the same
@egreg probably true
 
@DavidCarlisle No, tabbing behaves
\section{Title}\begin{tabbing}a\=b\end{tabbing}\showthe\everypar gives
> {\setbox \z@ \lastbox }\everypar {}\@endpefalse .
 
@egreg so a unique lt feature then:-)
 
@DavidCarlisle You share it with listings
16
Q: How do I force \section's to start on a new page if they don't fit on current page?

Peteris KruminsSo I am writing a programming book, and I use \section for each topic. For example, I'd write, \section{Another way to double-space a file} \begin{lstlisting} awk 'BEGIN { ORS="\n\n" }; 1' \end{lstlisting} BEGIN is a special kind of pattern which ... \section{Triple-space a file} .... Some...

 

« first day (2298 days earlier)      last day (2628 days later) »