Hello, I've never ventured into char before but cfr suggested that I ask in here about a better name for the quiz package that I am getting ready to submit to ctan (tex.stackexchange.com/questions/393309/…). Is something like latex2quiz too audacious?

Also, is anyone able to answer my questions about submitting to ctan, in particular about the format of the zip file and putting a configuration file in $TEXMFLOCAL/scripts/mathquiz/mathquizrc

@Andrew You want a flat zip, they don't want stuff in 'TDS format'

@Andrew If you have stuff for the TeX tree and stuff for scripts you might ask them first if they want two directories in the upload, but I suspect 'no'

It's flatish, but there are directories for scripts, latex and doc

@Andrew you can supply a separate .tds.zip file (inside the main zip) with the upacked files in TDS locations, but for ctan it's better to be all in one directory

Thanks Joseph and David. It is easy enough for me to repackage it, including adding a TDS file, but there are 55 files in total so do they really want them all in one bucket?

@Andrew You can check with them but usually 'yes'

@Andrew That's a lot of source files!

@Andrew yes (most likely but you can always mail ctan@ and ask them)

@Andrew have a look at ctan.org/pkg/oberdiek that's a lot more than 55 files all in one flat directory, but the tds.zip file puts them in multiple trees for scripts and lua and doc etc

@Andrew although as the person currently packaging that, I can't exactly recommend letting monsters get that big:-)

Thanks. I'll email first but it sounds like a flat file with a TDS included in the right approach. (There are about 10 files for the package proper and the rest are for the documentation -- all of the images in the manual are auto-generated from "example" source files. The zip file is also auto generated so there's no packaging overhead...)

@Andrew You won't need to upload those images then, only the final documentation

No the images are needed to build the manual from source - the way that I generate them uses tools from outside tex.

@Skillmon: Thank you for your awesome package!

@PauloCereda Wouldn't "penguin (right)" be better?

@samcarter Yes! :)

Does anybody have a spare reopen vote for tex.stackexchange.com/questions/393698/… ?

@samcarter done. :)

@PauloCereda Thanks!

@PauloCereda Thanks for enjoying it!

@DavidCarlisle Nice typo: been at a pub recently? “It is forcing a page break at a fixed pint”

@egreg there are nevr tipos in my ansers

@DavidCarlisle So you really meant “a pint”. Ale or lager?

@egreg see the edit log

@DavidCarlisle True, but I'm waiting for a student to discuss her thesis. ;-)

@AndrewSwann: Hi!

@PauloCereda Hand me one!

@egreg No wine today? :)

@PauloCereda By the way, doing a chat with a user yesterday I earned a beer in Hannover.

@egreg: speaking of wines, I heard that a Brazilian wine was in the top 5 list. :)

@PauloCereda Only on the train back home

@egreg ooh

@DavidCarlisle oh yes :)

@DavidCarlisle ^^ let's increase the wackiness of my comments by giving you a direct reply to my "oh no" :)

If I have an inline formula like $f^{(m)}$, the parentheses break the line spacing. What would be a good idea to avoid this?

@Bubaya probably but it depends on the font what to do, you could lower the superscript or use smaller brackets or increase the global linespacing or....

@Bubaya I think luatex has a command to force “cramped style”, which might solve the problem. Alternatively, you can lower the exponent a bit with f^{\raisebox{-1pt}{$\scriptstyle(m)$}} (modify the -1pt if need be).

@Bubaya (gotta go now, no time for followups on this one …)

@DavidCarlisle Font is lmodern. Unfortunately the document is already crowded with stuff like that (S&R should be no problem though), and I don't want to clutter the code with lots of \raiseboxes.

@DavidCarlisle Is there a possibility to make all superscript parenthesis smaller?

@Bubaya yes you could make them all smaller or as here make them all lower:

\documentclass{article}
\usepackage{lmodern}
\begin{document}


\hrule
 $f^m$ $f^{(m)}$  $\fontdimen14 \textfont2 =2pt f^{(m)}$ $f^{(m)}$

\end{document}

note fontdimens are global so I affected the last one even though I set it in the middle one in a local group

@Bubaya I can't reproduce the problem. What font size are you using?

@egreg that was my other thought: it should "just fit"

@egreg @DavidCarlisle Ah, I suppressed that it is actually $f^{(m)}_n$ :), so there are also subscripts.

@Bubaya No difference

@Bubaya you know examples should always start \documentclass and end \end{document} it leaves less to be guessed...

@DavidCarlisle I know; wait a moment please.

@Bubaya It happens if the subscript has an ascender like “d”.

@egreg @DavidCarlisle I already tried to avoid ascenders. Consider this MWE:

\documentclass[10pt]{scrartcl}
\usepackage{lmodern}
\usepackage{amsfonts}
\begin{document}
\noindent
If all indices are even, then all $\gamma_{i,i\pm1}=1$.
In this case the $\partial$-elementary symmetric polynomials
specialise to those from at $\gamma_{i,i\pm1}=1$,
which we recognise at the ordinary elementary symmetric polynomials $\varepsilon^{(n)}_m$.
The induction formula from indeed gives
\end{document}

you might get away with $\varepsilon^{\zz{n}}_m$ where \def\zz#1{\smash(#1\smash)} (it's OK here but could overprint in othercases)

Sounds reasonable. Alternatively, are there also parenthesis w/o ascender and descender?

@DavidCarlisle Am I going to get into trouble if I do the following for part of my document?

\let\oldparenl(
\let\oldparenr)
\catcode`(=\active
\catcode`)=\active
\def({\smash{\oldparenl}}
\def){\smash{\oldparenr}}

@Bubaya yes

Can somebody hint me up in this question?

drive.google.com/file/d/0BytPpTY5P_E1c1lta3FORHNrUW8/…

@DavidCarlisle OK, I think I'll use your \smashing

@Bubaya Please

@AnimeshAshish er that's just an external link (that I haven't followed) not a question, can't you ask a question on the question and answer site here?

Please take a look, I am having trouble in finding slope of the tangent to the parabola at (1,2,5) Given plane x=1 intersects the parabola z=x^2+y^2

@Bubaya -- you could also smash the superscript: $\varepsilon^{\smash{(n)}}_m$. but then you have to pay very careful attention to what's in the line above.

@barbarabeeton Mind if I poke you? I have something important to ask.

@barbarabeeton yup, I look forward to lots of fiddling :)

@PauloCereda -- okay. poke away. (by the way, do you know anything about glossaries? i'm having trouble forcing a "glossary" that is really an index, and should have been entered that way, into the required series style.)

@barbarabeeton Oh I know nothing, sorry. Maybe @NicolaTalbot can help.

@PauloCereda -- well, she's surely the expert, and court of last resort, but i thought i might be able to avoid bothering her if you could give a quick answer.

@Bubaya If you have several of these cases, it may make sense doing \linespread{1.1} in the preamble: 10/13 is not so bad.

@barbarabeeton poked.

@AlanMunn LOL

@AlanMunn ^^

@egreg that's about the limit for most fonts though

Am I missing anything important?

@DavidCarlisle I think Gitter is only useful for big projects with lots of 'live chat

@JosephWright I'd forgotten all about it but every couple of months it sends me an email saying I'm missing out. Oddly enough facebook and linked in do the same, as did research gate before I spam filtered RG:-)

@JosephWright ah thank you <3

@DavidCarlisle ^^ see

@JosephWright seems like I should have posted my git advice:-)

@DavidCarlisle Yup: at least as a beginner it's often easier

@DavidCarlisle I seem to have ended up knowing quite a bit of Git: all very odd

@PauloCereda I can see it being really useful with a full-time team or where an OS project has some full time devs. For smaller stuff though I'm not really seeing it.

@DavidCarlisle Don't get me started on RG!

@JosephWright Indeed, but it's fancy. :)

@JosephWright I love poking people at RG. :)

@PauloCereda Oh don't

@PauloCereda Sure, but for example for the whole of LaTeX you are likely better here :)

@JosephWright :D

@JosephWright you are the bloke described in the mouse over

@DavidCarlisle LOL

@PauloCereda so I guess you worked out the git reference:-)

@DavidCarlisle :)

@PauloCereda no ducks

@DavidCarlisle shocking isn't it

@PauloCereda bears eat ducks

@DavidCarlisle oh no

@DavidCarlisle uh-oh

@DavidCarlisle Regarding github.com/ho-tex/hyperref/issues/37, do you think that \textNFSSnoboundary would be okay as name? I don't want to use the suggested \textPUnoboundary as there is a similar definition in pdfx/l8uenc.def. And textnoboundary isn't imho good either, as it is more or less only an internal definition and not meant for users.

@UlrikeFischer why not use some @s ? :)

@Skillmon I'm not sure I should assume that the enc.def-files of such "virtual" fontencodings are read always with \makeatletter active.

@UlrikeFischer I think it should be OK to use @, I just looked at puenc.def and for example \DeclareTextCompositeCommand{\b}{PU}{\@empty}{\textmacronbelow}% so @ needs to be safe

@UlrikeFischer that said I'm not sure it needs to be an encoding specific command, if it is only used as \let\noboundary\zzznoboundary when you know the PU encoding is going to be in force, it could just be \def\zzznoboundary{..} couldn't it?

@DavidCarlisle I'm not sure. Imho pdfx will want to remap it to its l8u encoding. But \textzzznoboundary would be a good name too, then everyone will know who to blame if something goes wrong ;-).

@UlrikeFischer yes probably safer to make it an encoding command,

@DavidCarlisle But puarenc.def is actually only an extension of puenc.def, so it is quite possible to do \usepackage[unicode]{hyperref}\input{puarenc.def}. And while I used a lot @ in the chess encodings, since I saw you do \input{tuenc.def} in an example I'm not sure if it was a good idea ...

@UlrikeFischer not so keen on nfss in the name (since it's explicitly not fonts) how about \textencodingnoboundary ?

@DavidCarlisle fine with me. Or \pseudoencodingnoboundary?

@UlrikeFischer you might want to use if with a real encoding I suppose (eg TU might want to map it to Unicode rather than the tex primitive in some cases? (not sure)

@DavidCarlisle OK. So \textencodingnoboundary? Some uppercase letters somewhere?

@UlrikeFischer it's long enough probably to be safe (I realize as you're German it looks hardly long enough to be a word) but \textEncodingNoBoundary ?

@DavidCarlisle That looks good. Imho "internal" enough.

Hello!! Could someone of you take a look at my question: tex.stackexchange.com/questions/393849/… ?

TUGBoat has arrived

@JosephWright a couple of days ago

@PauloCereda vvv

@JosephWright someone called Mittelbach writing about table macros, that one?

@DavidCarlisle Yup, that's the one

@DavidCarlisle Come to home so I never see them until Friday evening

@JosephWright at some point you have to let "home" == the place where you spend more than half the week:-)

@DavidCarlisle On a life average, I am safe ;)

@DavidCarlisle Also, family are here

@JosephWright damn, been defeated by mathematics. From a chemist.

@JosephWright which means another week went by. I'll make another attempt to make some coherent proposal about filenames this weekend.

@DavidCarlisle Ah, good

@DavidCarlisle Speaking of which, I will get on to Trinity this weekend for UK-TUG

@DavidCarlisle Do trains qualify?

@egreg :)

Boris's 'editorial' is good

@JosephWright it seems to be the day for merge commits in pull requests. Does github's "squash and merge" make it all into a single commit anyway so the multiple commits in the PR don't matter or should I be doing the cherry picking stuff (not that the git history is so important here) github.com/ho-tex/hyperref/pull/45 (@UlrikeFischer)

@DavidCarlisle Yes it should do: have to say I hadn't thought of squash-and-merge (I tend to like to leave people's commits 'alone', but there is nothing wrong with the one-commit-per-issue model)

@JosephWright OK I'll eyeball @UlrikeFischer's code and then do that. I suppose I ought really check it all out to a clean repo and test something but I haven't any real tests for hyperref anyway

@egreg @egreg @egreg @UlrikeFischer has kindly produced a slide for your next talk

github.com/ho-tex/hyperref/pull/45/commits/…

ooops:(

(./README.tex
LaTeX2e <2017-04-15>
(using write cache: /usr/local/texlive/2017/texmf-var/luatex-cache/generic)(usi
ng read cache: /usr/local/texlive/2017/texmf-var/luatex-cache/generic /home/dav
idc/.texlive2017/texmf-var/luatex-cache/generic)
luaotfload | main : initialization completed in 1.728 seconds
Babel <3.13> and hyphenation patterns for 1 language(s) loaded.
)
*

@DavidCarlisle Ooops

@DavidCarlisle Too late!

@JosephWright I really think I should drop all the generation of README and ChangeLog in html and pdf versions it failed there as the xslt is version 1 and I've just upgraded to a version 3 engine, an dit's dropped 1.0 compatibility:-)

@DavidCarlisle Sounds reasonable

@JosephWright and after fixing that it dies on

! Undefined control sequence.
l.50 \begin{itemize}\GobbleTwocommit
                                   b312fe2176a4ec2c4b036ae8c04f489dfc82b6bb

@DavidCarlisle Ergh

because my cunning plan of dropping in the fragment of git change log clearly broke the (unwritten) rules about the format allowed in the changelog xml source. I'm going to kill it:-)

@DavidCarlisle Seen article in TUGboat about xdvipdk?

@JosephWright no, will look

@DavidCarlisle Page 197

ooh might make dviluatex usable...

@DavidCarlisle Yup

@DavidCarlisle Will ask on the TL list if it is 'coming soon'

oh its on github from vtex?

@DavidCarlisle Haven't checked yet

@JosephWright GPL3 github.com/vtex-soft/texlive.xdvipsk

@DavidCarlisle Ah: tug.org/svn/texlive/trunk/Build/source/texk/xdvik

@JosephWright that xdvi

@DavidCarlisle Oh, so it is