3:39 AM
@ChristianHupfer tex.stackexchange.com/questions/394697/… :):):) and the upvote is NOT from the OP!
@ChristianHupfer I don't know if to catalogue it under "devalued answers" or "ungratefulness" series :):):)

3 hours later…
6:39 AM
Don't know how to compile this tex.stackexchange.com/a/394786/67761

@VincentMiaEdieVerheyen context filename

3 hours later…
9:17 AM
@DavidCarlisle ooh

@PauloCereda Morning

@JosephWright Morning, Joseph!
Fletcher room, here we go!

@PauloCereda Yup, it's the 'dungeon' as usual

@JosephWright there's pizza, so we will be fine. :)

@PauloCereda Well I've not specified the form of lunch ...

9:25 AM
@JosephWright oh no :) @DavidCarlisle ^^

@PauloCereda Don't worry, I'll just book Pizza Express ;)

@JosephWright ooh

@JosephWright classy!
Besides, I love this bit so much:
> Beyond pizza, there are cookies – we use the non-edible kind on this site to help you get the most out of it. If you’re happy with that, keep on browsing.
2

@JosephWright Apart from grated parmesan cheese and the pork, those pizzas seem (almost) acceptable :)

9:35 AM
@CarLaTeX the best part of a pizza is the pizza. :)

@CarLaTeX I never go anywhere near such things on pizza, don't worry

@CarLaTeX Das Bakon
It's pig in German. :)

@PauloCereda I don't dare imagine how is the taste of the knead (I don't know the correct English word for "impasto") :)
@PauloCereda Ducks are polyglot

@CarLaTeX Meow.

@JosephWright Good :)

9:38 AM
We're making progress: An OP found a manual :) tex.stackexchange.com/questions/394958/…

@TeXnician ooooh that's great

@PauloCereda In Google Translate maybe, but not in German...

@TeXnician <3

@TeXnician @PauloCereda Ducks are not polyglot

@CarLaTeX oi you are mean

9:40 AM
@TeXnician surely you are not suggesting google translate can be wrong? How could that be?

@PauloCereda Oh no, I don't have a "you're not mean" message!

@CarLaTeX I will provide one later. :D

@CarLaTeX well, AFAIK I've got neither of the two messages :-)

@PauloCereda <3

@CarLaTeX <3

9:43 AM
@yo' How can it be possible?

Jan 12 '16 at 12:15, by David Carlisle

@DavidCarlisle oh no

@yo' You're right, I should ask @PauloCereda!

@DavidCarlisle Let's say it's not wrong, because Google Translate is always the right tool to make fun of languages and language skills ;)

9:48 AM
anyway I (finally :-( ) got the order of service (songs) for tomorrow, so it's time to go and quickly rehearse them

@yo' have a great day, Tom!
@DavidCarlisle: what are you going to talk in our meeting?

10:41 AM
@PauloCereda "on the benefits of using pineapple as a pizza topping"

@DavidCarlisle ooh
@JosephWright: are we expected to talk about something? :)

@PauloCereda Not necessarily

@JosephWright oh

11:09 AM
@PauloCereda usually yes, but you're exceptionally allowed to talk about nothing.

@yo' ooh a logical conundrum
@JosephWright: will try to come up with a talk

11:36 AM
@DavidCarlisle I still can't figure out on my own how to make it work tex.stackexchange.com/questions/394981/…

12:10 PM
0

When I compile this code: \documentclass{memoir} \usepackage{amsmath} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{mathtools} \begin{document} \begin{equation*} \begin{pmatrix*}[c] 0&\dots&0&1&\mu_{1,j+1}&\dots&\mu_{1,k-1}&\mu_{1k}&\mu_{1,k+1}&\dots&\mu_{1n}\\ 0&...

That one should probably marked duplicate of
30

The code below works, but when I add 10 more columns, it does not. So, how can I add ten more columns to the matrix below? \documentclass{article} \usepackage{amsmath} \begin{document} $$\begin{bmatrix} 1 & 0 & 0 & 0 & 0 & 1 & 0 & 0 & 1 & 0 \end{bmatrix}$$ \end{...

or
50

I'm using bmatrix environment from the amsmath package, and I'm trying to do something like this: \documentclass{article} \usepackage{amsmath} \begin{document} \begin{bmatrix} 1 & 2 & 3 & 4 & 5 & 6 & 7 & 8 & 9 & 10 & 11\\ \end{bmatrix} \...

1 hour later…
1:47 PM
@CarLaTeX ooooh
Go goose!

1:59 PM
@Sebastiano Please don't edit WYSIWYW to WYSIWYG, these are two different things and your edit invalids the question ask in tex.stackexchange.com/questions/394829/…

@PauloCereda Very close to Milan (but not ducks) :)

2:35 PM
\usepackage{xparse}
\usepackage{hyperref}

\ExplSyntaxOn

\NewDocumentEnvironment {PersonInformation} { m O{} } {
\begingroup
\tl_if_blank:nTF { #2 }{A #1}{B #2}
\endgroup
}{
\href{a.beautiful.url/?searchid=#1}{Click on me}
}

\ExplSyntaxOff
MWE.
:
\documentclass[a4paper]{article}

\usepackage{xparse}
\usepackage{hyperref}

\ExplSyntaxOn

\NewDocumentEnvironment {PersonInformation} { m O{} } {
\begingroup
\tl_if_blank:nTF { #2 }{A #1}{B #2}
\endgroup
}{
\href{a.beautiful.url/?searchid=#1}{Click on me}
}

\ExplSyntaxOff

\begin{document}

\begin{PersonInformation}{Test}{Test}
\end{PersonInformation}

\end{document}

@VincentMiaEdieVerheyen #2 is optional, so should be given in square brackets

2:49 PM
apart from that you want Click ~ on ~ me otherwise it will be Clickonme as spaces are ignored in expl3 code sections.

2 hours later…
4:31 PM
@JosephWright Thanks, but it still didn't compile correctly. No hyperlink.
@DavidCarlisle Thanks David. That's good to know. The following still doesn't show any clickable hyperlink:
\documentclass[a4paper]{article}

\usepackage{xparse}
\usepackage{hyperref}

\ExplSyntaxOn

\NewDocumentEnvironment {PersonInformation} { m O{} } {
\begingroup
\tl_if_blank:nTF { #2 }{A #1}{B #2}
\endgroup
}{
\href{a.beautiful.url/?searchid=#1}{Click ~ on ~ me}
}

\ExplSyntaxOff

\begin{document}

\begin{PersonInformation}{Test}[Test]
\end{PersonInformation}

\end{document}

@VincentMiaEdieVerheyen I have a hyperlink with that code

@VincentMiaEdieVerheyen I get a link too

4:47 PM
@VincentMiaEdieVerheyen @JosephWright @DavidCarlisle I get a link but not the correct one, even if I add the protocol http, the link has backslashes instead of slashes and the searchid is "Test.pdf" and not pdf (I won't try to figure out, what hyperref does with a colon with catcode letter ;-).)

@VincentMiaEdieVerheyen You shouldn't use \tl_if_blank:nTF, but \IfNoValueTF and with o argument type.
\documentclass[a4paper]{article}

\usepackage{xparse}
\usepackage{hyperref}

\NewDocumentEnvironment {PersonInformation} { m o } {%
\IfNoValueTF{#2}{A #1}{B #2}%
}{%
}

\begin{document}

\begin{PersonInformation}{Test}[Test]

\end{PersonInformation}

\end{document}

@egreg He actually want to test for blank arguments, see tex.stackexchange.com/questions/394991/…
@VincentMiaEdieVerheyen Do get a correct url-link you should add the protocol before the url, without it hyperref will this that you are trying to access a local file. And you should better not use \href is an expl3-context as the colon as letter confuse hyperref. This here works for me:
\documentclass[a4paper]{article}

\usepackage{xparse}
\usepackage{hyperref}

\ExplSyntaxOn
\cs_set_eq:NN\mytlifblankTF\tl_if_blank:nTF
\ExplSyntaxOff

\NewDocumentEnvironment {PersonInformation} { m O{} } {
\begingroup
\mytlifblankTF { #2 }{A #1}{B #2}
\endgroup
}{
\href{a.beautiful.url/?searchid=#1}{Click ~ on ~ me}
}

\begin{document}

\begin{PersonInformation}{Test}[Test]
\end{PersonInformation}

\end{document}
Be aware that special chars in the url can lead to problems.

5:03 PM
@UlrikeFischer That's a very different thing; if an argument is declared optional, omitting it is the right way to go.

@egreg @Johannes_B As you favour not closing questions like the polyglossia documentation one would you mind vote for reopening? I would write up an answer.

@UlrikeFischer I click on it and xpdf says Couldn't open file 'a.beautiful.url/?searchid=Test.pdf' oh, it dropped the http, I see.

@egreg I know, but if he wants it ...

5:22 PM
@TeXnician Done. Reading documentation is one thing, using google is another. ;-)

@TeXnician I voted for reopening, that's as much I can do.
@TeXnician I believe that \setmainlanguage was added when fontspec changed \setromanfont into \setmainfont, for uniformity.
I think it's semantically better than \setdefaultlanguage.

5:40 PM
@egreg Sounds good
Wondering what to talk about at UK-TUG meeting: thinking l3regex ...

6:18 PM
@JosephWright colour?

@DavidCarlisle Perhaps
@DavidCarlisle Should be ready by then

@egreg Ah, nice to know, thanks. But I just referenced the order from the docs.

@DavidCarlisle I guess I'll be able to say 'we are sorting colour and graphics, more-or-less a straight (code level) reimplementation of the 2e code'

6:53 PM
@JosephWright Short talk ;-)

@UlrikeFischer :)
@UlrikeFischer Have you met the UK-TUG regulars :)

@JosephWright ? Are they only waiting for the pizza?

@UlrikeFischer We tend to find a way to have long discussions over all sorts of things ;)

@UlrikeFischer mostly they are hoping that they will be able to escape and see daylight again one day
2

@DavidCarlisle :)
@UlrikeFischer It's an improvement on the previous food provision
@DavidCarlisle Let's hope that the venue doesn't fall to pieces this year!

7:01 PM
@JosephWright We had spanish tapas at the Dante meeting two week ago ;-).

@UlrikeFischer Oooh
@UlrikeFischer Probably have a larger turnout than us
@UlrikeFischer Also probably know in advance what the talks might be (I can guess but ...)

7:24 PM
@DavidCarlisle Both of them? ;-)

@JosephWright We were around 26. And quite a number stay overnight. So there are also dinners on the evening before (this time at my home ;-)) and after the meeting (pizzeria, but I think nobody had pineapples).
@JosephWright There is timetable and you know the titles of the talk, but it is not always clear what they are about ;-)

@UlrikeFischer Ah, cool
@UlrikeFischer More than we know in advance :)
@UlrikeFischer We tend to find out what people will talk about when they start!
@UlrikeFischer I can predict, of course: I'll take about something (l3regex/l3color), Simon Dales will talk about Lua, Kaveh will talk about something publishing-focussed, @DavidCarlisle might talk about something if I lean on him, Jonathan Fine will probably talk about grand plans, ...

7:45 PM
@JosephWright I could do a line-by-line run through of the indentfirst code.

@DavidCarlisle :)

@JosephWright if I'm there I could possibly talk about spaces and funny foreign characters

@DavidCarlisle The latter was what I was wondering about, yes (unless you mean a certain green square!)
I guess @PauloCereda might speak too, plus perhaps Phil Molyneux
@DavidCarlisle 'How do we escape the dungeon?' ;)

3

8:01 PM
@AlanMunn Cf. recent discussion on TeXhax

@AlanMunn Ich konnte mein ganzes Gespräch auf Deutsch geben

@DavidCarlisle Heveriko fa tokony handefa fiteny mahaliana kokoa ianao.

@AlanMunn আপনি জার্মান আকর্ষণীয় খুঁজে না?

@DavidCarlisle අදහස් දැක්වීමෙන් වලකිනවා.

@AlanMunn Mi rammarico dei non linguisti che dovranno incollare ogni riga in Google Translate per seguire la conversazione.

8:17 PM
@DavidCarlisle Uma boa idéia.

@DavidCarlisle "Mi rammarico dei non linguisti" means that you regret the non-linguists...

@CarLaTeX :)

@CarLaTeX Being a linguist, I endorse this sentiment. :)
3

@AlanMunn :)

@CarLaTeX I know I back translated it and that's what it said. Near enough (I started with "pity":-)

8:23 PM
@AlanMunn I think it might be lost on the audience
For a start, Google Translate doesn't speak Duck
4

@JosephWright If @PauloCereda 's thesis is written in Duck, then translation the translation function T to Duck is trivial. T (s) -> {}

Evolution of the just-do-it-for-me: tex.stackexchange.com/review/first-posts/133805

@CarLaTeX :D

9:00 PM
@DavidCarlisle Maybe you could add \ifnum\catcode\&=\active\begingroup\lccode\~=\&\lowercase{\endgroup\let~=&}\‌​fi to the start code of aligned. :-)

@egreg hmmm.... tempting.. I hope you were impressed by my knowledge of tikz keys

@egreg Ah, the 'we need to sort actives' issue

@egreg or perhaphs better use \let&\aligntab where \aligntab is defined globally as \let\aligntab& or in luatex is the primitive of that name

@DavidCarlisle Tested mine and it works. :-)
@DavidCarlisle And break old documents that define \aligntab?

@egreg they'll break on lualatex anyway, and are there any?

9:14 PM
@DavidCarlisle I don't know. ;-) But LuaLaTeX will break documents for many other reasons.

@egreg mine works as long as & is active when you do the \let&\aligntab ie for the aligned definition
@egreg you don't say:-)
@egreg alignedleftspaceno isn't the default (that's alignedspaceyesifneg ) although it should make no difference here.
@egreg [narcissus] ?

oh hey there folks

@Shalvenay hello

anyone seen an "Undefined control sequence" message from within the datatool package complaining about \GenericError ... ?
or should I toss a Q up on mainsite about it?

@Shalvenay that usually means you used a fragile command in a moving argument

9:23 PM
hrm....
that's strange, to say the least

@Shalvenay or in other words an earlier error has sent tex off on some accidental code path and at some point it falls over and gives a confusing and unhelpful error messge

because the only arguments I'm passing to the command that's failing are a plain string and a file name, no \ in sight in either of them, and it's coming out of the expansion of that macro
and if I comment out the failing command and the one command that depends on it, the file compiles

any non ascii characters? in any case it needs an example because that error is almost always not related to the real cause.

@Shalvenay Underscores or spaces in the filename?

@AlanMunn an underscore, yes. I was wondering if that was an active char...

9:26 PM
@AlanMunn you should come to my uktug talk!

@Shalvenay put \show _ somewhere close and see what it says, and use \string_ in the filename

ah yes, the maths mode subscript
dunno why that's left live in a filename argument by datatool tho
as \includegraphics doesn't break that way when you use an _ in a file name fed to it

@Shalvenay oh no, that should work. I was expecting it to say something else (anything else)
@Shalvenay \includegraphics was written by a brilliantly skilled tex practitioner:-)

I'll try the \string_ trick
and \string_ did nothing :/
(i.e. gave the same error message as a plain _)

9:30 PM
@Shalvenay yes it would if _ had the default subscript meaning, back to original suggestion, make a complete test document...

will try that with datatool
welp, figured out what hte problem is
apparently utf8 inputenc and datatool don't mix

@Shalvenay That's certainly not true generally. But it's often simpler to use a UTF8 aware engine instead.

there are no actual non-ascii chars in the file I know of, but I use utf8 inputenc anyway
in the TeX file at least
let me check the CSV :P

@Shalvenay inputenc does nothing to ascii characters

9:38 PM
@DavidCarlisle yeah, I just default to utf8 anyway :p
just in case I find myself in such a situation

@Shalvenay a good plan, in general

anyone using debian / ubuntu here? I was going to install from this PPA : sudo add-apt-repository ppa:jonathonf/texlive
it's a fresh install

@baxx I'm not on ubuntu but why not just install texlive 2017 from tug?

I thought that i needed a PPA
damn, i'm installing that now :/

ah yes, the CSV has a few non-ascii things in it

9:43 PM
i think i looked at tug briefly, and it had stuff about dvds and what not so i just found a ppa

@baxx texlive is a part of the base Debian distro btw

@baxx yes as @Shalvenay says you should have texlive anyway, although getting the upstream tug texlive gets you closer to bleeding edge updates

but yeah...my CSV has a handful of non-ASCII chars in it
and apparently utf8 inputenc causes them to give datatool indigestion o.O

@Shalvenay probably fixable given a test case

@DavidCarlisle will get one whipped up here in a bit, just let me make up a csv

9:47 PM
@Shalvenay i thought that the debian versions were quite out of data?
*date

@baxx I think they are texlive 2017 aren't they?

@DavidCarlisle probably now, i remember them being behind and being told not to use the repo version before, maybe that's changed

I think so, although check packages.debian.org to be sure

they most certainly didn't used to be suggested ( as in, using the default apt install )

@DavidCarlisle got a test case together. want me to throw it up on mainsite?

9:53 PM
@Shalvenay that's best yes

so do people here use the version like I just installed? Or an older one?
I'm not sure how mega brake once a week bleeding edge this is likely to be, that's all

I use the stock distro TeXLive

@baxx i'd guess most of us here are tex geeks and use the upstream installation from tug (or compile from source sometimes) but for the majority of people an OS supplied version will be fine

yeah - but what i was wondering more was how much it's likely to break

10:01 PM
@baxx not much:-)

i definitely remember the stock versions being suggested against, because they were behind ( a year or 18 months back i guess )
OK - well, i'm a fairly boring user, so hopefully my stuff isn't pushing any boundaries :)

@baxx well the page you pointed to was pointing at tl2016 so that's old too unless it's been updated since.

got this just as I reached 100 answers :-)

@Shalvenay Are you dead set on using pdfLaTeX as opposed to using LuaLaTeX or XeLaTeX?

@AlanMunn hrm, I'd be willing to look into LuaLaTeX at least a little bit, does it offer a better solution to my situation than the stock pdfLaTeX?

10:14 PM
@Shalvenay It makes handling UTF8 input much easier.
@Shalvenay I tried to get your file to work with pdflatex but failed. (Which doesn't mean it's impossible.) But I can post a lualatex solution if you like.

@ShreevatsaR nice!

@Shalvenay see it's easier with a test file:-)
@AlanMunn quicker to post a pdftex version:-)

@DavidCarlisle right, i'll wipe this then i guess

@DavidCarlisle we'll see if a more generic solution comes down the pike :)

@DavidCarlisle yes haha... I've installed a PPA of an older version :')

10:27 PM
@Shalvenay that is as generic as it gets for pdftex (well I should do the three and four octet versions too if you need the full unicode range)

@DavidCarlisle ah
@AlanMunn -- yeah, post your LuaLaTeX solution as well :)

@Shalvenay the Declare... declarations are unrelated to the datatool handling you would need those if you just used μ in plain text

@DavidCarlisle ah
but they're not needed to use the special chars in the CSV?

@Shalvenay which special chars? the ones I added were exactly the ones you had in the csv

@DavidCarlisle I mean, are the Declare... declarations prerequisites for the datatool handling to work?

10:33 PM
@Shalvenay Done. Mine requires a lot fewer @s :) And you don't need any declarations for the extra characters.

@Shalvenay you need the two lines with \zz to make dataool work, then μ survives dataool and will work in the same way as if you had typed μ into a paragraph in the main document, which is that it will give an (understandable) error message saying that it is not set up for latex so you need teh DeclareUniocdeCharacter line to tell latex what to do with μ

@DavidCarlisle I see

@AlanMunn not even a zz ?

@DavidCarlisle No. As soon as I see a zz I just want to take a nap.

@AlanMunn well it is nearly midnight (in the canonical timezone)

10:38 PM
@DavidCarlisle I see you also got the wrong omega. (I've fixed mine already.)
@DavidCarlisle You are mean.

@AlanMunn details
Jun 29 at 16:15, by Paulo Cereda
@DavidCarlisle you are not mean :)

@DavidCarlisle But in this case you really are. Grenwich mean, in fact.

well, thanks for all the help :)

@Shalvenay You could complain to Nicola, datatool shouldn't use an unguarded \edef really.

@DavidCarlisle that is a point, yes
datatool probably should not do that :)

10:44 PM
@DavidCarlisle Too late to add the missing 'e' to my last message. Details.

@egreg Thanks, yes you are right. Your code does provide the URL ... why do you think I "shouldn't use \tl_if_blank:nTF" with O{} argument type, but instead should use "\IfNoValueTF and with o argument type." please ?

I'm not sure if these need addressing at all : tex.stackexchange.com/questions/1092/…
texlive vs texlive-full << is it just language packages that are the difference?

@VincentMiaEdieVerheyen if you have a command with an optional argument the user already has a way to flag the argument is not used by not using the argument it is unnecessary and confusing to also treat having the option but empty in a special way. compare \section{zzz} with \section[]{zzz} and \section[yyy]{zzz} specifying [] acts like [yyy] and makes an empty toc entry it does not act like not using [] at all
@baxx not sure, I always get the full one, but the basic executables are the same and you can always update to get everything else later

10:59 PM
@baxx I'm not sure which answer specifically you're referring to, but I think in the context of the distro packages vs vanilla texlive, texlive-full refers unambiguously to a distro package, while texlive refers (possibly ambiguously at times) to the distribution as installed via the TUG download page.

@DavidCarlisle hrm, i'll just install full and save any future hassle I think

@DavidCarlisle But for coding purposes, and for later automaticity (for example when adding or re-writing parts of the text) I am looking for a structure to keep ... If I need [] for some chapters, I will want to keep the structure [] throughout the document, yet leave it empty (at some point in time) for certain chapters.

@DavidCarlisle bug report filed, bug report form encoding bugs confirmed

@VincentMiaEdieVerheyen I see no reason to use \mymacro{X}{Y}[]; just use \mymacro{X}{Y}: this is how optional arguments work.

@Shalvenay Just to note, from the docs:
As from version 2.24, the predefined handlers now have limited support for UTF-8 characters. This is still experimental. The support will automatically be switched on if the inputenc package is loaded with the utf8 option before loading datatool-base (which is automatically loaded by datatool). If inputenc is loaded after datatool-base, you can use the (boolean) utf8 option when you load datatool-base to enable it.

11:04 PM
@AlanMunn yeah, I'm on v2.27 here

@Shalvenay Not saying you shouldn't have reported, just noting that utf8 support is still experimental (with pdflatex).

@AlanMunn yeah :) hopefully she's able to make sense of the bugreport despite her form mojibake'ing my CSV when I copypasta'ed it into the bug report body

@Shalvenay Yes, me too. That's quote from the current version's manual though.

(I did put a link to the tex.se question in so she can get it from there or just ping me and have me email it to her)

@Shalvenay This is why I've switched to UTF-8 engines for all of my documents.

11:06 PM
Why does

\IfNoValueTF{#2}{A #1}{B #2}%

and

\IfNoValueTF{#3}{A #1}{B #2}%

create a different output (there are straight brackets, and in a different place)?
\documentclass[a4paper]{article}
\usepackage{xparse, hyperref}

\NewDocumentEnvironment {PersonInformation} { m m o } {%
\IfNoValueTF{#3}{A #1}{B #2}%
}{%
\href{a.beautiful.url/?searchid=#1}{Click on me}%
}

\begin{document}
\begin{PersonInformation}{Test}[Test]
\end{PersonInformation}
\end{document}
is different, in its output (it changes more than the A and the B) from
\documentclass[a4paper]{article}
\usepackage{xparse, hyperref}

\NewDocumentEnvironment {PersonInformation} { m m o } {%
\IfNoValueTF{#2}{A #1}{B #2}%
}{%
\href{a.beautiful.url/?searchid=#1}{Click on me}%
}

\begin{document}
\begin{PersonInformation}{Test}[Test]
\end{PersonInformation}
\end{document}

@VincentMiaEdieVerheyen \IfNoValueTF{#2} will always return false, because #2 is a mandatory argument.

and yes, lualatex works on my MWE OOB, or at least doesn't barf
so with that, I will \end{chat} and see you all some other time

@egreg I see ... Let me look at it again, it looks like I forgot to add a 3rd argument as well

TeX 3.14159265 (TeX Live 2015/Debian)
this is the result from tex version, this is old then?

@baxx a bit but unless you are using luatex, it doesn't matter too much

11:14 PM
@DavidCarlisle well i just want the repo version now... i purged the PPA, and installed texlive-full ( which seemed to go surprisingly quickly... ) and this is the version i have

@baxx well as I say I don't have ubuntu myself, I thought it was a bit newer than that, but if you don't want the system one I'd use the one from tug, either of those is "normal" and there will be people around who can help with issues, if you get some non standard build from somewhere then you're on your own:-)

yeah, just feared i'd bodged the PPA and installed half the repo now or something. I'm installing it on another system as well so i'll see what that says after

:40418779 The following does what I want: (note I had to use

\ExplSyntaxOn
\cs_set_eq:NN
\IfEmptyTF
\tl_if_blank:nTF
\ExplSyntaxOff
\documentclass[a4paper]{article}
\usepackage{xparse, hyperref}

\ExplSyntaxOn
\cs_set_eq:NN
\IfEmptyTF
\tl_if_blank:nTF
\ExplSyntaxOff

\NewDocumentEnvironment {mainentry} { m m O{} } {%
\IfEmptyTF{#3}{First argument returns #1 NO}{Second argument returns #2 YES}%
}{%
\href{a.beautiful.url/?searchid=#1}{Click on me}%
}

\begin{document}

A) There should be a YES here. FAIL.

\begin{mainentry}{
first
}{
second
}[
third
]
\end{mainentry}

B) There should be a NO here. OK.

\begin{mainentry}{

@VincentMiaEdieVerheyen but what @egreg and I meant is that you should not want that. Treat [] like a supplied argument not like the argument has been omitted.

anyone have any emacs advice? preview-scale-function isn't making equations any larger

11:25 PM
@baxx not sure I have ever used preview:-)
do you mean you set the buffer local preview-scale variable? or is there a function called preview-scale-function, I din't see one?

@DavidCarlisle I see ... but I like my document to be consequent throughout ... that is, to have the same syntax structure for every entry ... this will help keep the data structured for later programming applications.

@VincentMiaEdieVerheyen so have three mandatory arguments and test if they are empty.
@VincentMiaEdieVerheyen that is replace O{} by m and keep your IfEmpty test, then the input for the unused case is always {} not allowing [] or no argument and then having to document that those two possibilities do the same thing

@DavidCarlisle i have the preview function yeah... what else is there to get equations to be larger in something like org?
stackoverflow.com/a/11272625/3130747 < seems to work for org at least

@baxx I use emacs all the time but I prefer to see my tex as tex so never tried the preview functions much.

@DavidCarlisle @baxx Yeah, David lives in a TeX Matrix :-)

11:40 PM
I was using org-drill and it insists that the latex is seen :P (maybe there's a setting to remove that )

@yo' actually I live in a tex free world filled mostly with xml

@DavidCarlisle c'mon, work does not count.

@DavidCarlisle OK ... I performed changes according to your advice ... but now somehow the URL isn't becoming clickable any longer? See the following MWE:
\documentclass[a4paper]{article}
\usepackage{xparse, hyperref}

\ExplSyntaxOn
\cs_set_eq:NN
\IfEmptyTF
\tl_if_blank:nTF
\ExplSyntaxOff

\NewDocumentEnvironment {mainentry} { m m m } {%
\IfEmptyTF{#3}{
First argument returns #1 NO
}{%
Second argument returns #2 YES
}%
}{%
\href{a.beautiful.url/?searchid=#1}{Click on me}%
}

\begin{document}

A) There should be a YES here. OK.

\begin{mainentry}{
first
}{
second
}{
third
}
\end{mainentry}

B) There should be a NO here. OK.

\begin{mainentry}{

@yo' hmm this isn't work:-) github.com/OpenMath/OMSTD/commits/master
@VincentMiaEdieVerheyen I get 5 clickable links