I love the help line. :)

@PauloCereda your instructions are clearly rubbish

@DavidCarlisle <3

Does anybody know how to prevent the error in the first run of the following code? It seems that on the first run the coordinates for \tikzmark have not been yet calculated, so the calculation of the intersection point fails:

\documentclass{article}
\usepackage{tikz}
\usetikzlibrary{intersections}

\newcommand\tikzmark[1]{%
  \tikz[remember picture,overlay]\node[inner xsep=0pt,inner ysep=10pt,outer sep=0pt] (#1) {};}

\begin{document}

\[
\tikzmark{c}2:\tikzmark{a}3 + 5:\tikzmark{d}3+\tikzmark{b}2:3+5:3
\]

\begin{tikzpicture}[remember picture,overlay]
\draw[name path=line1,red,->] ([xshift=2pt]a.north east) -- +(0pt,10pt) -| ([xshift=2pt]b.north east);
\path[name path=line2] ([xshift=2pt]c.north east) -- +(0pt,23pt) -| ([xshift=2pt]d.north east);

I am contemplating that somebody went crazy, even the simplest things don't resolve.

www.math.uconn.edu does not resolve.

@PeterGrill perhaps you can help me with this problem?

@egreg: your profile has +10k views!

@PauloCereda Let's hope no one tries that. Or, perhaps, let's wish someone does.

@GonzaloMedina There's a saying about it, "the first lesson is always the hardest". :)

!!/eightball are you there?

!!/eightball any idea on how to make the code in chat.stackexchange.com/transcript/message/8970200#8970200 work on the first run?

@GonzaloMedina we need to kick @PauloCereda into starting the bot

@DavidCarlisle Hold on. :)

!!/help

@PauloCereda Psmith, the TeX bot: Good night! I'm Psmith, the friendly TeX bot - the p in my name is silent, as in pshrimp. I'm here as a companion to our fellow users in the typographic land. As you probably noticed, I always reply under Paulo's account, but do not despair, I say, my replies are always preceded by my own name. Enjoy your stay at TeX.sx! If you need any help, just ask our chat residents. Cheerio!

@GonzaloMedina, @DavidCarlisle: the bot is on. :)

!!/eightball will @MartinScharrer get to 100k today?

@DavidCarlisle Psmith, the TeX bot: The great 8-ball says: better not tell you now.

@DavidCarlisle I could connect finally!

!!/eightball any idea on how to make the code in chat.stackexchange.com/transcript/message/8970200#8970200 work on the first run?

@GonzaloMedina Psmith, the TeX bot: The great 8-ball says: yes. Beware, Alan might stab you.

@PauloCereda So \path[stab]{Alan}{Gonzalo} should be enough! Thanks.

Is it possible to build the complete filename mcv-name-company.pdf for \includepdf with a given \company. I think there is a little bit tex necceccary. See this MWE:

I am moving to the terminal based browser now... but, before that, I would like to know if this chat can be accessed from the terminal... I'll try now.

Please bear with me as I test...

you can thank me later :P

@percusse I'm sure you can help me.

Can you please see the code in chat.stackexchange.com/transcript/message/8970205#8970205

@GonzaloMedina \tracingall on tikz code is interesting... vvv

\documentclass{article}
\usepackage{tikz}
\usetikzlibrary{intersections}


\newcommand\tikzmark[1]{%
  \tikz[remember picture,overlay]\node[inner xsep=0pt,inner ysep=10pt,outer sep=0pt] (#1) {};%
}

\begin{document}


\[
\tikzmark{c}2:\tikzmark{a}3 + 5:\tikzmark{d}3+\tikzmark{b}2:3+5:3
\]

\expandafter \ifx\csname pgf@sys@pdf@mark@pos@pgfid\the \csname
pgf@picture@serial@count\endcsname\endcsname\relax
\typeout{no}%
\else
\typeout{yes}%
\begin{tikzpicture}[remember picture,overlay]
\draw[name path=line1,red,->] ([xshift=2pt]a.north east) -- +(0pt,10pt) -| ([xshift=2pt]b.north east);

@DavidCarlisle your answer here says \makeatletter twice: tex.stackexchange.com/questions/108566/…

I'd edit it myself, but stupid SE says edits must be at least 6 characters

@kahen oops betrays that I changed my mind half way through:) will fix thanks

@DavidCarlisle Hmmm, so in effect there's a problem with the coordinates of the marks on the first run. Thanks!

@GonzaloMedina well they are completely unknown, basically it's like \ref but \ref can typeset ?? but if you are planning to do arithmetic on the values that's harder to fake.

You could just test \expandafter\ifx\csname pgf@sys@pdf@mark@pos@pgfid1\endcsname (ie test the first one is defined which is a test for "first run" but that fails if you add things the version above uses a counter so it checks the last mark set is actually known which is probably a bit safer

!!/cricket

@DavidCarlisle I see. Yes, I will use the other one. Thanks again. One day I have to do my first \tracing... to see it in action.

Psmith, the TeX bot, in fixed font mode: Let's take a look at the last cricket results:

- Namibia 276/10 &  336/10 * v Netherlands 293/10
- Kolkata Knight Riders v Sunrisers Hyderabad
- Rajasthan Royals v Kings XI Punjab

Our cricket expert David might explain these results later on.

!!/formulaone

@DavidCarlisle Psmith, the TeX bot: I'm deeply sorry, old chap, but the command formulaone does not exist.

!!/eightball should I implement a F1 command?

@PauloCereda Psmith, the TeX bot: The great 8-ball says: reply hazy, try again.

@DavidCarlisle Wow! The output of \tracingall is fascinating... it froze TeXworks, but what I could see was, in effect, TeX tracing it all!

!!/eightball Should I implement a Formula One command?

@PauloCereda Psmith, the TeX bot: The great 8-ball says: my sources say no.

@GonzaloMedina that's the idea, if tikz (or l3exp for that matter) is loaded it's best to keep it in a group, unless you have a big disk.

@David: upvotes for you done. :)

@PauloCereda thanks:-) I suspect that Martin is going to beat me to 100k though:-)

@DavidCarlisle We could downvote him. :)

@PauloCereda nah he probably deserves to get there. stick to downvoting egreg.

@DavidCarlisle oh! :)

!!/eightball Should we downvote egreg?

@PauloCereda Psmith, the TeX bot: The great 8-ball says: cannot predict now.

!!/eightball Should we downvote egreg?

@PauloCereda Psmith, the TeX bot: The great 8-ball says: don't you think emacs is great?

@DavidCarlisle: apparently the bot is against it. :)

@FrancescoEndrici: Is that you? :) github.com/franen

@GonzaloMedina Sorry for the delay. I did something similar to what David suggested by getting the intersections twice; first for the no of intersection checking and second if nonzero drawing things. So then it would be something like

\documentclass{article}
\usepackage{tikz}
\usetikzlibrary{intersections}

\newcommand\tikzmark[1]{%
  \tikz[remember picture,overlay]\node[inner xsep=0pt,inner ysep=10pt,outer sep=0pt] (#1) {};}

\begin{document}

\[
\tikzmark{c}2:\tikzmark{a}3 + 5:\tikzmark{d}3+\tikzmark{b}2:3+5:3
\]

\begin{tikzpicture}[remember picture,overlay]
\draw[name path=line1,red,->] ([xshift=2pt]a.north east) -- +(0pt,10pt) -| ([xshift=2pt]b.north east);
\path[name path=line2] ([xshift=2pt]c.north east) -- +(0pt,23pt) -| ([xshift=2pt]d.north east);

The lame \pgfextra{\xdef.....} part is for getting the intersection number available outside the current path. It's not impressive but rather pragmatic.

@percusse No problem. Thanks for this solution. I will also add it to my answer.

@GonzaloMedina My pleasure. Sorry for the mediocre solution :)

@percusse When using intersections and remember picture,overlay, a mechanism to prevent error messages while the coordinates are calculated on the first run, could be useful.

@GonzaloMedina Indeed. But one thing I can't get my head around is how to understand if it is the first run. There might be previous pictures already stabilized and that would effect them too.

@GonzaloMedina Just saw this now. Looks like percusse has solved it. Problem is that there is no intersection in the first run so somehow you have to check that an intersection exists before attempting to access the point of intersection.

I think he can't understand us: tex.stackexchange.com/questions/95257/…

huhu

@DominicMichaelis Hi.

I guess i gonna post a question on main, I hope I will be soon able to answer some questions too :(

@DominicMichaelis Don't worry: I didn't have TeX.sx when I was learning TeX. ;-)

what package to i need exactly for being able to use \[ \] ?

@DominicMichaelis None. It's built in the LaTeX kernel.

oh this one gives me an error and says I don't have a math environment

@DominicMichaelis Did you forget closing a formula? What's the precise error message?

It only has an error when compiling it there you can see the error by clicking going to first error. It says "Missing $ inserted <inserted text> 1.3 [ h_{1} \ ... " argh its difficult to writ it

wait now it doesn't compile at me let me search it

ah it doesn't like standalone

@DominicMichaelis standalone doesn't like centered material: it uses, by default, only one line.

@DominicMichaelis use option varwidth

it's just for a minimal working example i guess i compile it and use a screen shot which i will cut

@DominicMichaelis When you answer here, click on the small arrow at the far right of the line you want to link to.

@egreg Oh ok thanks :)

@DominicMichaelis See? In this way one is notified about the answer

anybody here?

@user4035 no sorry noone here ;)

I was looking at expl3 package, trying to understand, how to use it

but found no decent tutorials

only official pdf and this text: texdev.net/2011/12/11/…

anything available for the newbies?

@user4035 I wrote some beginners stuff in my blog

@user4035 texdev.net/index.php?s=programming+latex3

@JosephWright looking: texdev.net/tag/programming-latex3

Do we have a question about the differences of \trivlist and \list? I know both are using \@trivlist. But what exactly is the differences.

I hope my MWE is good

@JosephWright How to decipher this without referencing the pkg manual: "\cs_new:Npn \SayHello #1"?

\SayHello - function name, #1 - 1 parameter

but the 1-st part?

@user4035 \cs_new:Npn creates a new function \SayHello (the N) token, with parameter spec #1 and replacement test whatever you have in the { ... } (the n part)

Did you read expl3.pdf?

I hope my question isn't to localized

@JosephWright "Did you read expl3.pdf?" Yes, glanced, will look carefully now.

@percusse I prefer the following to PSY. :-)

Every time I want to open TeXworks, I go to the (Gnome) shell and type the first keystrokes of the app, tex. Gnome suggests me to send an email to Bruno Le Floch instead. :P

@Bugbusters Oh my! :) Another name? :)

@PauloCereda :-)

@egreg I like it when you give a better answer but I get the 15 points:-) tex.stackexchange.com/questions/108624/…

@DavidCarlisle :P

@MarcoDaniel list is less triv(ial) than trivlist

@DavidCarlisle Why not \renewcommand{\checkmark}{{\usefont{U}{msa}{m}{n}\symbol{\string"58}}}? (I always add \string because of risks with babel)

Oh no, today I have to play bass in the mass. :(

@DavidCarlisle that's simple

@egreg or I could just use \usefont{U}{msa}{m}{n}X :-)

@MarcoDaniel essentially list has hooks for setting the label format which trivlist doesn't use (as it's normally used in things like center where the label/item is hidden. or theorems where there is only one item). Its premature optimisation in a way @egreg had to patch a command (theorem I think) in an answer the other day to use list not trivlist so that the formatting could be changed. Had the original used list it would have been a trivial setting.

@DavidCarlisle Thanks. This could be an answer to an upcoming question ;-)

@MarcoDaniel I preferred the first answer myself.

@DavidCarlisle :-)

@DavidCarlisle I prefer your second answer here: tex.stackexchange.com/questions/108662/…

@MarcoDaniel You got the first answer as well...

@DavidCarlisle :-)

@MarcoDaniel If I edit the question (as I just did with a link to @egreg's answer) do you as OP get a notification with the edit comment? (I'm never sure if I am supposed to ping the OP with an explicit comment on site after a major edit)

@DavidCarlisle No. I saw the edit but I got no notification.

@MarcoDaniel ah thanks. I assume you get a notification for each new answer though? (my experience with the question interface is a bit limited:-)

@DavidCarlisle It seems that a new edit isn't handled like a new answer.

@MarcoDaniel OK thanks, then my comments with @ pings to the OP are not wasted:-)

@DavidCarlisle I provided an other starting point ;-)

@PauloCereda Yes.. but I've just pushed some wrong button on your github page! :-) I didn't mean to fork arara, but now I don't know how to remove it from my (useless) account :-)

I was about to access this on the main site, but then I remembered a question about it that I saw long ago. Have you ever see diagrams in books where sets are just amorphous blobs? Isn't there a question already out there that generates this amorphous blob? I can't seem to find it.

something like tex.stackexchange.com/questions/69359/… but less regular

aha! (if there was anyone wondering now but me: tex.stackexchange.com/questions/54985/amoeba-shape-with-tikz)

Hello there, everyone....

I have a burning question :P Is there any canonical way of putting a figure and a table inside a .......environment?

Should I define a hybrid float?

@percusse Do you use the package caption?

caption package gives a nice alternative with hybrid captions

@MarcoDaniel Hahaha. Great!

Yes I do

@MarcoDaniel But is it considered to be a good practice?

@percusse Use \captionsetup{type=figure} picture \captionsetup{type=table} table

The manual says the usage of something like

\DeclareCaptionLabelFormat{andtable}%
{#1˜#2 \& \tablename˜\thetable}
. . .
\begin{figure}
\centering
\includegraphics{. . . }%
\qquad
\begin{tabular}[b]{\ldots}
. . .
\end{tabular}
\captionlistentry[table]{. . .}
\captionsetup{labelformat=andtable}
\caption{. . .}
\end{figure}

And that leads to Figure 40 & Table 4: A figure and a table with a common caption

That's really great but I was curious if this has any side effects before I convert 10+ cases in the thesis

@percusse: Ask @axelsommerfeld ;-)

@MarcoDaniel Indeed :-) Maybe I should post a question instead ... if I can make it general enough.

@percusse he's online ;-)

@FrancescoEndrici Nah don't worry! :) It's time for us to follow you! :)

@cgnieder: we have another TeX.sx user to follow on GitHub! github.com/franen < @FrancescoEndrici :)

@PauloCereda: you are back. How was it ;-)

@MarcoDaniel I played bass! Yay! :)

@PauloCereda Did you get applause?

@MarcoDaniel In church, during the mass? Nah. :)

@PauloCereda :-)

I just knew that the bounty of 200 will not be available if we have exactly 200 reputation. See the following:

@Bugbusters I upvoted a question now you can ;-)

@Bugbusters When do you plan to stop the spread out of the points ? :-) I appreciate your view on points but if the points are meaningless then giving them away is also not helping, because, ehm, they are still useless.

@MarcoDaniel Oh. Thank you. You understand my body language. :-)

@percusse No problem. :-)

@Bugbusters ooh we have some dance going on. :)

@Bugbusters But the linked question was well answered by egreg ;-)

@percusse: I feel happy when offering bounties. I don't know why. :-)

@Bugbusters :) Must be the color choices of SE.

It is especially nice when the things we regard as beautiful are also regarded by other people as useful. (D.E.K)

how can you make two different numberations, i already saw sometimes the introduction pages was numbered with I II .. IV and so on and the real pages with 1,2 ... and so on

Do we have a question about the meaning of \ExplSyntaxOn ... \ExplSyntaxOff?

@PauloCereda: biblatex has no logo. I think such a great package needs one. Can you provide logo and suggest to @JosephWright ;-)

@MarcoDaniel Don't think so

@MarcoDaniel How about a bear?

@MarcoDaniel PL was pretty clear on this: just 'biblatex', capitalised normally

@JosephWright No bears?

@PauloCereda I think we want to stick to Pl's 'spirit' if we can

@JosephWright No we have: tex.stackexchange.com/questions/108696/…

@JosephWright Bear bear bear bear. :P

@JosephWright: or a blancmange! :)

^^ MP reference :P

@JosephWright: I meant "Now" -- Sorry

@egreg: a friend of yours wrote me an email today. :)

@MarcoDaniel Will you self-answer?

@JosephWright I would prefer a LaTeX3 team member would answer the question ;-)

@MarcoDaniel : Nice question!

@PauloCereda Sent u an email and a bunch of reading in Trello! LOL

@Brent.Longborough Uh-oh. We need another Trello board to manage our Trello board. :)

I was thinking it needs a bit of a tidy...

@MarcoDaniel OK

@Brent.Longborough Yep. :) You are our CEO.

@PauloCereda Nono, CIO maybe. Or better, you're our Pope, I'm just a bishop

@PauloCereda Tommaso?

@MarcoDaniel I tried this

  \begin{tikzpicture}
  \tikzset{mynodestyle/.style=
    {minimum~height=1cm,minimum~width=1cm,outer~sep=0pt,rectangle,draw=black}}

@Brent.Longborough By the power vested in me by the big parrot, I hereby grant you the title of Cardinal Longborough. :)

@MarcoDaniel Of course defining a style is better and it's perhaps clearer if it is outside the \ExplSyntax... part

Small trivial question. By default, " typesets as two curly quotes. However, if I use \usepackage[T1]{fontenc} it is now displayed as two straight quotes. The approved way (vide tex.stackexchange.com/a/43562/3406) seems to be to use \textquotedbl. So why does " work?

@egreg Yes. :)

This seems a bit too trivial for the site.

@egreg Yes, but it's good to know that ~ will work, too.

@egreg Sometimes I get the impression that there's a relationship between TikZ and glossolalia

@Brent: I actually got the 3rd line wrong - pdflatex if changed(myfile.idx) . The correct would be pdflatex if changed(myfile.ind) , which is the resulting file from makeindex.

I'm too slow these days. :)

@Brent.Longborough :)

@FaheemMitha Oh. I've always used two backticks for open and two single-quotes for close. I think I might be upset if " didn't give me upright quotes. Then I only have to remember \textquotesgl(I think)

@PauloCereda Tommaso was initially skeptic about arara, but now he's a convinced adept of the Holy Bird.

@egreg Oh my! :)

@PauloCereda Sure, but the idea is extremely valid, for helping latexmk with those things it doesn't (yet) know about

@Brent.Longborough :) changed is actually a hash calculation behind the scenes. :)

@PauloCereda Thanks

I believe the paylod is not too big.

@MarcoDaniel OMG

@egreg Per il santo ucello!

@PauloCereda Yes, I imagined it was.

@Brent.Longborough Expect a lot of email exchanges between us again. :) Marco and I are the muscles of the project, you are the brain. :)

@PauloCereda Which hash are you using? For this it can be the fastest you can find - no need for cryptography-level security!

Speaking of brain, I read something funny on 9gag yesterday: The brain is the most important organ of the human body... according to the brain.

@PauloCereda Did you olike my proposed solution to the Halting Problem?

@Brent.Longborough I didn't pick one yet, but maybe md5?

@PauloCereda LOL

@PauloCereda Yes, I'd think that plenty good enough, maybe even overkill

@Brent.Longborough Oh no, halting problem! Damn you Turing! :) I actually liked your solution, I think we can come up with something really nice. :)

@PauloCereda Yes, we could find ourself with a nice API...

@MarcoDaniel Hope the answer is roughly what you want

@Brent.Longborough yes! :)

We can even add logical operators!

if changed(foo) and !exists(bar)

@JosephWright Indeed. I thought \ExplSyntaxOn changes more catcodes?

@PauloCereda Let's not run before we can walk...

@MarcoDaniel Sort-of: there is some code to make sure other things don't catch us out

@Brent.Longborough Run Forrest run! :)

@PauloCereda ... Or we'll end up with Komascript

Hey if Marco is the Camerlengo, if I die, he's the one in charge! :)

@Brent.Longborough ooh I'll add that to the TODO list. :)

@MarcoDaniel Currently they are done by number, which makes reading the code a bit tricky: I'm not sure what is best (I suspect ```\<char>`` might be better long-term)

@Brent: and we need another colour for arara! Colours change between versions. :)

@PauloCereda BRG

@JosephWright: By the power vested in me by the big parrot, I hereby grant you the title of Cardinal Protodeacon Wright.

@egreg: By the power vested in me by the big parrot, I hereby grant you the title of Cardinal Gregorio, Prefect of the Congregation for the Doctrine of the Parrot Faith.

@JosephWright I think users who use ExplSyntaxOn are interested in the main behaviour. So ~, : and _ is important. Without testing: Does ExplSyntaxOn change the catcode of @

@MarcoDaniel No

Doesn't change @ as we don't use it

Should I add that?

@JosephWright To combine LateX2e macros and LaTeX3 commands I have to use makeatletter too.

@MarcoDaniel Yes

@JosephWright Perhaps you should add something about ~ at the beginning or end of lines. And possibly about the difference between \~ and \<space>

@egreg I promote this. And also the hint to @ would be nice.

@PauloCereda \definecolor{arara4}{rgb}{0.004,0.2,0}

@PauloCereda : Oh, by the way, starlogs has to do with Star Trek, not Star Wars

I claim my Nerd/Geek badge...

@PauloCereda BTW, careful with this Holy Bird stuff, until you've read The Unpleasant Profession of Jonathan Hoag

@Brent.Longborough ooh! :)

@PauloCereda I've archived the announcements and the Done

@Brent.Longborough Thanks. :)

OK, gotta go. Cheers all

@Brent.Longborough See ya, Brent!

TeXworks 0.4.5 had been released.

@StefanKottwitz Yes, lots of e-mails

@StefanKottwitz Thanks. Unfortunately no Mac version :-(

@MarcoDaniel not yet

@MarcoDaniel In progress, I believe

@JosephWright Scary to read about losing preferences, but distro version vs. manually installed version is an explanation.

@Brent.Longborough Ok, but I'm just wondering whether this change caused by \usepackage[T1]{fontenc} is documented. I haven't seen it anywhere.

Is there a way to change the numberation of pages in adobe reader ?

No arara support. :(

@DominicMichaelis What do you want to achieve?

@PauloCereda What? Where?

@MarcoDaniel TeXworks.

@PauloCereda You should open a ticket ;-)

i am writing a script and wrote an introduction and have the timetable, with a pagenumbering in roman, and the rest in arabic, and i would like that the numbering in adobe is the same as on the pages

@MarcoDaniel Where?

@PauloCereda code.google.com/p/texworks/issues/list

@DominicMichaelis Normally it works

@MarcoDaniel Ah.

@MarcoDaniel adobe only counts from 1-20 instead of "don't count ... I... III ... 1 ... 14"

@DominicMichaelis I tried the following example:

\documentclass{article}
\usepackage[a5paper]{geometry}
\usepackage{hyperref}
\usepackage{lipsum}
\begin{document}
\pagenumbering{roman}
\lipsum\lipsum
\clearpage
\pagenumbering{arabic}
\lipsum\lipsum
\end{document}

And the output provides i

i am using \document{scrbook} it gives me no i

but your works

@DominicMichaelis It works with scrbook too.

oh i am such an idiot i commented hyperref out -.- sorry

it gaves me warnings at first that why I commented them out

@marcoDaniel Thanks a lot

@PauloCereda Nice :) the network keeps growing...

hello

Gonzalo Medina gave me a solution about trees: tex.stackexchange.com/questions/106499/…

unfortunately, the editor has removed the picture

I works, but for the complex trees it becomes bad

What do people think I should talk about in France?

@JosephWright the state of the Euro?

@JosephWright I guess most people won't have seen expl3 syntax at all so you wouldn't get much further than an introduction to that and xparse in a single talk? unless you want to start at the other end and say why l3 is useful when coding siunitx and similar packages...

@JosephWright I read your posts about expl3 today, but couldn't find an example of program, looping through array elements or defining an array. Is it possible in expl3?

@user4035 What sort of 'array'

@user4035 You might want a sequence or a property list, depending on use case

@JosephWright like in programming languages

yes

a sequence of values

@user4035 That's a bit advanced for beginners

It's one of the fundamental building blocks for programming

@user4035 TeX is a typesetting system, so our focus is somewhat different :-)

I know, sorry, but when I see programming features I can't stop myself

@user4035 As I said, depending on the exact use case I'd use either a sequence or a property list

@user4035 For example, if you want random access then a property list is probably best

Well, I think people need to learn, how to use both of them

@user4035 Not all programming languages have arrays in that sense. I was just reading the lua manual the other day and similarly there that doesn't really have arrays it just has tables (like l3 property lists) where the key might (or might not) be numeric.

@DavidCarlisle Quite

@DavidCarlisle yes, php uses hashed for everything

in low level languages like C there is a strict distinction

@user4035 and in macro expansion languages like TeX you are lucky if you get any structure at all:-)

As far as I understand, you write functions, that will output your data in expl3 format. And then define what data you want to output

@user4035 We've provided data structures that fit with what we need to use: the one I still need to finish is data tables

@DavidCarlisle Exactly

@user4035 As David says, as everything is built in macros we have some trade-offs

Here is tikz code:

{1,..,15} can be viewed as an array of numbers

and we have a loop

@DavidCarlisle I was wondering if 'big picture' thoughts, expl3, or xparse would be best

@user4035 That doesn't need an array at all :-)

@user4035 It's not really an array (implying constant time access) it's a sequence. L3 has sequences, it just doesn't have arrays.

\int_step_inline:nnnn { 1 } { 1 } { 15 } { <code> }

@DavidCarlisle Property lists though

@JosephWright yes property lists and sequences seem a reasonable foundational datatype layer given the macro based implementation.

@DavidCarlisle As I said, I do think a data table/record type structure is handy, for example for what datatool or pgfplots do plus things like storing author data. The problem is performance: pgfplotstable is fastest at the cost of csnames

Tried this:

worked

@JosephWright hard to know, depends on the audience (perhaps the organisers can give you a tip:-) if they are hard core tex hackers they may appreciate xparse rather than lots of expandafter, but if they have never seen expandafter being told they may see less of them isn't much of a sales pitch.

@JosephWright What does the 1-st { 1 } stand for?

@user4035 First value

@user4035 So here starting from 1

@user4035 it's a fortran DO LOOP start step and end

hm, there are two {1} -s

that's what confused me

ah step

got it

!!/answer Happy Birthday Leonhard Euler!

@JosephWright see this:

@DavidCarlisle what if I what to use if operator in the loop?

what should I write inside the condition?

@user4035 I didn't follow the discussion, but it should be

\ExplSyntaxOn
\int_step_inline:nnnn { 1 } { 1 } { 15 } {
\bool_if:nT { \int_if_odd_p:n {#1} } {#1}
}
\ExplSyntaxOff

Or, more simply,

\ExplSyntaxOn
\int_step_inline:nnnn { 1 } { 1 } { 15 } {
\int_if_odd:nT {#1} {#1}
}
\ExplSyntaxOff

This would produce 13579111315

@egreg worked

both odd and even

@DavidCarlisle Oh no, someone did his homework for evil purposes - look for bugs in dandelion!

Help L3!

@Joseph: how about a keynote on breaking news in the TeX world, with, say, a bug in longtable? :)

@JosephWright: ^^

I got that joke! Aha!

I think we could add Karl and Norbert to the list. :)

@PauloCereda more or less, yes:-)

@PauloCereda Please can we have @PauloCereda as next victim^H^H^H^H^H^H guest in the interview room?

@DavidCarlisle The next interviewee will be Nicola. :)

@cgnieder: Another one! github.com/Qrrbrbirlbel

:)

@PauloCereda Ooh, it's getting crowded! :)

@cgnieder LOL

Why does Barbara Beeton not insert \! between $ and \begin{aligned} in her answer here?

@Bugbusters because the AMS classes are designed to produce documents according to the AMS house style (where barbara has worked since the dawn of time) People only insert explicit spaces if they want to modify that style.

Hey @DavidCarlisle. Seen J&M?

@FrankMittelbach sent a reply a bit ago! (hadn't been online all afternoon:-)

@DavidCarlisle cool. neither have I ... first real spring day here so we did a 3 hour cycling tour ... only issue, seems everybody else was on the road as well :-)

!!/fortune

@PauloCereda Psmith, the TeX bot: Here is your fortune: David deserves 10 bucks.

Not again. :)