TeX, LaTeX and Friends - 2018-03-22

« first day (2699 days earlier) ← previous day next day → last day (2221 days later) »

12:21 AM

Randall forgot to include me :(

In competition, you can only play a name if you know who the person is. No fair saying "Frank ... Johnson. That sounds like a real person! Let me just Google him real quick."

2 hours later…

2:01 AM

Did I miss much? Nope. :)

@HaraldHanche-Olsen It was a Beatles reference...since you brought up holes.

3 hours later…

5:04 AM

@cfr This is the kind of message I like!

1 hour later…

Christian Hupfer

6:25 AM

@marmot It's der Bär ... as it is der Kater oder der Hund...

@HaraldHanche-Olsen: Regarding cats: There is cat content here on TeX.SE, e.g. about lions

1

Q: Why don't add a lion in our background image?

The proposal to add a duck was rejected because it is an inner joke which could be interpreted in a different way by chat residents and by new users. However, I found our background a bit boring. Why don't add a lion? It is universally known as the TeX mascot (Does the TeX lion mascot have a na...

Harald Hanche-Olsen

6:49 AM

@barbarabeeton Yes, I really did feel sorry for you and your hip. How is your new one performing, by the way? To me, the only trauma was psychological, and I am fast getting over that. (Plus the deductible on car insurance, maybe an increased premium in the future, and the inconvenience of being carless for a few weeks.) I am lucky after all.

@AlanMunn Ah, afraid I am not well versed in Beatles lore.

@ChristianHupfer Oh yes, \ifcat\Large …

2

@cfr Much better. Thank you.

Harald Hanche-Olsen

7:23 AM

By the way, my sister's cats have catcodes that let them use the cat door while shutting other cats out.

6

2 hours later…

8:57 AM

@cfr <3

@PauloCereda I just bought a sandwich at the student's café. They had a duck at the counter desk. They said something that "when you explain something, you should explain it so that the duck understands". I wasn't aware of that.

@mickep ooh duck logic

Harald Hanche-Olsen

@PauloCereda Here is one for you.

(Sorry, I am not linking directly to the image for copyright reasons.)

@HaraldHanche-Olsen :)

Harald Hanche-Olsen

Rubber duck debugging

In software engineering, rubber duck debugging or rubber ducking is a method of debugging code. The name is a reference to a story in the book The Pragmatic Programmer in which a programmer would carry around a rubber duck and debug their code by forcing themselves to explain it, line-by-line, to the duck. Many other terms exist for this technique, often involving different inanimate objects. Many programmers have had the experience of explaining a problem to someone else, possibly even to someone who knows nothing about programming, and then hitting upon the solution in the process of explaining...

9:10 AM

What are by default active characters except for ~?

Could somebody explain why I there is (or I have), as if, copies of packages in User\xxx\AppData\Roaming\MikTeX\2.9 worth 1/7 of the total size of the default installation path?

@Skillmon Imho only ^^L (\def^^L{\par}% ascii form-feed is \par).

@vlg That's the user install tree. Packages go there when you install on-the-fly as user.

@UlrikeFischer :( so no other useful stuff for code golfing in TeX.

@HaraldHanche-Olsen Oh, thanks for the reference!

I thought that might be it, so now I have copies. If I let MikTeX ask me for every package, would it change the path, or does every package have to be installed via the console?

9:22 AM

@Skillmon would it be cheating to write a cg.tex with usefull shortcuts and upload it to CTAN? It could contains things like the \let\def\c and \newcount\b and `\let\r\RequirePackage?

@UlrikeFischer better would be something like \catcode`\.=13 for many tokens, than \let.\let and stuff :)

@UlrikeFischer and something like \catcode\:=13` and \def:#1{\catcode#1=13}` or \def:#1#2{\catcode#1=13\def#1{#2}}`

Christian Hupfer

user image

Oops, I did it again ;-)

@ChristianHupfer Britney Spears?

@vlg sorry I don't understain your question. If you install packages as user they go in the user install tree. If you want to have them only in the main tree you should install them with the admin version of the console and deinstall them in the user version of the console.

you understood correctly. Sorry for the bad wording, thanks

Christian Hupfer

9:27 AM

@PauloCereda No, that other female blonde singer... ;-)

@ChristianHupfer ah :)

Christian Hupfer

@PauloCereda Pretty obvious and unambiguous, isn't it? :D

@ChristianHupfer exactly. :)

@AlexG we can shorten our code golfing code by 10 bytes :)

@AlexG using ~ instead of \c:

\RequirePackage{tikzducks}\newcount\b\let~\def~\0{red}~\1{orange}~\2{yellow}~\3‌{green}~\4{blue}~\5{cyan}~\6{violet}~\7{gray}\loop\b\uniformdeviate8\tikz{\duck[b‌ody=\csname\the\b\endcsname]} \ifnum\b<7\repeat\stop

Christian Hupfer

@PauloCereda I meant Heather Nova ;-)

@HaraldHanche-Olsen \ifcat\Large\feeditwithWordusers\else\servemice\fi ;-)

Harald Hanche-Olsen

9:36 AM

@ChristianHupfer :)

9:54 AM

@DavidCarlisle that thing about \includeonly vs \include vs pdfTeX warning (dest) is not too bad to fix for normal xrefs. It is much worse if you use a ToC in the doc as well. But since I rarely use the toc or the bookmarks when I edit, it is probably easier to simply remove the hyperrefs from the toc in that case

And hyperref it self explains how to do that in the main aux

10:27 AM

@Skillmon s/\\c\>/\~/g? Done. Thanks for this suggestion!

@AlexG and I'm working on cg.tex (thanks to Ulrike), which tries to introduce weird syntax for code-golfing :)

10:42 AM

@Skillmon ooh :)

11:01 AM

@Skillmon look in xii for a few tips

@DavidCarlisle I don't have to obfuscate that much. I'll post some code here, in a few minutes/hours

@Skillmon no but the initial macro that makes a character active and defines its active version might be useful

@DavidCarlisle already implemented. As of now ;#1#2; is \catcode`#1\def#1{#2}

@Skillmon :-)

@Skillmon doesn't that need #1 to be already active?

@DavidCarlisle missed a 13 there

@DavidCarlisle And I guess you're right.

@DavidCarlisle stupid scanning fixing cat codes :(

11:12 AM

@Skillmon that's why xii uses a double character each time

@DavidCarlisle So I'd have to use ;..foo; to define . to expand to foo. That's a byte too much :(

@DavidCarlisle anything someone could do with \futurelet?

@Skillmon make everything active in the cg.tex to start with

@Skillmon for a start if you make everything above 128 active and use latin1 encoding that's a bunch of active characters you can use straight away, you could define them to be \loop and \if and ....

11:29 AM

@DavidCarlisle but who uses latin1 as encoding? My vim certainly doesn't.

@Skillmon people who are trying to halve their bytecount for code golf. If you used a real editor you could save the buffer in any any encoding that had the characters used.

@DavidCarlisle since I have a real editor, that is not a problem at all. It's just not my default :)

@Skillmon default settings and codegolf don't mix:-)

@DavidCarlisle how about using \scantokens? So \def;#1#2;{\catcode`#113\scantokens{\def#1{#2}}}` :)

@Skillmon too modern for me

11:39 AM

@DavidCarlisle and doesn't like to be used in plain (which is sometimes better for code golfing)

@Skillmon works with pdftex plain just classic dvi tex is set up not to use etex

i seem to have \ref{blah} putting 2 in the doc, although the figure is "figure 1", i'm not sure what kind of error this is / what to look for

@Skillmon the main disadvantage is that it will scan with the current "weird" settings so the content of #2 will be interpreted with the settings when the active #1 is used not when the definition is made

@baxx is 2 the section number?

ah yes

you have \label before \caption

11:43 AM

i didn't realise there was an order to that

this is the most frequently asked latex question of all time:-)

ha :')

@DavidCarlisle good to know. But somehow it inserts a white space during the definition

\begin{figure}[H]
\centering
\includegraphics[width=0.45\textwidth]{./images/initialScatter.png}
\caption[]{blah blah blah}
% \label{initialScatter.png}
\end{figure}\label{initialPlot}

@baxx \label records the "current reference number" at teh point of teh label,

11:44 AM

this should be the correct ordering then?

ah no, ok

it's done, sorry ! thanks

@baxx no that's completely wrong !!!! put it where it was commented out (or better) inside the argument to caption

inside caption ok, cool. thanks

88

Q: Why does an environment's label have to appear after the caption?

\documentclass{article} \usepackage{algorithm} \begin{document} \setcounter{section}{54} \section{A section with no label} \begin{algorithm} \label{myalg} \caption{An algorithm with the label `myalg'.} \end{algorithm} Reference to myalg: \ref{myalg}. \end{document} With this code, the referenc...

cross-referencing numbering captions algorithms

@baxx either immediately after \caption{...} or inside \caption{fooo\label{fig:foo}}. But be sure to not put a blank before \label

@Skillmon inside is better than after

11:47 AM

@DavidCarlisle yes, I know, but never had issues with after.

what's a blank? ( a blank line? ) I've put it within caption, and it seems to work fine

@baxx anything interpreted as a white space by TeX.

hrm, well it's working now so all good

@Skillmon are you sure? :-)

15

A: When is leading/opening whitespace of a line in a tex file important?

The check program is clearly being over cautious (ie wrong) to say page breaks can always happen at spaces before \label. However the form you give is not optimal,. but for a different reason: If you put the \label after the caption (whether or not you leave a blank space or newline before it) t...

@DavidCarlisle never noticed that. Thank you very much (and now I have to rewrite some old code).

@DavidCarlisle do you know why \def;#1#2;{\catcode`#113\scantokens{\def#1{#2}}} inserts a white space if I use it like ;.foo;%?

@DavidCarlisle nope ;.?;;?foo;. produces only two spaces and ? as output.

@DavidCarlisle \def;#1#2;{\catcode`#113\scantokens{\def#1{#2}}\unskip} eats the space (though I still don't know where it originated from).

11:59 AM

@Skillmon hmm. I think you meant "blame Ulrike.

@UlrikeFischer no, the idea is funny. And it's somehow challenging to create :)

@Skillmon because \scantokens is less than optimally designed. luatex has a different one \scantokenstext or some such that doesn't do that (it's the fake end of line at the fake end of file that \scantokens uses. Set the endoflinechar to -1 to stop it

@DavidCarlisle or just use \unskip...

@Skillmon well that depends.

@Skillmon if the command you defined is used in horizontal mode so the spurious space makes a skip then you can unskip it, but if it is used in vertical mode teh space is ignored anyway and the \unskip will remove the previous vertical skip or generate an error if you are on the main vertical list

@DavidCarlisle I could put a \kern0pt at the beginning of the macro, to prevent \unskip from being to eager.

@DavidCarlisle didn't consider that

12:06 PM

@Skillmon welcome to my world

@DavidCarlisle I learn way to much during this little experiment

@DavidCarlisle how do I actually do that?

@Skillmon put \endlinechar=-1 before the scantokens

Harald Hanche-Olsen

@DavidCarlisle Will he need a space after the -1 too? In case more digits appear in the scantokens text?

@Skillmon well if you need challenges: there is a Alice and Bob question which could perhaps be done with tikzpeople.

@HaraldHanche-Olsen no \scantokens isn't expandable

@HaraldHanche-Olsen but I'd use \m@ne to be safe:-)

Harald Hanche-Olsen

12:13 PM

@DavidCarlisle Thet etex manual says otherwise.

@DavidCarlisle Good idea.

@HaraldHanche-Olsen reading manuals is cheating

Harald Hanche-Olsen

@DavidCarlisle I know, but I'm too lazy to go read the source.

@HaraldHanche-Olsen @Skillmon sorry I lied, \scantokens is expandable (but you can almost never use it in an expansion context as you need to set \endlinechar and \everyeof to make it safely usable)

Harald Hanche-Olsen

@DavidCarlisle I was surprised at your claim because I expected it to act a bit like \input (the primitive), which (if I recall correctly) is also expandable, with an empty expansion.

@HaraldHanche-Olsen yes except it is more like \write the contents to a temp file and \input them back, and \write isn't expandable, but you are correct it is expandable.

12:39 PM

@DavidCarlisle but wouldn't that need to be grouped?

Harald Hanche-Olsen

@Skillmon You can do that, and replace \def by \gdef, or just reinstate the old \endlinechar afterwards.

@Skillmon that is why \scantokens is so hard to use in practice, rather than group just set it back to 13 after or look at \@xtypein in latex.ltx

1:08 PM

@HaraldHanche-Olsen that's what I ended up with.

Christian Hupfer

The magic of maths: facebook.com/8FunnyWorld/videos/253390805285786 ;-)

anyone know the how to write the unicode (utf8) rho in Emacs?

@daleif that highly depends on your keyboard layout. If you're using neo2.0 it's just "Shift+Caps+r" (well technically it's "mod5+r")

Christian Hupfer

@daleif For me it is CTRL-SHIFT U and then 03C1... ρ

hmm, none of those worked. Nevermind, I only needed it once, so I copied it from wikipedia

Harald Hanche-Olsen

1:57 PM

@daleif C-x 8 C-m GREEK SMALL LETTER RHO will do the trick. You do have completion, so you don't need to type all of that.

@daleif Or turn on the TeX input method (M-x set-input-method) and type \rho.

2:14 PM

Turns out to be much harder than I thought.

In order to avoid a duplicate Q I first ask here: How can I write unbalanced { and } in 2 separate PS \specials to DVI/Postscript?

Harald Hanche-Olsen

@AlexG Temporarily changing some catcodes might be the way.

Christian Hupfer

@HaraldHanche-Olsen Do those cats use encryption for their cat codes as well? ;-)

Harald Hanche-Olsen

\begingroup
\catcode`\{=12
\catcode`\}=12
\catcode`\(=1
\catcode`\)=2
\gdef\lbraceother({)
\gdef\rbraceother(})
\endgroup

@AlexG ↑↑ And the use the defined commands inside your \special.

@barbarabeeton I deleted my question on Meta because it's not the moment to ask it, reading your comment, thank you for pointing it out! Now I can't see my post anymore but another user commented about a "we hate fun policy" on SE. Does this policy exist? Where can I find it? Thank you!

2:29 PM

NOW IT'S OFFICIAL: I WILL BE AT RIO FOR TUG 2018

@PauloCereda Was there the possibility you would not? Argh!

@CarLaTeX I would go anyway, but now I have confirmed logistics. :)

@PauloCereda Ah, well!

Christian Hupfer

@PauloCereda Shaky Bus?

@HaraldHanche-Olsen Works perfectly, thanks a lot!

2:35 PM

@ChristianHupfer Shaky plane. :)

Christian Hupfer

@PauloCereda I thought you would use your duck wings :D

Ok, cg.tex will not happen any time soon. It doesn't work out the way I intended it to. Back to other procrastinating stuff, which will hopefully be of more success:)

@ChristianHupfer But then I'd lose the onboard biscuits. :)

Christian Hupfer

@PauloCereda Oh my, how could I forget the onboard crumbled biscuits ;-)

@ChristianHupfer <3

Christian Hupfer

2:39 PM

@PauloCereda Is is it a bird? No, it's araraman in his shaky plane full of free onboard biscuits ;-)

@ChristianHupfer ooh more biscuits

Christian Hupfer

Citing @PauloCereda: I am writing a thes.... oh biscuits ... nom, nom, nom ...

@PauloCereda: :-P

@ChristianHupfer awwww

I'm trying to work out how to cite this thing : ec.europa.eu/competition/mergers/cases/decisions/m1406_en.pdf , when I search for "cite pdf" i keep on getting things about citing websites though. What kind of thing should i search for for this?

Harald Hanche-Olsen

2:56 PM

@ChristianHupfer I really don't know, but expect it's just a dumb RFID chip. It's also used by vets to identify the cats when needed.

@CarLaTeX -- as far as i know there is no official "we hate fun" policy, but it does seem that the "powers" lack appreciation for the type of silliness that this group often indulges in. here's the link that describes the upcoming site changes; this may be what's being referred to: meta.stackexchange.com/questions/307862/…

3:12 PM

@HaraldHanche-Olsen -- i'm glad to hear that you're coping reasonably well. as for my new hip, yesterday, for the first time, i was able to put on my own socks and shoes without assistance. and on level surfaces, i am less reliant on the cane -- but i will continue to carry it for the foreseeable future, if only as a signal to others to stay away. (it's a pretty good weapon, if needed.)

@barbarabeeton: Rio, please? <3

Christian Hupfer

3:45 PM

@HaraldHanche-Olsen I was actually joking...

@HaraldHanche-Olsen Why are you using plural in "marmots" and "Englishmen"?

Phelype Oleinik

4:02 PM

Can someone check what I said here, please: tex.stackexchange.com/a/422603/134574

It makes sense for me, but I'm not sure it's correct...

@marmot Because the plural in English means "one or more", not "more than one".

@marmot And the generic plural names the kind of thing, not a plurality of the thing.

@marmot TLDR: Because linguistics. :)

4:17 PM

@PhelypeOleinik it is more or less correct, but the questions seems to ask how to get \testd to work and print a&b while your command takes the a&b as argument.

Phelype Oleinik

@UlrikeFischer Ooh... I did what was under the I would like to do this but it doesn't work:.

I'll edit to add the other cases later. I'm not at the computer right now.

Thanks for checking it :)

(But why "more or less"?)

Harald Hanche-Olsen

@AlanMunn :)

@AlanMunn I see. So "The Alan Munns are very good at English grammar" is actually correct.

@marmot -- +1

Harald Hanche-Olsen

@ChristianHupfer What, joking on tex.se? Heresy!

Christian Hupfer

4:27 PM

@HaraldHanche-Olsen blame \someone\else ;-)

@marmot Not quite, since defintite plurals in English can't be generic. But without the "the", yes.

Christian Hupfer

@CarLaTeX: Where's the Lion question?

@ChristianHupfer Deleted.

@ChristianHupfer "Where are the lion questions?" would also be correct. ;-)

@ChristianHupfer I deleted it because @barbarabeeton pointed out that some major changes are in progress: chat.stackexchange.com/transcript/message/43560947#43560947

Harald Hanche-Olsen

4:46 PM

@AlanMunn Thank you. I couldn't have explained it that way; it just came naturally from whatever language feeling I have. Good to see that it's not too far off the mark.

@marmot :) Of this isn't true, which means that we need to add a bit more to the story: pragmatically the singular is stronger than the plural, since the plural includes the singular but not vice versa, so if you know that singular is true, you must use the singular. This is why "The questions were tough" isn't ambiguous, but "Do you have questions?" can be answered affirmatively even if you only have one.

@AlanMunn Got it. I can use singular or plural as I like unless you have objections. ;-) ;-)

Harald Hanche-Olsen

@barbarabeeton A cane is a formidable weapon indeed. For even better effect, you might consider exchanging it for a quarterstaff.

@marmot Exactly. I am the arbiter of all plurals. Even the pluralia tantum :)

Harald Hanche-Olsen

@AlanMunn For a moment there, I thought you said pluralia tantrum. Think the Donald.

4:53 PM

@HaraldHanche-Olsen lol

user image

Harald Hanche-Olsen

@marmot It seems I am the only insect here.

Somehow the crazy pattern continues.... where is @PauloCereda?

@HaraldHanche-Olsen Think of this as a medal ceremony at the Olympic Games... ;-)

@marmot quack <3

@PhelypeOleinik The problem is the catcode of the & at "definition time" (at least in some of the examples, I didn't check all of them) and not when the command is called.

4:59 PM

@HaraldHanche-Olsen I bee-lieve in you. <3

5:31 PM

@HaraldHanche-Olsen A cane in Italian is a dog :)

@CarLaTeX -- so, cave canem. (old latin saying.)

@barbarabeeton Hic sunt anatum

5:54 PM

@PauloCereda -- better, anates hic sunt. (you're mising singular and plural, never mind nominative vs. some other case.) (yes, i can get pedantic.)

@barbarabeeton Two grammar geeks in the chatroom make ducks and marmots nervous.

@AlanMunn :) (and red squares too, for that sake)

anyway, the Indian place at the corner starts it evening 40% discount in 2 minutes, so I gotta go :-)

@barbarabeeton Yes :):):)

@barbarabeeton How much snow did you get?

6:10 PM

@barbarabeeton <3

@AlanMunn /quacks linguistically

@marmot -- actually, this nor'easter was a bust. by this morning, what there was could be swept off with a broom. we were expecting 8-12in of heavy wet stuff. and with all the scary predictions, the office closed at 2pm yesterday, and didn't open until 9:30 this morning. so more "quiet time" when i could actually get something done, since i live only 15 minutes away from the office. (but there are hints that there may be more bad weather this weekend.)

@marmot -- if i may be so bold, where exactly are you located? from various clues, i'm guessing maybe slac. that's a really neat burrow, even if it is perpendicular to the san andreas fault.

@barbarabeeton It's a bit south of it. Hint: anteaters and middle earth are very popular here. But the weather is also bad (or, given the drought, good?) here.

user image

6:25 PM

@marmot ooh an anteater

@PauloCereda let ooh = ooh +1 ;-)

Not sure how to ask this question in the official think, so here goes- would it be possible in a tabular environment, at end of every entry, to append a row in a .dat file with the same entries as in the tabular row but comma-separated?
This is so that it may be later used for a graph using pgfplots.

@vlg Look at pgfplotstable or csvsimple packages: from a .csv file they create a table. You could start from a .csv and from it create both the table and the plot.

@vlg An example: tex.stackexchange.com/a/398352/101651

6:46 PM

Yeah, I'm reading all three manuals, I guess I'll get back here once I ascertain how to ask my question

2 hours later…

8:51 PM

@DavidCarlisle thanks for your quick answer. I wonder why I keep forgetting about \pdfstrcmp...

@Skillmon I don't think I've used it before:-)

@egreg Thanks. However an expl3 solution would've been easy in the first place :) I want/wanted to know how to do this without packages.

@Skillmon Sorry, that's nonsense to me.

@egreg could you please elaborate that?

@Skillmon Why should one avoid packages and reinvent the wheel?

8:56 PM

@egreg for academic purposes.

@Skillmon Sorry, in academia there are libraries. ;-)

@egreg when you were still studying you proved stuff mathematicians proved some hundred years ago. And I hope you did not just copy a prove out of some textbooks everytime.

@Skillmon I didn't ask for help on the net. ;-)

@egreg there was no net...

@egreg and it's nothing I can show in my CV as I don't think any future employers care whether their newly recruited engineer can parse stuff in TeX :)

@Skillmon Objection overruled. :-D

9:03 PM

@Skillmon he's just upset because he thinks you are going to give the tick to me

@DavidCarlisle If you manage to provide the space ignoring version :)

Phelype Oleinik

9:18 PM

@UlrikeFischer Thanks a lot! Now I understood what happened there :)

@Skillmon I might do it later just needs to juggle a couple of extra macro calls to drop the spaces. But it seems a bit odd thing to do so i'm not that motivated:-) is there a use case behind the question?

@DavidCarlisle there is a super stupid use case behind the question. Which in fact is so stupid I don't want to talk about it. When I created the template for my bachelor's thesis I made a deal with the professor that it will look almost identical to his existing MS Wort template and I'll share it with every other student who wants to write his thesis in LaTeX. As a result I'm the maintainer of that unofficial package.

@Skillmon but why the {} test?

@Skillmon One should never make deals with the professor. ;-)

@DavidCarlisle The MS Wort template has a setup that uses the same ugly Arial font for both text and math. In order to achieve that I used helvet and mathastext. But unfortunately mathastext doesn't play well with \mathaccentsV (above all letters taken from helvet in math the accent is shifted to the left). So I wrote a patch for that yesterday. Now I need to parse the input to my patch in order to display stuff right.

@DavidCarlisle Some of the tests need to know whether the content is in an extra group just to deal with stupid users input.

9:27 PM

@Skillmon That's why you have the documented source ...

@DavidCarlisle and I thought "why not do this using only plain or LaTeX to see whether I can do that"

@Skillmon almost certainly it would be more robust to fix the macro to have correctly braced arguments than to do the kind of tests that you ask for in that question

@DavidCarlisle the problem is, what if the user puts some more braces around something which is perfectly fine syntax normally (nothing wrong with \hat{{x}} as far as I know).

@Skillmon well \let\xhat\hat\def\hat#1{\xhat{\@firstofone#1}} would drop any spurious braces

@DavidCarlisle what about \hat{{{x}}} if we have a very reluctant user?

9:36 PM

@Skillmon well if you are using a non standard accent command to make some nonsense font choice work, just document that spurious braces don't work I can't see any point in recursively testing for brace groups and stripping them one by one, what goes wrong with the brace groups anyway?

@DavidCarlisle the choice whether to move the accent a bit to the right or not. Something I could easily fix with an optional token. It was just for the sake of it :)

@DavidCarlisle circumvent the problem is easy. But then I tried to solve it and ran into problems while doing so.

9:50 PM

@Skillmon but basically the test I suggested with pdftsrcmp works and to make it ignore more levels of grouping and extra spaces outside the group is just a matter of adding \@firstofone in enough places, because \@firstofone{x} is \@firstofone x is \@firstofone {x} but it would be simpler and more robust to skip the pdfstrcmp test and just insert \@firstofone... somewhere in your redefined accent macros so extra braces are ignored

@DavidCarlisle yes, I know. But the experiences I got from questions of my fellow students is: No one reads documentation or readmes. In 20 years someone is going to come to me and say: Hey, why is that not working :(

@DavidCarlisle and going-to-future is the correct tense here.

@Skillmon the purpose of documentation is not that the users avoid problems, it's so that when they fall in to a hole you can say it's not my fault, I told you so. that's why every month for the last 25 years I've pointed someone at the footnote on page 6 of grfguide :-)

10:06 PM

@DavidCarlisle one wrote me about 6 emails concerning stuff covered in the readme. That's so annoying.

@DavidCarlisle can I give half of a tick to you, and the other half to @JosephWright?

@Skillmon oh another answer? I'll look

@Skillmon as long as you don't give the tick to @egreg I don't mind. You can give it to Joseph or wipet:-)

@DavidCarlisle My solution is much shorter and to the point.

@DavidCarlisle wipet uses \futurelet which isn't expandable (though his code is nice), so it's you or Joseph, imho.

@DavidCarlisle Does htlatex automatically read foo.cfg for document foo.tex or do I have to specify it in the command line?

@AlanMunn pass (I have read the docs, but not since last century:-)

10:19 PM

@egreg shouldn't it be \NewDocumentCommand{\MyIfGroup}{+m}{\regex_match:nnTF { \A\s*\cB\{.*\cE\}\s*\Z } { #1 } } in your code?

@egreg it doesn't meet the spec.

@DavidCarlisle and isn't long.

@Skillmon Better specifying all expected arguments, if you aren't doing nasty things. You can make any argument long, if you so wish.

@DavidCarlisle At least you've read the docs. :)

@AlanMunn didn't understand them. the only way I ever got tex4ht to do anything was to ask Eitan Gurari who'd almost always send a config file by return of post:-)

No one mention 58 runs

10:27 PM

@Skillmon I added the edge case, so now it's clearer why to absorb all arguments.

@egreg in that case it makes sense. But in the first case, I don't see any drawbacks. It looks to me like doing \newcommand\myif[3]{\ifnum#1>0\expandafter\@firstoftwo\else\expandafter\@second‌oftwo\fi{#2}{#3}}.

@DavidCarlisle Yeah, I'm not having any luck with them either. They're very cryptic.

@DavidCarlisle Makes the tabulary documentation seem like a dream. :)

@AlanMunn hebrew actually rather than cryptic

@Skillmon That's a team policy.

@DavidCarlisle Not any more. They're now Czech. Similar lack of vowels though. :)

10:34 PM

@egreg is there any performance difference in the two (I never can tell what performs fast and what slow in TeX)?

@Skillmon Well, slightly slower when you absorb the arguments just to pass them to another macro. But definitely clearer under the expl3 approach.

This community is infecting me — without quite consciously thinking about it (was just picking “random” examples), wrote an answer earlier containing ZZ and ducks: tex.stackexchange.com/a/422629/48

3

@egreg clearer it is, no question.

@ShreevatsaR you should get help to avoid mentioning ducks

@Skillmon try running the regex version and mine with \tracingall to see what they are each doing....

10:51 PM

@DavidCarlisle let me reword that: Clearer top-level code it is, no question.

user image

ZZ Duck

4

haha brilliant!

11:08 PM

@AlanMunn the hat is missing, @samcarter.

@Skillmon Oh, I forgot about that. Neither of the existing hats really do the trick, since it needs a tall rounded top.

@AlanMunn Is there any connection between the ZZ and the person that is always using \zz in macros?

@marmot Well that was @ShreevatsaR 's connection; I've turned that into an 80s rock icon duck.

11:36 PM

@AlanMunn why 80s? They still play.

@AlanMunn and the song is called "Gotsta get paid" not "gotsa get paid".

« first day (2699 days earlier) ← previous day next day → last day (2221 days later) »

Transcript for

Mar '1822

$TeX - LaTeX$ TeX, LaTeX and Friends

\begin{chat}

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

$TeX - LaTeX$

site design / logo © 2024 Stack Exchange Inc; legal

mobile