TeX, LaTeX and Friends

I have basic question on tickz. I read it uses pt for units. So when I write this:
\documentclass{article}

\usepackage{tikz}
\begin{document}
\begin{tikzpicture}
\draw [<->] (12,0) -- (0,0) --(0,12);
\end{tikzpicture}

\end{document}

Then the length of the lines should be 12pt? But then I read that in Latex a pt is 0.013 inch. So the above will make a line that is 0.165inch. But when compiling the above, I see it extend to almost the full page width. Which is about 8 inch in physical size. So I am confused.

..I think the default units are cm, not pt. I did read pt somewhere, but lost the link now. ok, I am set.

cis

2:54 AM

@nasser: Basic tikz-unit is cm. you could use \begin{tikzpicture}[x=1in,y=15mm] or something else.

Nasser

3:20 AM

@cis thanks! Yes, cm it was. I am learning tikz now, it is nice actually.

Joseph Wright

6:05 AM

@ChristianHupfer As expected, those votes from yesterday got reversed

Christian Hupfer

6:39 AM

@JosephWright: Thanks a lot!

@JosephWright: I know who did the downvoting... it's possible to track back ;-)

Have a nice day... I've got an appointment with an architect, for my own home ...

Paul Gessler

7:18 AM

Very exciting @ChristianHupfer! When will it be built, soon?

Joseph Wright

@ChristianHupfer Yes, I'm sure you do

@ChristianHupfer I will keep an eye on this

Christian Hupfer

7:57 AM

@JosephWright: ;-)

@PaulGessler: Well, actually, the house is already built -- my parental home, a typical German house with t wo floors and a huge attic... My father wants to give the house into my custody, but there is no place for me to live in, so I want to reconstruct the attic -- and German Law requires an architect for this :-(

So, I am off again... the architect will be here in a few minutes

Have a nice day

Joseph Wright

8:21 AM

Right, i've shipped siunitx v2.6: now wait for the queries about the changes!

egreg

8:32 AM

@JosephWright :) How many incompatible changes?

Joseph Wright

@egreg None incompatible, but some changes to the 'out of the box' behaviour

@egreg In particular, I've switched back to 'always math mode' and altered the space between a number and a unit (to a full width one)

egreg

@JosephWright I disagree with the latter: it should be a thin space.

Joseph Wright

@egreg That's what I thought, but I've checked the official position

@egreg Essentially, there is no support in the BIPM docs or similar for using a space other than a 'standard' one. The user may decide that they wish to emphasise the link between the number and unit by using a thin space, but I can't have that as the standard setting without some official justification.

egreg

@JosephWright Manuals of style? Decades of typography?

Joseph Wright

@egreg NIST are the other people I check: they again simply say 'space' (physics.nist.gov/Pubs/SP811/sec07.html)

egreg

8:42 AM

@JosephWright Just because this comes from the typewriter era.

Joseph Wright

@egreg Publishers in my area use a standard space

@egreg Can you point to anything official?

egreg

@JosephWright Of course not: as I said, those directions were written during the typewriter era, so there was no choice.

Joseph Wright

@egreg Doesn't make any difference, and more importantly they are aimed at 'proper' publication not just manuscripts

@egreg OK, I'll reconsider :-)

egreg

@JosephWright I can't look now, but I'll check with “Physical Review” whether they use a full or thin space.

Joseph Wright

@egreg I took a look at Nature: I think you have a point there

@egreg OK, revised release sent to CTAN: don't say I don't listen ;-)

@egreg I'd actually spotted an issue with the first attempt anyway

I did say 'wait for the issues to come in' :-)

Does anyone know a guy called 'Günter Milde'?

Johannes_B

8:59 AM

@JosephWright German user, very active

Joseph Wright

@Johannes_B Ah

@Johannes_B He left a couple of comments on my blog (texdev.net/2014/07/10/…), and I'm wondering if I should try to talk to him directly

Egreg says: This seems to be a job for l3regex: :-)

Johannes_B

@JosephWright He is very active on trennmuster@dante.de concerning hyphenation patterns

egreg

@JosephWright I'm waiting for it to exit the experimental stage.

Joseph Wright

@egreg Tricky one

@egreg I'm not sure it ever will

egreg

@JosephWright I'll talk about it at the GuIT meeting

@JosephWright Why not? There are several use cases.

Joseph Wright

9:04 AM

@egreg Raise on LaTeX-L: there are a couple of questions

@egreg Question 1: does this belong at kernel level at all (part of expl3)?

@egreg Question 2: what should the interfaces actually be?

@egreg Need to sort out some of the string stuff first

@egreg If this looks like a good idea, I can move up my priority list a bit

Johannes_B

@JosephWright I think it would be a good idea to adress the list directly, there are some others active, including Werner Lemberg, Herbert Voß and Stephan Hennig.

egreg

@JosephWright Q1: probably not part of the kernel, but an official package anyway. Q2: I'm not dissatisfied by the current interface.

Joseph Wright

@Johannes_B Huh? Which thing are we talking about?

@egreg On the interface, I'm mainly thinking of things like \regex_new:N (which I'm not so sure about)

Johannes_B

@JosephWright Instead of contacting Günter Milde, i think contacting the list (trennmuster, see above) is more productive.

Joseph Wright

@Johannes_B On the 'what level' part, I'm in two minds

@Johannes_B Ah

@Johannes_B The question at hand wasn't about hyphenation :-)

@egreg Like I say, probably one for LaTeX-L

egreg

9:11 AM

@JosephWright Having precompiled regexes can come handy. Consider a case where one needs to check whether an argument is a <factor> according to TeX syntax.

Joseph Wright

@egreg Yes, I can see that

@egreg It's the part about whether this is an independent data type or just a tl that's more tricky. Same with 'stored' strings.

Johannes_B

@JosephWright I havent't read your post, but the comments. I still think this is relevant for the list as they are discussing things like case changing as well.

Joseph Wright

@Johannes_B OK, I'll perhaps write something over the weekend

egreg

@JosephWright Well, it's definitely not a tl, in my opinion.

Joseph Wright

@Johannes_B Will take a while to translate into German

Johannes_B

9:13 AM

@JosephWright No need for translating it, i think.

Joseph Wright

@egreg In a sense no, but a tl is simply an unstructured holder for 'stuff'. The problem is there are loads of special case data types, and if you say each one needs to be fully defined you need a lot of function names, etc.

@Johannes_B OK, I'll think about it

egreg

@JosephWright You'd still need \regex_set:Nn

@JosephWright Additional packages may define their own data types.

@JosephWright Bruno has shown functions to deal with array manipulations; a package out of them would define the array data type.

@JosephWright Also a sequence is a tl, after all. Defining a data type for them hides the implementation details. And defining new data types is very common in programming languages for that precise reason.

percusse

@egreg May yes but it is still complicated for me.

My experience as an amateur is that it's not so straightforward to handle them. I tried for vectors and matrices

Bruno has an asnwer somewhere about it too

38

A: Perform matrix operations (addition, product, transpose, etc.) in LaTeX?

Here is some code to manipulate matrices of any size. Currently, it can perform additions, subtractions, and multiplication (as well as fetching individual entries, and transposing a matrix, for instance). Entries are floating points that l3fp supports (16 digits of precision). % Programming-l...

Joseph Wright

@egreg Yes, I know: it's a question of where the balance is

egreg

9:33 AM

@percusse This is the code I referred to.

David Carlisle

@ChristianHupfer I could sell you one:-) nag.co.uk/numeric/mb/design.asp

egreg

@JosephWright The question is: can something be used as a tl? If not, then it's a different data type.

\regex_new:N \my_reg_ex_regex
\regex_set:Nn \my_reg_ex_regex { [01].* }
\tl_show:N \my_reg_ex_regex

> \my_reg_ex_regex=macro:
->\__regex_branch:n {\__regex_class:NnnnN \c_true_bool
{\__regex_item_caseful_equal:n {48}\__regex_item_caseful_equal:n
{49}}{1}{0}\c_false_bool \__regex_class:NnnnN \c_true_bool {\__regex_prop_.:
}{0}{-1}\c_false_bool }.

@JosephWright Hard to say it's unstructured input. ;-)

tohecz

@PauloCereda Double or halve the number of backslashes depending on your context ;)

Christian Hupfer

9:49 AM

@DavidCarlisle: Ah, a Matlab library? Thanks, but I am not working with Matlab any longer (well, about 15 years now )

percusse

@DavidCarlisle Do you also base those on BLAS and other usual suspects? Or it's top secret tech?

Christian Hupfer

@Johannes_B: Good morning

David Carlisle

@percusse the nag library can use either a system blas/lapack (eg acml or mkl) or its own implementation although actually in the matlab version we use the same blas/lapack as matlab does (otherwise you get symbol clashes) and don't expose the blas interfaces (as they don't really add anything in matlab) (Some of the original blas and lapack work was done here)

percusse

@DavidCarlisle yes I saw jack dongarra's name somewhere there

@DavidCarlisle That's really cool stuff what you are doing there :)

Joseph Wright

@egreg I see that

percusse

9:57 AM

By the way I heard from mathworks guys that finally in 2015 the grids will be adapting the zoom level with @handle2 in the plots

Joseph Wright

@egreg What I'm trying to get at is there are lots of cases where you need a piece of data in a particular form, but deciding if they are different data types is not so easy. The str case is probably more tricky: do we need str to cover what e.g. \edef\foo{\detokenize{bar}} produces?

Christian Hupfer

Till later.. bye

percusse

10:14 AM

Ehuehe, I got a downvote for a very old answer. New generation don't appreciate basic TikZ anymore :)

Johannes_B

@ChristianHupfer Good Morning

egreg

@JosephWright \tl_set_string:Nn

@percusse Sorry, already upvoted.

Joseph Wright

@egreg That's the thing, isn't it: currently Bruno has a whole \str_set:Nn architecture that I'm far from sure about

@percusse Odd

Christian Hupfer

@JosephWright: Sorry, now I have the reverse problem: Someone is upvoting my answers in a row

Joseph Wright

@ChristianHupfer Not so easy to track down: wait and see if it gets reversed

Christian Hupfer

10:25 AM

@JosephWright: Yes, I see... ok, some answers get more than just one upvote in a few seconds.. this can't be the same user

percusse

@egreg Haha, thanks. That reminds me that I still have to fix that hand-drawing thingy. Pfff, that's a tough one.

egreg

@JosephWright Something is a tl when \meaning produces just a representation of the contents.

percusse

@JosephWright No problem at all. :)

Joseph Wright

@egreg Reasonable position. It comes down to whether you feel that a function which requires a string-like tl therefore needs a separate data type or not. I don't (it's too open-ended).

@egreg Nice defintion

egreg

@JosephWright A string does, a regex doesn't, like a seq

percusse

10:27 AM

@ChristianHupfer Or you uncovered the secret of popularity

Joseph Wright

@egreg Yes, I see that

@egreg I like this but as \tl_set_str:Nn (we always use str)

Christian Hupfer

@percusse: Me and popular: Good joke! :D

Joseph Wright

@egreg I'll prod the rest of the team on this

Paulo Cereda

@ChristianHupfer guilty (for some). :)

I said the Voting Squad was afoot.

Apparently no one listens to me. :P

percusse

@PauloCereda When did that happen ? I didn't see the cloudlight

Paulo Cereda

10:29 AM

@percusse <3

Christian Hupfer

@PauloCereda: Well... no, do not just upvote my answers ... I feel embarassed, but thanks anyway

This looks so weird .. +100 in about 10 min :D

@PauloCereda: I hope, you kept some votes for David? ;-)

Paulo Cereda

@ChristianHupfer He has no recent answers.

Christian Hupfer

@PauloCereda: Do you have upvoted all the other ones already? ;-)

Paulo Cereda

@ChristianHupfer Yep. :)

Christian Hupfer

@PauloCereda: You're a vote - addict

Paulo Cereda

10:33 AM

@ChristianHupfer I am surely not. /says the top voter of the whole StackExchange universe

Christian Hupfer

Sorry, must go for half an hour or so.... see you later

Paulo Cereda

@ChristianHupfer talktoya!

Christian Hupfer

@PauloCereda: There must be a Platinum badge for you ;-) It can't be earned only once and by one person only

egreg

@PauloCereda How are you today?

Paulo Cereda

@egreg I'm feeling better, thanks. :) Yesterday was quite complicated, I had to lay down because my back was hurting. Now it seems the flu is in the final stage.

Christian Hupfer

10:36 AM

@PauloCereda: Hope, you will be better finally !

Bye...

egreg

@PauloCereda Going to SP with the flu surely didn't help.

Paulo Cereda

@egreg You are absolutely right. :(

David Carlisle

11:42 AM

@ChristianHupfer excellent thought, shame @PauloCereda didn't have same consideration.

Paulo Cereda

@DavidCarlisle oy

:)

Christian Hupfer

I hope this architect will arrive in some minutes -- I don't like waiting

@DavidCarlisle: My thoughts are always excellent /irony off

David Carlisle

@ChristianHupfer of course

Christian Hupfer

@DavidCarlisle :D

Oh, the door bell... see you later

Paulo Cereda

12:02 PM

@DavidCarlisle: how about writing more answers? :)

LaRiFaRi

12:20 PM

was the second question in his updated post a joke? tex.stackexchange.com/q/199661

never mind... reproduced his desired output as good as I could

David Carlisle

@PauloCereda no point if you have no votes left

Paulo Cereda

@DavidCarlisle :P

David Carlisle

@PauloCereda have you read ISO/IEC 40314 yet? It's an excellent read.

Paulo Cereda

12:46 PM

@DavidCarlisle o.O

Christian Hupfer

1:07 PM

Bye, till later

Paul Gessler

2:24 PM

Is there an established go-to dupe for viewer-related problems with shaded cell backgrounds "covering/obscuring" cell borders?

I thought there was, but I can't find it atm.

The question in question:

0

Q: Filling a cell of the table with color and preserving the border

I used \cellcolor{} to fill the cell of my table. However, the black borders of the cell disappears. The tex viewer shows borders, but I don't see the borders when I open as a pdf. Can you help me about it? Thanks. \begin{table}[h] \centering \begin{tabular}{|c|c|c|c|c|} \hline \backslash...

egreg

Duplicate (old one, I already voted time ago): see OP's comment

2

Q: "Literal" numbering of chapters but not in contents

Thanks to this macro \makeatletter \let\old@makechapterhead\@makechapterhead \def\@makechapterhead{\def\thechapter{\words{chapter}}\old@makechapterhead} \makeatother \newcommand\words[1]{\expandafter\xwords\csname c@#1\endcsname} \def\xwords#1{\ifcase#1\or one\or two\or three\or four\or five\...

table-of-contents numbering

moose

Hello everybody

Does anybody know if people use \rightarrow, \longrightarrow and \shortrightarrow in a semantically different way?

egreg

7

Q: What is wrong with Comic Sans?

There's a great TEDxExeter talk by a colleague of mine, Simon Peyton-Jones, about the recent advances in the English lower school 'computer science' curriculum: http://youtu.be/Ia55clAtdMs Like all of his slide decks he uses Comic Sans throughout. Depressingly, though inevitably, one of the YouTu...

fonts

eiterorm

2:40 PM

@moose I know there are various different uses of arrows in group theory, but from what I gather, not everyone agrees on the different uses.

Christian Hupfer

Back.. Good MAEN;-)

moose

@eiterorm That's interesting! Could you please tell me what \rightarrow, \longrightarrow and \shortrightarrow are used for in group theory (and one author that does so), please?

eiterorm

@moose I don't know much about it, I'm afraid, hence the rather vague answer. On different math forums and in other online articles, I've seen e.g. \rightarrow used in place of \mapsto.

moose

Ok, I know that \rightarrow and \mapsto are used different. But I rather wanted to know if \rightarrow, \longrightarrow and \shortrightarrow are used different. So the question was if the length of the arrow contains information or if it is just used when people think it looks better.

Christian Hupfer

@moose: I am not into university any longer, but I remember, that 10 mathematicians/physicists etc. use 10^2 different notations .... there might be typographical aspects, what should be better in this or in that case, but it's often a matter of taste. Somehow annoying, somehow it's nice, since there is still some free will ;-) (Ok, there might be real weird example, but @egreg will show them the typographical rules ;-) )

eiterorm

2:50 PM

@moose I've seen people prefer the use of longer arrows when they want to visualize more than a single bijection, if that helps.

egreg

@moose The consensus seems to go to \rightarrow for a map, \mapsto for its action on elements. But it's not universal.

The arrow length doesn't convey information. Beginners tend to use longer arrows, just like they don't want wide margins.

moose

Ok, thank you very much!

Christian Hupfer

@egreg: I silly question of mine, perhaps I could/should look into the questions on TeX.SX! Is there a Plain TeX \if... command in order to test, whether a macro, say \somemacro already exists? I don't mean etoolbox etc. commands of course?

egreg

@ChristianHupfer \ifdefined\somemacro (it's of e-TeX, actually), or (reverse), \ifx\somemacro\undefined (assuming \undefined is undefined).

Christian Hupfer

@egreg: I don't mind if's e-TeX (There is hard to find a non-e-TeX compiler nowadays, anyway!) Thanks!

Joseph Wright

3:04 PM

@ChristianHupfer Very easily: tex is always Knuth's TeX without any extensions

Christian Hupfer

@JosephWright: Perhaps my naming was misleading, sorry -- of course, I meant TeX as shipped out with TeXLive/MikTeX. It's e-TeX then, or am I completely errant? I am quite happy with e-TeX ;-)

Joseph Wright

@ChristianHupfer TeX systems today ship with tex, pdftex, xetex and luatex. The first is Knuth's TeX unchanged: the other three have the extensions. pdftex the binary is used as the default engine, in DVI mode to implement e.g. latex and in PDF mode for e.g. pdflatex.

@ChristianHupfer Extensions are enabled by default for all engines bar tex

@ChristianHupfer Those of us who build formats from time to time might have to do pdftex -ini -etex :-)

Christian Hupfer

@JosephWright: Ok ,thanks again.. I was still using wrong terms... I always use pdftex, so I know its e-TeX ready ;-) And pdflatex of course...

Joseph Wright

@ChristianHupfer No problem: this is somewhat tricky to follow

Christian Hupfer

@JosephWright: I'll build my format in my next life ;-)

Christian Hupfer

3:55 PM

@PauloCereda: I managed it: David answered a question :D Just upvote him ;-)

Paulo Cereda

@ChristianHupfer I need my votes back. :)

Christian Hupfer

@PauloCereda: Oh Paulo... :D

@JosephWright,@DavidCarlisle,@egreg,@Werner: Thanks for the answers ... makes me feel dumb ...

Joseph Wright

@ChristianHupfer No problem: more-or-less free rep :-)

egreg

@JosephWright What rep? ;-) Ah, you were talking about @DavidCarlisle who needs it.

Johannes_B

There was a comment addressed to me, but the question was deleted. As i have so very low rep, can anybody tell me if there was anything important in the comment? -> tex.stackexchange.com/questions/199657/…

Christian Hupfer

4:07 PM

@Johannes_B: I look after

@Johannes_B: Shall I copy it or provide a screenshot?

Johannes_B

@ChristianHupfer A copy seems sufficient

Christian Hupfer

This was the comment
@Johannes_B, seems to be a better alternative in KOMA script there. It is very interesting if you could upgrade your comment to an answer though.

Johannes_B

@ChristianHupfer Ha, well. ;-) There are so many on addchap lately.

Christian Hupfer

@Johannes_B: I must admit, that I am responsible for that chapterthumb questions -- somewhat, but the whole thing was blown up by requests of the OP to turn a screw there and another one there...

Johannes_B

@ChristianHupfer i never looked at chapterthumbs.

Christian Hupfer

4:12 PM

@Johannes_B: Well, that .sty is not meant to be extended really. The original code by me did not look like it was posted by the OP today... he made a lot of statements inside it.

@JosephWright: ;-)

@PauloCereda: David got already 5 upvotes ;-)

Johannes_B

@ChristianHupfer Have a look at koma-script.de. Markus gives support there. There might be any new versions that never made it to CTAN.

David Carlisle

@egreg I don't need rep:-)

3

Christian Hupfer

@Johannes_B: Well, yes, on occasion -- as I said -- I had to abuse chapterthumb.sty to fulfill the OP's requests, but anything after that is not in my responsibility :D

egreg

Speaking of rep! Harish report: -365 (rep capped for today). There will be a party on Sunday, I guess.

6

Christian Hupfer

Of course nobody cares about rep here ;-)

Johannes_B

4:24 PM

Something i noticed: There are many questions on bibliographies (and changing their visual appearance) in german. Opposed to it, there often arises the question in english-speaking places, why picture appear where the smurf they want.

David Carlisle

4:37 PM

@egreg I hope you'll be arranging a party for other people passing multiple of 100k barriers?

@Johannes_B Germans do well referenced research but Englishmen just like to look at pictures?

2

Paul Gessler

@DavidCarlisle Englishmen just like to look at pictures located on the page where they damn well please. :-)

Johannes_B

@DavidCarlisle Well, that could be an explanation ;-)

Paulo Cereda

5:35 PM

I learned in QI that the French don't believe in the existence of custard. Just crème anglaise. :)

egreg

@DavidCarlisle Is somebody doing it soon? :P

Paulo Cereda

@JosephWright: will get a PS4? :)

Johannes_B

6:51 PM

@ChristianHupfer There are some special users. It begins all harmless, but then ...

latex-community.org/forum/…

Christian Hupfer

@Johannes_B: I will take a look immediately.... warp 9.9999999 ... engage

eiterorm

My thing. It works. Celebration!

Now for the next thing.

Johannes_B

@eiterorm Yeay

Christian Hupfer

@eiterorm: What thingy?

eiterorm

@ChristianHupfer Code.

Christian Hupfer

6:58 PM

@Johannes_B: I did not use doxygen for years now... I have never used doxygen package outside of code documentation

eiterorm

@ChristianHupfer I'm doing image processing, and I've been struggling to get all the indices and conditions right.

And now, onto the next function. =P

Johannes_B

@ChristianHupfer Used the package for the first time today. Horrible

Christian Hupfer

@Johannes_B: I am not sure whether the is designed for direct usage at all... Doxygen is a doc generator for html/tex mainly, documenting C/C++/Java(?) and Qt - Code

Johannes_B

@ChristianHupfer He said he is documenting code, and up till now he was talking about autogenerated code. btw: this is the backseat banana guy.

Christian Hupfer

@Johannes_B: Yes, my name-recognition-synapses were alarmed ;-)

@Johannes_B: Let us call him the Banana-On-The-Backseat-Guy

Johannes_B

7:05 PM

@ChristianHupfer If i remember a username it's one that got my intention. There are only two ways to do that, being good nice and funny; or beeing stubborn and uneinsichtig.

@ChristianHupfer He never replied to my request for information.

Christian Hupfer

@Johannes_B: ...and probably will never do;-)

Johannes_B

gnumeric cookbook.xls

Christian Hupfer

@Johannes_B: New liver - recipes? ;-)

Johannes_B

@ChristianHupfer It's a very special cookbook ;-)

Christian Hupfer

@Johannes_B: How's the dog?

Johannes_B

7:10 PM

I'll be out for a while with the dog <- @ChristianHupfer Fine ;-)

Christian Hupfer

@Johannes_B: See you later then...

Johannes_B

7:36 PM

@ChristianHupfer Powerwalk. The dog got my knee right in the head. Bam.

Christian Hupfer

@Johannes_B: Wait a minute, I am just looking for the phone number of Animal rights group ;-)

Johannes_B

@ChristianHupfer Better than having a real fight with a two times bigger dog. My Whisky tried to walk through me, Fail.

Christian Hupfer

@Johannes_B: Stupid dog ;-) .... the dog, not you ;-)

Johannes_B

@ChristianHupfer Now he is looking at me.

Christian Hupfer

@Johannes_B: Hm... I thought he is somebody's else dog, not yours?

Johannes_B

7:45 PM

@ChristianHupfer He is not mine. Why do you ask?

@ChristianHupfer The owners are on vacation, i am staying at their place so the dog is in his usual environment.

And under my close watch. ;-)

Christian Hupfer

@Johannes_B: Because you wrote "My Whisky" ... misunderstanding of mine

Johannes_B

@ChristianHupfer Now that you are setteling down. Plans for a dog? Or do you already have one?

Christian Hupfer

@Johannes_B: Settling down...sounds ... spießig:D

Johannes_B

@ChristianHupfer Sind wir nicht alle ein wenig Bluna? ;-)

Christian Hupfer

@Johannes_B: No, I am not a dog-owner or will be ... I like dogs, but I prefer cats...

@Johannes_B: Bluna aus Buna ;-)

Johannes_B

7:51 PM

@ChristianHupfer youtube.com/watch?v=UiYGX4Wbgtc

Christian Hupfer

@Johannes_B: But we don't have a cat so far... We are living at a street with much traffic and not much garden and green yards, not a good place for a cat (or any other animal)... or a human Landei :D

Johannes_B

@ChristianHupfer To be honest, i don't care. I can take care of dogs, cats, children.

Christian Hupfer

@Johannes_B: I hope, you are not confused with the corresponding rules

Johannes_B

@ChristianHupfer I am confused, but that's not your fault.

Itś the Banana-on-the-backseat-guy

Christian Hupfer

@Johannes_B: It's never my fault :D

Stefan Kottwitz

7:56 PM

@Johannes_B Great! You can have my dog and the two children for a day or so :-)

Johannes_B

@StefanKottwitz I think all at once would bring me in trouble ;-)

49 mins ago, by Johannes_B

gnumeric cookbook.xls

Christian Hupfer

@StefanKottwitz: :D

Stefan Kottwitz

@Johannes_B Now all three are sleeping finally, yesss :-)

Johannes_B

@StefanKottwitz Want to have another kid? Try the doxygen user. ;-)

Christian Hupfer

@Johannes_B: That's not fair to compare the Banana-On-The-Backseat-Guy with a kid... kids are clever ;-)

Johannes_B

8:00 PM

I know it's not right to talk about people behind their backs, but ...

Christian Hupfer

@Johannes_B.. it makes fun?

Johannes_B

@ChristianHupfer he might be clever as well, but is distracted by other things.

Stefan Kottwitz

@Johannes_B Oh, it's great that already you are so patient there

Christian Hupfer

Sorry, I am off for a while ... have a nice time ...

Manuel

    \expandafter\expandafter
    \expandafter\endgroup
    \expandafter\the
    \expandafter\tmpbtoks
    \expandafter{\the\tmpatoks}

Johannes_B

8:07 PM

@StefanKottwitz Our user KeksDose. Am i right in suspecting?

Manuel

What would the expl3 approach to that? Assuming we have available the \toks_set:Nn, \toks_use:N etc. and family (functions to set toks, and also variants wit h :o :x etc.).

Stefan Kottwitz

@Johannes_B I don't know, what, I just read the one answer some days ago

Manuel

By the way, although may be not supporting it, or even no documenting it. Wouldn't it be useful to have an interface for \toks?

Johannes_B

@StefanKottwitz I think this is the same user posting as a guest sometimes.

Stefan Kottwitz

@Johannes_B Aha, if yes, fine

David Carlisle

8:12 PM

@Manuel why? there used to be:-)

@Manuel given etex \unexpanded there really isn't any time that you need a token register where you could not use a macro

Manuel

@DavidCarlisle I know. I had to search in the web, to obtain a .dtx which I don't know how it works, and then search google to obtain a .sty file from 2011. However, most of the work can be done with token lists (macros), but still there are sometimes where toks might be useful.

@DavidCarlisle If you want to save #1 in a token list you need ##1, or is there a way?

David Carlisle

@Manuel vvv

$ etex
This is pdfTeX, Version 3.14159265-2.6-1.40.15 (TeX Live 2014) (preloaded format=etex)
 restricted \write18 enabled.
**\edef\foo{\unexpanded{#1}}
entering extended mode

*\show\foo
> \foo=macro:
->##1.
<*> \show\foo

?

@Manuel essentially \unexpanded is an anonymous token register just as \dimexpr is an annoymous dimen register

Manuel

@DavidCarlisle Mmm…

@DavidCarlisle But then, how does one uses \tl_set:Nn \l_tmpa_tl {#1}

@DavidCarlisle Isn't there an “interface” and you need \tl_set:Nn \l_tmpa_tl {\exp_not:n{#1}}?

egreg

@Manuel There's no function for manipulating token registers. You say \tl_set:Nn \l_tmpa_tl {#1} and, if you don't need it expanded more than once you use, for instance, \exp_not:V \l_tmpa_tl

Johannes_B

8:35 PM

@StefanKottwitz Oh i just love pdfs generated by excel/libre office/gnumeric etc.

@StefanKottwitz tiful documents Hä? Ah, there is a beau on the next page. ;-)

Manuel

@egreg Mmm… I have to think about it a bit. I'm trying to write the “\bgroup … \egroup argument” code in expl3 to understand it better.

Johannes_B

@StefanKottwitz At which userbase is the current project based?

egreg

@Manuel You surely don't need that mess. ;-)

@Manuel Some more context is needed.

Stefan Kottwitz

@Johannes_B LaTeX users with some experience who already read an introductional book or have one

@Johannes_B So either very nice and recommendable should-know-basics or beyond

doesn't need to be hard - can be simple if just not included in normal intros

Manuel

@egreg It took a while to process to my mind, but now it's (almost) fully translated to expl3 and is more understandable. Just grabing tokens one by one, in case you enter a group do something extra until you end that group, and keep going until you find something that means exactly the defined delimiter you are looking for.

Stefan Kottwitz

8:48 PM

@Johannes_B showcase like, hands-on like, practical, not really teaching but all will be explained

Manuel

@egreg No context, I said that it was just curiosity. Now I can do \begin{lcase} Code To MAke LOWerCASE \end{lcase}. Nah, just a joke.

Stefan Kottwitz

@Johannes_B The most efficient is doing an impressive and useful example which simply uses a new package and so is pretty short :-)

Johannes_B

9:08 PM

@StefanKottwitz I just read the outline in more detail. I have some remarks, and even one or two additions. I try to make them into real words tomorrow.

Christian Hupfer

Have a nice time ... good MAEN etc ;-)

Stefan Kottwitz

@Johannes_B Great!

Joseph Wright

9:31 PM

@Manuel We dropped it because you really don't need a public toks interface with e-TeX: this was not a snap decision

@Manuel The only place we use toks is some tricks by Bruno inside l3regex, and that is all done by number and inside a group

@Manuel \exp_args:NNV \endgroup \tmpbtoks \tmpatoks, which used to look like \exp_args:NNNo \group_end: \toks_set:Nn \l_tmpb_toks { \toks_use:N \l_tmpa_toks } (I don't think we had V-type expansion when we were still using the toks module).

@Manuel If we still had it, would be \exp_args:NNNV \group_end: \toks_set:Nn \l_tmpb_toks \l_tmpa_toks

@PauloCereda Unlikely

@Manuel That will work, assuming you are not inside another function

@Manuel \tl_set:Nn uses the trick @DavidCarlisle points out: this was something I pushed for, and in general I think it's worked out :-)

@Manuel Note that one reason for the V-type argument was originally that it allows both macro and toks based variables to be expanded without needing to know what the underlying implementation is

Manuel

@JosephWright Okey. Thanks for the info (all of them).

@JosephWright Also in a hopefully to come environment declaration like environ command, isn't it?

@JosephWright However I'm not sure, but you need to expand also the \tmpbtoks, does \exp_args:NVV work? Yes, you did have V at least in the l3toks.dtx file in GitHub.

Joseph Wright

@Manuel No, to assign to \tmpbtoks you don't want to expand it

Manuel

9:46 PM

@JosephWright No, the code I uploaded has two \the, no assignment.

Joseph Wright

@Manuel I saw that, and assumed that the first \the is present in error (otherwise what are you up to)

Manuel

@JosephWright The question was meant: \exp_args:NVV works? Or do I need somehow to expand first the \tmpatoks and then \tmpbtoks (which I don't know how to do in expl3). I'm not sure what I'm up to, but the thing is that, before ending the group, I want to expand two token lists (or toks) with the second inside a group \endgroup\tmpb{\tmpa}. Then, close the group, leaving <contents of \tmpb>{<contents of \tmpa} and let the typesetting continue.

Joseph Wright

@Manuel \exp_args:NVV will expand the value of the two variables, leaving the results in braces in the input stream. Thus \exp_args:NVV \foo \tmpatoks \tmpbtoks becomes \foo { <content-of-tmpatoks } { <content-of-tmpbtoks }

@Manuel We've never (or at least extremely rarely) needed to do that: the much more common case is \endgroup \assignmentcmd \variable \stuff_to_assign_to_variable

Werner

This will now become active:

13

Sometimes, when users improve a post through the review interface, they see a reasonably poor edit suggestion, improve it and leave it ticked because they've either forgotten to remove it, don't know what it does or can't be bothered to. This can result in awarding reputation to suggested edits ...

discussion feature-request review suggested-edits user-interface

Manuel

@JosephWright Then I just thought… nested \exp_args or may be a tricky \exp_args:NV… \use:n? The problem is that \tmpbtoks and \tmpatoks may have whatever inside of them. It's not necessary an asignment, it's just adding braces around the second one and leaving it there. And both are just defined inside the group (that's why I need to expand them before closing the group).

@JosephWright This is getting too long and “breaking” the chat. Don't worry, I will investigate. Thank you.

Joseph Wright

9:55 PM

@Manuel I need to see a real case: this feels like the wrong approach to a problem to me

Manuel

5

A: Arguments possibly delimited by \bgroup and \egroup

The question gives me a sense if it is read from its end: give the possibility of creating a macro which expands its parameter during parameter scanning. Then the variants } or \egroup as a delimiter of the parameter is serviceable. I've created the \eparam macro with this syntax: \def\mymacro ...

@JosephWright Crazy thing. With no use. I'm just trying to translate it into expl3 and trying to understand it.

Joseph Wright

@Manuel Well that's a bad idea to start with, so ...

egreg

@Manuel Which is a really bad idea to begin with. :)

@JosephWright Hey!

Manuel

@egreg @JosephWright Yeah, both!

Joseph Wright

@Manuel Remember that for some low-level stuff the team have to use primitive syntax, for example read the FPU, regex engine, ...

Manuel

9:59 PM

@JosephWright All of it is more or less straightforward, really. \tl_map_function:NN, \token_if_expandable:NT, \token_if_space:NT, \token_if_group_begin:NT, \peek_after:Nw… and all of that divided in 8 or 9 auxiliar commands.

The only thing that I don't know how to do is that code I posted above, which seems a lot more TeX (all of those \expandafters) just with “expl3 names” (\exp_after:wN).

@JosephWright Would this work? \exp_args:NNVV \use:nn \group_end: \l_tmpb_toks \l_tmpa_toks

@JosephWright Or may be it's easier to expand inside the group and then use a \group_insert_after? I will think about it and probably post an answer to that question just translating that solution into expl3 (which is muuuch more readable for me).

@JosephWright Nah, unnecessary crazyness. \edef\gtmp{\the\tmpbtoks{\the\tmpatoks}}\engroup\gtmp

Johannes_B

10:29 PM

Sometimes, giving support is like youtube.com/watch?v=jl3uBNmpf10 please concentrate

eiterorm

@Johannes_B "Shut up!"

Johannes_B

@eiterorm :-p

egreg

11:28 PM

@ChristianHupfer Serial downvoting reversed!

Transcript for

$TeX - LaTeX$ TeX, LaTeX and Friends