@ChristianHupfer I got some "infoboxes" to work with a manual solution. Are you interested in the result? I can send you an example document.

Guys, I want to write an etex rule for arara. Any hints on a good command line invocation?

@egreg: is there any engine restriction for frontespizio?

@PauloCereda ?

@Werner thanks for the retag; I saw the main suggestions pop up immediately and didn't even think to look for others.

I am in need to make some comments to W0rd document right now, but only got LibreOffice. As i am used to vim, i thought i search for any kind of plugin. And what did i find multiple times? Use LaTeX instead!

Gi

*Hi, I'm looking for good practice on how to align equations on the right side of the "=" sign instead of the classical left side. This is particularly important for multiline equations. For now to compensate the spacing I need to manually insert "\,\," after "=&", which is not convenient at all...

@s__C not sure what you mean, but perhaps ={}& is what you are looking for?

@Johannes_B If you are used to vim your brain will be so corrupted really there is no point trying anything. You should have used emacs:-) (ask @PauloCereda)

Yes that's it @DavidCarlisle

But I still wonder why there is a different spacing for $={}& -..." VS any other expression not starting with a sign

@DavidCarlisle Using emacs would make me feel just as dirty as using LibreOffice.

@s__C a binary operator reverts to a mathord if it doesn't have something both sides, compare $-2$ and ${}-2$

@Johannes_B I've used emacs almost every day since 1987 and it hasn't done me any harm (I claim:-)

@DavidCarlisle @Johannes_B Yeah, he's able to make the friendship knot on his fingers :)

@PauloCereda No, but it assumes pdflatex, by default.

@egreg gotta love the usage of \therefore here :D math.stackexchange.com/a/950779/43247

@tohecz Is writing “therefore” really too long?

@egreg no, but everyone can do that. Only morons well-educated people know to use symbols ;)

btw, tempted to suggest an edit on it, with edit reason: "Improved formatting" :)

@DavidCarlisle I meant the command line thingy. :) Like, what flags, etc. should I use. :)

@egreg Got it. :)

@DavidCarlisle And I have some suspects on who defined ~ in the LaTeX kernel. :P

@egreg: is plain TeX also valid for frontespizio?

@PauloCereda No, it assumes LaTeX.

@egreg Got it. I'm adding a list of valid engines to the rule. :)

'pdflatex', 'pdftex', 'latex', 'xelatex', 'xetex', 'lualatex', 'luatex'

@egreg: ^^ do you think it's ok? :)

@PauloCereda Why pdftex?

@egreg Dunno. :)

So xetex and luatex go away as well then. :)

@PauloCereda etex file ?

@DavidCarlisle No options?

@PauloCereda not by default, but it's got all same options as latex --shell-escape etc (since it's the same executable)

@egreg blaming Frank is traditional policy

@DavidCarlisle No justification?

@egreg oh I seem to recall there was a bug that a caption with foo~ bar got written as foo\nobreakspace bar so one space not two so foo\nobreakspace{} bar

@DavidCarlisle That's a good reason. Bad input is everywhere. :(

@egreg we should probably have made the internal command have a delimited . or take an unused argument or something so \nobreakspace. or \nobreakspace{} but with the {} being gobbled.

paulo@alexandria Arara$ java -jar arara.jar enrico_test.tex
  __ _ _ __ __ _ _ __ __ _
 / _` | '__/ _` | '__/ _` |
| (_| | | | (_| | | | (_| |
 \__,_|_|  \__,_|_|  \__,_|

Processing 'enrico_test.tex' (size: 642 bytes, last modified:
09/29/2014 07:50:25), please wait.

(Frontespizio) The engine ............................... SUCCESS
(Frontespizio) The frontispiece ......................... SUCCESS
(Frontespizio) The DVIPS program ........................ SUCCESS
(Frontespizio) The engine ............................... SUCCESS

@egreg: ^^ :)

@DavidCarlisle OK thank you for the hint regarding spacing around signs

@JosephWright xparse has the same defect as siunitx: there's no public interface; \__xparse_declare_cmd:Nnn should be \xparse_declare_cmd:Nnn

@egreg No

@egreg Unlike siunitx, this is deliberate

@egreg Plus there are issues with the name anyway

@JosephWright So it's impossible for a package writer to define a family of commands without using deprecate tricks with \expandafter and \csname; maybe with the equivalent \exp_args:Nc \DeclareDocumentCommand{name}{<arg>}{<code>}, but it's the same.

@JosephWright I don't see why Frank insists in \NewDocumentCommand using an internal (public) version and doesn't do the same with that command.

@egreg xparse is a bit odd as it's not meant to be used at the document layer, it's meant to define the document layer

@egreg Well that's one of the issues: if you are defining a document level syntax you shouldn't be creating commands in an ad hoc fashion

@egreg Of course, this is as I've tried to indicate in various places something that almost certainly still needs discussion and proper examples of 'real life' use cases

@JosephWright Why not? Think to \myalpha \mybeta and so on. This is document level syntax.

@egreg Yes, but you should be documenting this syntax so need to know what commands you are creating

@egreg I guess I'm saying that perhaps 2e is too forgiving in terms of letting the user create 'stuff'

@egreg But as I say, what's likely needed is good real cases to base discussion on

@egreg Are these 'document commands' in that sense? A tricky issue not yet addressed is short-cuts and the like that don't take args or are variable-like

@JosephWright Yes, of course; the definition of those commands will be made by a package, not by a user. And it would be silly having to say \exp_args:Nc...... instead of \xparse_declare_cmd:cnn{my#1}{}{code depending on #1}, with #1 coming from, say, a clist, for defining new shapes for Greek letters.

I know that one can do \cs_new_protected:cpn, but is \cs_new_protected:cpn{my#1}{code depending on #1} a preferable way? I'm pretty sure it isn't.

@egreg I do see that case

@egreg It's probably one for LaTeX-L: Frank may well have some plan in this area, or at least has a lot more experience than I can claim with it

@egreg One issue, which is perhaps not as trivial as it looks, is the name of the module itself. xparse is fine as a 2e package, but is it really the right prefix for public functions at a LaTeX3 format level? I suspect not.

@egreg The thing here is that xparse is mainly about dealing with LaTeX2e-like arg syntaxes. Something like \mybeta presumably doesn't have any args, so it's less clear-cut. I see though that a single 'point of contact' for these defintions is a good thing.

@topskip: yay Pat is here!

Old man is visiting the chat! Hi all

@topskip Hello

@egreg Will you raise on LaTeX-L or should I?

@JosephWright You do. ;-)

@egreg OK, will do this evening

@egreg I'll try to lay out the arguments as best I can

Anything happening in the LaTeX world in the last six months?

(broad question mode)

@topskip TUG2015 will be in Germany

tug.org/tug2015

cool, I'll save the date

@topskip We've made expl3 work on formats other than LaTeX2e

@topskip pgf 3.0 release in that window I think

@JosephWright cool!

that is lolipop, plain and /or context?

@topskip Can I also interest you in expandable case changing? We've added that experimentally to expl3 and I'm hoping for feedback :-) (Not Lua-dependent, BTW)

I am currently so far away from regular (La)TeX usage....

@topskip Lollipop not tested so can't say, plain definitely yes, ConTeXt should be MkII/MkIV but is a bit trickier (MkII doesn't search the tex/latex tree so the code works but might not get found! MkIV should be fine)

@topskip Frank's up to something with river detection:

expl3 on plain sounds good.

@topskip Request from Heiko

@JosephWright I've seen that, @PauloCereda has pointed me to it. I wish I had some more time to implement at least a simple line breaking algorithm

.... the context one seems too impossible to port to plain LuaTeX

@topskip <3

I've had a look at Taco's and Hans' impementation. But no, too complicated without deeper understanding of what's going on

@JosephWright The conference is right at the beginning of our summer holiday... This usually conflicts with vacation, but I try to be there

@topskip Same here

@JosephWright I've been playing with some ideas on linebreaking last week. not rewriting the algorithm in lua, but detecting overfull boxes and rebuilding paragraph with different value of tolerance. and river detection, but that's not perfect yet.

github.com/michal-h21/linebreaking

Volunteers for testing arara 4.0? :)

@PauloCereda Have you already updated the manual?

@egreg Er... no. :(

I'll do that. :)

@PauloCereda A list of the new features?

@egreg I can write one in a couple of minutes. :)

Good maen ;-)

Scanning over Nature takes me to ploscompbiol.org/article/… where TikZ/pgf combination as shown off at TeXample.net gets a mention :-)

Notably, I don't see things like QtiPlot in there

@egreg: This will take way more than I expected: github.com/cereda/arara/wiki/New-feature-highlights-in-4.0

:)

@JosephWright Wow indeed.

@ChristianHupfer Hello. :)

@JosephWright No ducks. :)

Kjell Magne Fauske really gave the initial kick to a lot of people back then.

@JosephWright I particularly like the advice to avoid boxed diagrams (the “sine-cosine” figure).

@PauloCereda: Hello,... sorry I am busy :-P

@1010011010: Sorry, I am interested in your 'manual' solution, but right now the work amount is overwhelming ...

@PauloCereda Yay, multiline directives! :-)

@NicolaTalbot I hope this won't introduce any arara insert hacks ;)

@tohecz :-)

@NicolaTalbot Specially for you. :)

@PauloCereda Ahh! :-) Do you have a binary for v4.0?

@NicolaTalbot I can build you one. :) Just a minute. :)

@Nicola: By the way, look at the easter egg here: github.com/cereda/arara/blob/master/application/pom.xml :)

Hat enthusiast :)

@PauloCereda Duck enthusiast :-)

@NicolaTalbot :)

All these different forms of markup confuse me!

can someone explain me why Czech Airlines can't have a working reservation system?

@Nicola: I think we finally got rid of the filename with spaces issue. :)

@tohecz Adventure is more exciting!

@tohecz It won't be fun if everything works

@PauloCereda Yay! :-)

@egreg An italian and a turkish... This world is doomed

@NicolaTalbot app.box.com/s/o9lzpqzpyud7nnwuq0m8

I think I've just worked out a way of fixing flowfram (after having said it might take me ages!)

@percusse well, yeah. Currently they own me 25 euros. I'm not sure if I won't simply go for EasyJet. It's less comfy for the same price, but unlike CSA, they do want clients.

@NicolaTalbot Yay!

@NicolaTalbot Did you happen to have space travels lately?

@percusse cough cough TARDIS cough cough

@tohecz Why don't you just take the easyJet?

@NicolaTalbot: I would love to include your datatool rules in this new version. :) We might need to tweak it a little. :)

@tohecz Even being terrible takes some experience to excel. EasyJet and RyanAir is becoming very good at being mediocre. I didn't have any problems lately other than the usual crappiness.

@percusse it's much less comfy, especially in Prague, what they do there is ridiculous, even SmartWings do better. As well, I'll surely not fit in 20kg of luggage, which adds more and more to the price

@percusse Oh no, I thought my recent visit to Mars was classified! ;-)

@tohecz Wear two coats and a sweater, fill their pockets with socks and take them off in the plane :) Poor strikes back.

@PauloCereda Yay :-)

@percusse I already hate flying, you obviously want me to hate it even more :p

@NicolaTalbot That's a little too ambitious. You must have incredible handbreak skill to bend time.

@tohecz Being broke is pretty innovative.

@percusse ?

You devise interesting ways to cheat the cheaters

@percusse well, they are not quite cheaters, they just do their business. And as long as they obey all the safety rules, they do a good job I'd say

@tohecz Clearly you don't have any pilot friends which I hope you keep it that way :)

@percusse sometimes it's better not to know.

@tohecz My friend just started in Emirates (previosly was on Transavia). The stories are mind-boggling

@egreg Boxes are not really meant for plotting functions: much more intended for experimental data. Indeed, the pgfplots docs have an example suggesting exactly this.

@JosephWright But boxed function plots are everywhere. :(

@egreg Like pie charts.

Trivia: Florence Nightingale is the inventor of the pie chart.

@PauloCereda Erm... I don't get this error with 3.0 with datatooltk.yaml:

I have spotted an error in rule 'datatooltk' located at
'/home/nlct/.arara/rules'. I could not parse the rule, something
bad happened. Apparently, the provided YAML file is invalid. I
will do my best to help you in any way I can. There are more
details available on this exception:

DETAILS ---------------------------------------------------------
Cannot create property=command for
JavaBean=com.github.cereda.arara.model.Rule@16f5b392
 in
'reader', line 1, column 1:
    !config
    ^
Unable to find

@NicolaTalbot oopsie. :( Could you send me the rule?

Ah! I remembered! command is now deprecated. Now we have:

commands:
- command: <here comes the command>

Ah, okay. This is what I have at the moment:

@Nicola: but I can help with the rule in general. :) There are a couple of features. :)

!config
# DatatoolTk rule for arara
identifier: datatooltk
name: DatatoolTk
command: <arara> datatooltk @{options} @{output} @{csv} @{sql} @{input} @{sqldb} @{sqluser} @{name} @{probsoln} @{sort} @{sortcase} @{seed} @{shuffle} @{sep} @{delim} @{csvheader} @{noconsole} @{debug} @{owneronly} @{maptex} @{xls} @{sheet}
arguments:
- identifier: options
  flag: <arara> @{parameters.options}
- identifier: output
  flag: <arara> --output @{parameters.output}
- identifier: csv
  flag: <arara> --csv @{parameters.csv}

@PauloCereda :-)

@NicolaTalbot Thanks! Gimme a minute. :)

@NicolaTalbot: I rewrote it close to your original code:

!config
identifier: datatooltk
name: DatatoolTk
authors:
- Nicola Talbot
commands:
- name: The DatatoolTk software
  command: >
    @{
        return getCommand('datatooltk', options, output, csv, sql, input, sqldb, sqluser, name, probsoln, sort, sortcase, seed, shuffle, sep, delim, csvheader, noconsole, debug, owneronly, maptex, xls, sheet)
    }
arguments:
- identifier: options
  flag: >
    @{
        if (isList(parameters.options)) {
            return parameters.options;
        }
        else {

I can provide more info on what is happening. :)

Try running with java -jar arara.jar yourfile.tex -n, -n is the dry-run option.

So you can see the command expansion.

Processing 'sample-datatooltk-sql.tex' (size: 1 KB, last
modified: 09/29/2014 19:03:33), please wait.

[DR] (DatatoolTk) The DatatoolTk software
-----------------------------------------------------------------
Author: Nicola Talbot
About to run: [ datatooltk, --output, products.dbtex, --sql,
SELECT * FROM books, --sqldb, samples, --sqluser, sampleuser,
--noconsole-action, gui ]

[DR] (PDFLaTeX) PDFLaTeX engine
-----------------------------------------------------------------
Authors: Marco Daniel, Paulo Cereda

@PauloCereda Thanks! :-)

@NicolaTalbot ooh! Did it work? :)

@PauloCereda Yes :-) I'm going to try the multiline thingy now.

@NicolaTalbot Uh-oh. :)

@Nicola: Once we had mastered the system execution thingy, the sky is the limit. :)

@PauloCereda Yay, this is a lot neater. :-)

% arara: datatooltk: {
% arara: --> output: products.dbtex,
% arara: --> sqluser: sampleuser,
% arara: --> sqldb: samples,
% arara: --> sql: "SELECT * FROM books" }
% arara: datatooltk: {
% arara: --> output: customers.dbtex,
% arara: --> sqluser: sampleuser,
% arara: --> sqldb: samples,
% arara: --> sql: "SELECT * FROM people" }
% arara: datatooltk: {
% arara: --> output: countries.dbtex,
% arara: --> sqluser: sampleuser,
% arara: --> sqldb: samples,
% arara: --> sql: "SELECT * FROM countries"}

@NicolaTalbot woooooo

@PauloCereda You'll have to make sure arara v4.0 is ready before my new LaTeX book :-)

@NicolaTalbot I believe the code is ready (apart from the translations). :) But I need to write the manual, and that's the tricky part. :)

@PauloCereda I find writing the manual always seems to be more time-consuming than writing the code!

@NicolaTalbot 100% agreed. :)

@PauloCereda Howso is it tricky?

@SeanAllred The manual writing, you mean? :)

@PauloCereda Yes :)

@PauloCereda Also, I just took a look at the new features business. Example 2 (pdflatex until !found...) is genius syntax.

@SeanAllred Well, I usually find a hard time when writing technical text in English. :) And sometimes I don't know how to manage to put on paper the implemented features. :)

@SeanAllred Ah that one was good. :)

@PauloCereda Sometimes I forget you aren't a native speaker :)

And yeah, even native speakers have a hard time writing technical text from the looks of it. I swear some of the stuff I'm maintaining was written by a twelve-year-old.

@SeanAllred <3

@SeanAllred Oh my!

@SeanAllred It's hard to tell what's obvious and what isn't obvious to a user who isn't familiar with the code. Sometimes I find I've overrun the time I allocated to write the code so I don't have much time for writing the manual. I either have to delay the release until I have more time or hurry the manual.

@PauloCereda I'm always very impressed by anyone who can write technical stuff in a language other than their native one.

@NicolaTalbot Scheduling is a bane in any context, I suppose. And yes, understanding what's obvious and what isn't is a hard skill to learn -- one that might even be impossible to do 100%. Being in a teaching position definitely helps with that, though.

@NicolaTalbot Technical stuff pushes our limits. :)

@PauloCereda I think if I tried translating any of my manuals, they'd just say: pato, chapéu, pato, chapéu...

@Nicola: did you like the new justified layout? :) It's surely more pleasant to the eye. :) Even the verbose mode is more organized (or less cluttered). :)

@NicolaTalbot LOL

@PauloCereda Yes, it looks nice and tidy :-)

(DatatoolTk) The DatatoolTk software .................... SUCCESS
(DatatoolTk) The DatatoolTk software .................... SUCCESS
(DatatoolTk) The DatatoolTk software .................... SUCCESS
(PDFLaTeX) PDFLaTeX engine .............................. SUCCESS
(PDFLaTeX) PDFLaTeX engine .............................. SUCCESS

@NicolaTalbot Yay!

@NicolaTalbot: ^^ This is in a paper of mine. I'm serious. :)

@PauloCereda I needed that in my life.

@PauloCereda :-) tikz?

@SeanAllred <3

@NicolaTalbot Of course! :)

@Nicola: I think I could poke our translators in the next days. :) I simply love the Russian version we had in 3.0. :)

@PauloCereda Structured stack automation which recognises sentences containing duck, duck, goose?

@NicolaTalbot Yep! :) The duck, duck, goose game. :)

Then I propose a syntactic injection in which I can extend the language. :)

@PauloCereda Cool! :-)

@NicolaTalbot Ducks and hats, what else could one want? :)

@PauloCereda Erm, araras? :-)

Hmm, I've just uploaded a new package to CTAN, but they don't know what topic to assign to it.

@NicolaTalbot Oh my! I forgot to tell you! I went to Campo Grande, in Mato Grosso do Sul. The city is crowed with araras flying everywhere!

@PauloCereda Oh wow. Did you take photos?

@NicolaTalbot Sadly no. :( I had no camera and my cellphone had a draining battery. :(

Good evening. Is there a command to disallow displaybreaks?

@RafaelWörner \allowdisplaybreaks[0]

@tohecz allright thanks. So this is locally at this spot or for the rest of the document?

@PauloCereda duck -> lots of food -> foie gras

@RafaelWörner for the rest of the document I suppose

thanks

@tohecz ducks -> lots of food -> fat ducks. :)

@RafaelWörner well, if you enclose it in a group, it will be local

yea, right :)

@PauloCereda well, you do know foie gras, right?

@tohecz I do. :)

@PauloCereda btw, I should get some soon, and have an expensive dinner :)

@PauloCereda Oh, that's a shame.

@tohecz Mon Dieu. :)

@NicolaTalbot Indeed. :(

@PauloCereda well, you can get it as cheap as €70/kg, so a small package is then within reasonable price :)

(Most email signatures these days: USE ALL THE FONTS)

Evening, everyone.

@SeanAllred Just an exercise before going to dinner. ;-)

@henry Hi!

@egreg Oh, you like to show off :) <3

@SeanAllred :)

@egreg I honestly would consider turning that into an expl3 support package (a la lt3graph)

@NicolaTalbot Once again thanks for your quick assistance. I don't think I appropriately expressed how much I like this package so far, but I can tell you that unlimited access to the page layout is the most fun I've had since I started using LaTeX. Since I can't give you additional reputation for the most recent question, I'll rep you through the old one. Thanks again. :-)

If a such old question exists. I swear you answered another question of mine earlier.

Can't seem to find it. I'll try to bounty when the question cooldown time expires.

@PauloCereda Without looking at the code, I'm a little concerned with arara/security

> exists ... checks if the file exists. Note that this is a proper call to the File object in the Java API, so it should be used as exists(new File('foo.pdf'))

Do you ensure that you can't pass arbitrary code into the exists function? A la shellshock?

@JosephWright The documentation about \ior_map_inline:Nn is wrong. It says “TeX also ignores any trailing new-line marker from the file it reads”, but it isn't true: an empty line is converted to \par, so the end-line is there. Maybe a trailing space is stripped off when passing the line as argument, but \par is not nice to have for an empty line in several cases.

@egreg How am I supposed to finish xor with all this low-level stuff to fix ;-)

@JosephWright Leave xor to @DavidCarlisle.

@egreg Oh I wish

@JosephWright It could be \ior_map_inline:Nnn, where the second argument is an integer to which \endlinechar is set.

@egreg Tricky: at the moment we've not mentioned \endlinechar at all for the 'user'

@egreg The plan to date I think has been not to expose \endlinechar at all: we'd rather have appropriate internal setting where it is needed and otherwise leave 'standard'.

@JosephWright Then just a choice between “normal \endlinechar" and “no \endlinechar” (with the bug already discovered in the middle).

@egreg Yes, I guess so

@egreg I'll raise with the team

@SeanAllred Sadly there's nothing much I can do. The only possible way to (poorly) overcome a situation like that is to have, at some extent, a fine grained control on the abstract syntax tree, and I'm not inclined to do that. :( Code injection is possible at so many levels (notably reflection) that most of our efforts worth nothing. You are right on the spot that arbitrary things can happen, but I think now it's up to the user to trust its source. :(

At least we share concerns with other tools as well. I'm pretty sure we can inject $pdflatex='rm -rf /' in latexmk at some point. :)

@Werner nice one :)

@tohecz seems like @Werner needs some downvotes to get back to real palindromes

cube is a nice thing, too ;) and it is a $\Theta$-palindrome, with $\Theta=(1\; 2)$

@DavidCarlisle He turned red for the emotion!

@egreg you squares all look alike to me

@DavidCarlisle True...

@PauloCereda :/ I suppose you're right on both counts. Actually creating an AST is a lot of work for something that's got to be at the head of the file anyway :P

Anyway, am I just missing something or does xparse not support optional verbatim arguments?

Hello chat. Another question. I'm in doubt about what exactly is \endlinechar. \escapechar is the character that is printed when using \string\foo before the token name (usally a backslash, in that case the token name is foo so in the end one gets \foo printed). But I don't know where does \endlinechar fall. Isn't the end line char already ^^M, and, by setting it to the correct catcode everything should work correctly. Where does \endlinechar come into account?

@SeanAllred We are already doomed, I guess. :)

@Manuel \endlinechar is the code of the character that TeX adds at the end of every line it inputs. It's usually 13 (ASCII CR), which has category code 5. That's what makes end-of-lines behave like spaces or \par.

@Manuel When TeX reads a line, it removes the OS end-of-record signal and all that follows it; then it removes all the trailing spaces and tabs and adds the character with code \endlinechar (if ≥0).

@egreg But, ^^M is the carriage return which is the end of line. So it is in the file. So I understand that TeX sees that ^^M so why would TeX add \endlinechar. That's where I'm lost.

@Manuel The reason for all of this is how computers behaved in the 70's. Some of them had fixed length records (for instance IBM mainframes) and records were padded by spaces or NULs. Other OSs had variable length records and the record terminator could be CR, LF, CR-LF or LF-CR.

So Knuth decided to “normalize”.

Still an OS that I won't mention uses CR-LF.

In their wisdom, Apple guys decided to use LF for the Lisa (maybe also for the Apple II) and then for the Macintosh.

@egreg \endlinechar issue raised with the team

@egreg So the normalization is done by “gobbling” any of those CR, LF, CR-LF, LF-CR, and instert a general \endlinechar. Is that more or less the key? I don't know about ascii, 70's, or Lisa, so this is chinese for me.

@egreg I've always thought Windows is more logical here that Unix

@Manuel Yes, that's the idea.

@JosephWright Sorry, that's impossible. ;-)

@egreg :-)

Still too broad?

@egreg In that case, I'm with Joseph, this shouldn't be known by the user. I don't think that's relevant for anyone :P Just creatting appropiate macros and forgetting about \endlinechar for ever and ever.

@JosephWright You gotta be kidding me. :)

@Manuel It is needed for some effects, the question is can we wrap it up properly

@PauloCereda No, I'm serious. A new line just advances the scroll, it doesn't send you back to column 1. So CRLF/LFCR seems to be 'required': just LF should really disappear off the screen.

@Manuel The problem is when reading files with \read: the \endlinechar is added also there, which in some cases is a nuisance.

@JosephWright Yeah, I know (that's why I have seen it many times). But I hope you find a way of hiding it.

@tohecz could you write an answer on showing how to use that \the? The OP will have to do it for all his penalties by him self.

@JosephWright Also VMS used CRLF, IIRC; the model of course were the teletypes, which needed both characters.

@josephWright yes, it is to broad, still

But for denoting end-of-record one character suffices.

But showing him how to read out the default or current definition would give a pleasant answer

@JosephWright @tohecz or maybe just point to a dupe on the topic dimensions. (which would leave it closed nevertheless)

@LaRiFaRi but well, that's an answer to a completely different question. Moreover, this one is closed (and I won't vote to reopen it)

@JosephWright Yes.

@LaRiFaRi Look for “The mighty \the”

@egreg I was sure that this has to exist

@JosephWright Isee. :) It's a compromise with visualization as well, back then.

@LaRiFaRi: arara 4.0 is near its release. :)

@PauloCereda after reading your summary of new features, I'll definitely be testing out the new release. :-)

The conditionals are very appealing to me. Thanks for all your work on this!