@daleif Yes, but RStudio assume MacTeX, so if MacTeX is set up correctly, it should work.

hello. How to change thickness of line produced by \line(0,1){\LenToUnit{\textheight}}?

I am using eso-pic if related

@Sigur Example?

@AlanMunn, here:

I want o produce a vertical line between the body area and the marginpar area

@Sigur I don't know much about eso-pic. Could you use the backgrounds package instead?

@AlanMunn, well, I just need a vertical thick rule.

@Sigur Something like this:

@AlanMunn, yes. But that one uses tikz. I think it is too much. I'd like a very basic package.

My code produces the line. Now, I'd like to change its thickness only.

@Sigur I see. Can you make your code a complete MWE?

@AlanMunn, please, see this one. Not tested.

@Sigur This produces no line for me.

@AlanMunn, let me see

@AlanMunn, strange. So, I don't know why it works in my huge file.

Oh, it needs twoside option, of course.

\documentclass[twoside]{report}

But even with that, no line. Strange, very strange

\documentclass[twoside]{article}
\usepackage{lipsum}
\RequirePackage{eso-pic,eepic}
\allinethickness{2.7pt}
\makeatletter
\AddToShipoutPictureFG{
\begingroup
\AtTextLowerLeft{
\if@twoside
\ifodd\c@page
\put(\LenToUnit{\textwidth+.5\marginparsep},0){%
\line(0,1){\LenToUnit{\textheight}}}%
\else
\put(\LenToUnit{-.5\marginparsep},0){%
\line(0,1){\LenToUnit{\textheight}}}%
\fi
\fi
}
\endgroup
}
\makeatother
\begin{document}
\lipsum
\end{document}

@AlanMunn, you got it, I missed \makeatletter

@Sigur This gives you the thick line you need too.

yes.

I worry if \allinethickness will affect other lines also

its name is very powerfull

@AlanMunn, anyway, thanks.

@Sigur Better solution: load the pict2e package, and use \linethicknes{2.7pt} inside the group.

@AlanMunn Believe it or not, I was about to suggest something similar. (Of course, in a real document one will use forest, which loads TikZ, so one could draw the line with TikZ right away. ;-)

@marmot Yes, well some people apparently like the simpler things in life. :)

@Sigur wait for @DavidCarlisle. He knows everything about picture mode. But beside this you can also use \rule to print a line of a specific width.

@UlrikeFischer you can use \linethickness without any package for horizontal or vertical lines of any thickness, you just need pict2e for slopes.

Do y'all reckon there is a way to disable tabcolsep for one side of one column in a table only?

@Magisch {lll@{}}

I've seen a similar construct before but I never understood what it does

I have a table defined as \begin {tabularx}{\textwidth}{t{0.21\textwidth}t{0.39\textwidth}t{0.15625\textwidth}t{0‌​.1875\textwidth}}% so assuming your solution it would then be \begin {tabularx}{\textwidth}{lll@{}t{0.21\textwidth}t{0.39\textwidth}t{0.15625\textwid‌​th}t{0.1875\textwidth}}%, correct?

@Magisch you appear to have tabularx with no X column so that would be wrong

if you put @{hello} between two columns it will write hello instead of having the \tabcolsep spacing in each column (so @{} removes 2\tabcolsep` space except in the first and last position where it removes \tabcolsep

ah

so @{} in the positon where I don't want the tabcolsep

I could in theory then do @{\rule{\tabcolsep}{0pt}} to change the 2\tabcolsep between 2 columns into 1?

@Magisch yes but also don't use tabularx if you are not using X

@Magisch \hspace{\tabcolsep} would be simpler and clearer, but yes.

I don't remember why, but there was some issue with normal tabular

@Magisch there may be an issue with tabular but using tabularx is simply wrong, so use tabular:-)

Interestingly, ` \begin {tabularx}{\textwidth}{@{}t{0.21\textwidth}t{0.39\textwidth}t{0.15625\textwid‌​t‌​h}t{0.1875\textwidth}}%` throws an error while compiling

While ` \begin {tabularx}{\textwidth}{lll@{}t{0.21\textwidth}t{0.39\textwidth}t{0.15625\textwid‌​‌​th}X}%` does not, but causes \multicolumn{3}{X} to no longer span the last 3 columns.

@Magisch you won't get an error message if you do not use an X column but any output you get from such a table is just accidental nonsense, the code will do nothing sensible. If you are not using a variable width X column do not use tabularx.

@Magisch anyway what is your definition of t ? there is no t column by default

\newcolumntype{t}[1]{>{\raggedright\arraybackslash}p{#1}} %

I modified my tabularx to be \begin {tabularx}{\textwidth}{@{}t{0.21\textwidth}t{0.39\textwidth}t{0.15625\textwidth}‌​X}% and at some point in the code I define a multicolumn as \multicolumn{3}{X}{Some Text}, prior to adding the X column at the end of the tabularX, this worked fine. But now the multicolumn behaves as if it was a single column

@Magisch I would never dare something like \multicolumn{3}{X}. What do you expect it to do?

generating a column like ´multicolumn{3}{>{\raggedright\arraybackslash}p{whatever is left after the non multicolumned columns}}`

I just did that math by hand and now it works

Is now \multicolumn{3}{t{\textwidth-0.21\textwidth-2\tabcolsep}}

@DavidCarlisle bidi does contain code to encrypt the pdf and change the security settings and probably there was some error in an older bidi version.

@UlrikeFischer oh great

@DavidCarlisle well as it seems to be gone after an update ...

@UlrikeFischer oh I thought you meant it had just started encrypting. (I'm not really doing tex at the moment, just half watching the conversation:-)

@DavidCarlisle no it is gone now. Can one trace the amount of glue inserted during the typesetting? I need to find out where some 12pt glue comes from.

@UlrikeFischer vertical? \showoutput shows the kind of glue added baselineskip or \lineskip or something, otherwise \tracingall and search for glue...

@DavidCarlisle I meant, which of all the tracingcommands in tracingall is showing the glues?

@UlrikeFischer \showoutput shows the resulting glue but if you want to see where it is added you need \tracingmacros and look for something inserting \hskip` or \tracincommands and look for glue.

@UlrikeFischer or just ask @egreg, finding white space being his main skill.

@DavidCarlisle found it. beamer changed partopsep ...

@UlrikeFischer well it changes everything else as well:-)

@DavidCarlisle yes, but here it did it between 2017 and 2018.

Hi. This is ugly, isn't it?

Do you have any idea to improve?

@JouleV use a wider column...

@JouleV Use more words? "it is at its maximum between the faces labelled with P in the diagram"

Say I have somethng like this $\sum_{A_{B^C}}$ B and C are both the same size, which looks bad, but there is not scriptscriptscriptstyle available. I'm playing with a macro that formats A_{B^C} and is wondering what is the best way to detect when you are already in scriptscriptstyle at the B level, and thus need to apply something extra to C. I assume we need to use several mathchoices?

@daleif or change the notation? Or if you must have that, I assume it's in display mode not text mode, so you could set the subscript on the \sum in \textstyle so B would be in script and C scriptscript

@daleif You can see a difference in \[\sum_{A_{B^C}} \sum_{\text{$A_{B^C}$}}\] provided the main size is 12pt. For 10pt you need to modify the default \DeclareMathSizes{7}{7}{5}{5} to use, say, 4 for \scriptscriptsize.

@barbarabeeton Thanks again for your help with the TUG19 ad! It seems to be doing rather well: tex.meta.stackexchange.com/ads/display/8048 :)

@DavidCarlisle that is not my call, I'm just editing this for publication. It is even worse since C is often (n)

I'm thinking about using scalerel to scale C in the situation where B is already in ss, but I need to be able to detect this.

xstring is rather nice in deconstructing their input. It often has the form of D_{f^{(n)}}, the macro I'm rewriting it to uses \Macro{D}{f^{(n)}} as the input which I split up and clean further using xstring. (I could manually clean it, but I'd rather leave the ()'s in the input for clarity

@daleif I guess luatex is not an option? It has \mathstyle command to determine the current mathstyle.

@daleif simplify the notation? :) Oh it's not your work. Then: suggest simplifying the notation?

@UlrikeFischer I generally stay away from lualatex and the solution used should work on the systems the author is using and hardly any of our researcher use lualatex or even know it exists.

@yo' I'm going to mention it in may comments. At least write it as a macro with arguments instead of hardcoding D_{f^{(...)}} every time.

@samcarter -- well, it wasn't very hard to do. thanks for the idea. (any chance you might attend the conference?) i see that detexify is at the top now, as well it deserves.

@daleif true but that's an implementation detail, you can use it as if you can choose (hence the name:-) the fact that it evaluates all four branches and then picks one rather than just evaluating the appropriate branch shouldn't matter too much.

@barbarabeeton Maybe I'll be able to come, but I don't know for sure yet. There is another conference in the US the week before and if I attend this one, I could just stay a bit longer for TUG19.

@DavidCarlisle One cannot use it to say, set a boolean, such that in the D case of A_{B_{C^D}} one could act according to that boolean. If we execute something at C can we have that detect the current style and then use that information in the typesetting of D.

@daleif no everything that depends on the choice has to be in the argument so \mathchoice{\zz}{\zz}{\zz}{\zz} where \zz is A_{B_{C^D}} then apply styling in each case as required

@DavidCarlisle I hadn't realized that there are already so many tabu issues ...

@UlrikeFischer it's just a matter if chasing down the colour groups (but also not adding them if old array package has been loaded) but I have had no time

@UlrikeFischer التوقف عن التسبب في المشاكل

@DavidCarlisle hey, I did look at the *** tabu .... Btw: test t400 misses the brace at \end{documen but I couldn't push a change.

@UlrikeFischer :-)

@UlrikeFischer thanks: b271a95..e38db3d master -> master

@UlrikeFischer you have push access now

@DavidCarlisle I'm not sure if I should say thanks here ;-)

@UlrikeFischer what's the point in having all those organisation members if they have no write access and can't be blamed?

@DavidCarlisle ^^^^

hmmmm

Hello!

I would like to know if any of you has knowledge of the TARINGA! page created about a decade ago by Argentines. I just saw a video on YouTube and it is said that the page has died due to political inclinations, elimination of downloads of all kinds (music, video games, etc.), among others, which is true by empirical verification. And if you believe that our community may "disappear" (not be as active as it has been until now)

@daleif Did you get your Rmd issue sorted?

@manooooh As with many things, the things that will make this community disappear will be either (i) people completely stop using LaTeX (or at least stop to the point of their being too few users to support the community). (ii) something better than TeX.se shows up.

@manooooh When the site design was changed recently, there were some people who thought that we should start our own version of the site (since all the material is available) but it would take a lot of users to do it, and most thought it wouldn't be productive.

@manooooh When Microsoft bought GitHub, many of us immediately set up GitLab accounts just in case they did something tragic with it. And some projects did make the move, but most of us stayed when it became clear that there weren't going to be any major changes to GitHub as a result of the acquisition.

@AlanMunn the lecturer will not interact with the students until Thursday, then we'll see, thanks for your help

@daleif Ok. Feel free to ping me again if you have other questions.

@AlanMunn thank you for your opinion!! :D

Anyone else?

@manooooh the community isn't tex.se that's just a currently useful medium, it used to be on texhax mailing list then comp.text.tex then usenet news got completely messed up by google, and it sort of landed here, but if stackexchange becomes evil it could move elsewhere or if word starts typesetting documents better than tex, perhaps the focus would move around there. Who knows.

@DavidCarlisle thank you David!

When one googles something about (La)TeX the first results are sent directly to the TeX.SE site. I hope it is always like this

@manooooh it wasn't always like that, not least because latex is older than google

@DavidCarlisle that a site is newer or older than another has nothing to do with popularity. Good ideas last in time, but then they go out like a candle after being lit for several hours

@manooooh I was referring to your comment that you hoped that people would always be able to google and land here, just stating the (obvious) fact that that has not always been the case.

@DavidCarlisle oh yes, of course. Sometimes for a quick search Overleaf or ShareLaTeX tutorials are more efficient than Q&A

@manooooh when I started with tex there was no internet and people used ink pressed on to dead trees.

@DavidCarlisle LOL

@manooooh the UK didn't adopt internet until 1991 so tex had been in widespread use for almost 10 years by then

@manooooh Sometimes, sites are closed (MonkeyTeX, SpanDeX.io, or even Google+ these days), sold (Github) or taken over (ShareLaTeX) or they change significantly (even WhatsApp gets advertisement now). Noncommercial sites are not so endangered since they don't have to produce money for owners or investors.

@manooooh If such things happen, the community can just move to another place that's nice.

@StefanKottwitz do you mean that there are different types of sites? Any non-public domain (not government pages, for example) can be sold to another company. Are you saying it will not happen with StackExchange?

@StefanKottwitz perhaps, or another way to look at it is that sites with commercial backing are less endangered as they don't rely on so many donations of time and money to keep the back end infrastructure running.

@manooooh no he is saying it could happen to stackexchange

TARINGA! it is free access and has nevertheless gone through some stages of sale. Anyway, a free and crowded site began to fall to the point of very low levels of audience. I think the possibility that TeX.SE (or any other community) can pass the same thing, for whatever reason. What I do not want is for it to happen in a few years, but decades

@DavidCarlisle oh

Thank you David and @StefanKottwitz!

Can you give your opinion about TARINGA!, please? (Of course if you know what has happened). Do you think that people who have stopped participating have moved to other places? I believe that a small part has moved to YouTube, and little else (in addition to the communities that are created with respect to video games, mainly). I think nobody has created a new site

@manooooh I have never heard of it

And even the great majority has not returned to participate of any other site; not at least as actively as many users did

@manooooh I too, have never heard of it. Is it a kind of Argentinian Reddit?

@DavidCarlisle do you want to take an overview? taringa.net Of course it is not mandatory (even I only participated one or two days, literally). What do you think of the site?

@AlanMunn yes!

@manooooh no:-)

Ohh, thanks to @AlanMunn he remember me about Reddit. I think I was wrong; a lot of the people that who stopped participating moved to Reddit

@DavidCarlisle LOL, just an overview. I can not guarantee that there will be ducklings or marmots (or any animal) :(

@manooooh Quick sample page: taringa.net/+juegos_online/… how many advertisements can you count?

@StefanKottwitz three

@manooooh looks more like 300 to me

@manooooh I count 34 at first sight

@DavidCarlisle @StefanKottwitz you are right, sorry, AdBlock activated

It was not like this in the past. That's why I say, before it was much better; now it became garbage

@manooooh I would not participate in a site that shows such "content" on just a normal post page:

People think something different and already have 10 people behind who insult him. That's how we Argentines are ...

@manooooh what's the point of a site that is almost all adverts if you can only use it by activating software that hides most of the site? But don't ask me I don't use any social network sites (unless you count this one)

@manooooh Social media kind of sites are much more sensitive to the whims and trends of users. While some site like FB have an extremely entrenched user base, others are more ephemeral. How many people left Instagram for Snapchat, for example? The SE sites serve a much more useful purpose; they are only secondarily social.

@DavidCarlisle the ads are annoying for most users. One enters to see good content, good opinions, and no junk content

@StefanKottwitz yes, I know... That is one of the reasons why it has stopped being as popular as it was a few years ago

@manooooh that way, harvesting money by massive advertising, can kill a site

@manooooh I can't see me using such a site (even if it was in English) If it is just gossip and chat what's the point? This site is about answering questions on a specific subject, even if there is a bit of chat around the edges

@AlanMunn until a few years ago the page worked more like a Megaupload site than as a social network. The design has changed a lot, some users have evolved badly, and opinions are basically unfounded

@StefanKottwitz SE is a private company that does not have ads (T! is private but with ads and still died). How is SE maintained?

@manooooh it has lots of ads, look at stackoverflow

@manooooh A few links: tex.tips/some-bookmarks

@DavidCarlisle there are posts that talk about economy, politics, videogames, entire communities. If you do not like none of these is clear that the site is not for you. You are lover of LaTeX :)

@DavidCarlisle I have never noticed

@manooooh that is also why the site design changed last year to provide bigger left and right sidebars so they could put more adverts and links to money earning parts of the site network

@manooooh I guess TeX.SE lives because it's part of a big network where other bigger sites are actually generating money for servers, development, employees, marketing

@manooooh There is also the minor problem that although @DavidCarlisle knows how to use Google Translate is multilingual, Spanish has never been his strong suit.

Cue Spanish sentence in 3..2..1...

@AlanMunn Eu posso entender um pouco de espanhol porque meu português é excelente

@AlanMunn not spanish, but yes

@DavidCarlisle :)

@AlanMunn playing with people's defects is not pleasant, at least for me. I apologize if my English is not acceptable enough to understand me. In that case, I would like you to tell me which part you did not understand

@manooooh Alan just meant that I could not comment on that site as I can't read it (which is true:-)

@manooooh No, no, I'm explaining jokingly about another reason why @DavidCarlisle wouldn't be interested in Taringa. Nothing to do with you at all.

@DavidCarlisle yes, that is very probably. I have disabled AdBlock but there is no ad in tex.stackexchange.com ???

@manooooh there are "community adds", as currently being chosen now tex.meta.stackexchange.com/questions/8048/… they fill the space on this site that is taken by actual adverts on the money earning sites in the network.

@StefanKottwitz yes. I do not want to be a sucker, but all I see is a Q&A site where people are respected and the community is very respectful. That, for me, is what makes SE great

@DavidCarlisle man... I love the community adds!! I have never been in a site such that the community makes its own ads! That is amazing! And there is one ad, not 59 ¬¬ hahaha, I love it!

I did not even consider that ad

@AlanMunn oh, my apologies

@DavidCarlisle the only thing you should see are the memes!

@manooooh as a general rule, if you are not sure what someone means when chatting here just assume they are being rude about me. Chances are you will be correct.

@DavidCarlisle how can I think that with a person who has studied the core of (La)TeX more years than I have? :P

@marmot Happy Groundhog Day! ^^^^

@UlrikeFischer One hour to go till it is groundhog day on the main site.

@marmot that's a french marmot, it doesn't care about Greenwich time ;-)

@UlrikeFischer Yes, but groundhogs are not marmots. (Have you been at Au Pays des Marmottes? ;-))