TeX, LaTeX and Friends

8:34 AM

@vlg finally you start to show some code..,, I assume (I have to assume as you have given no hints) that there is no \par or \newline but that as I showed you have a bad group or space. The {..} you show in that fragment looks suspicious for example as \color does not take a text argument (but you may mean the ... is the color name such as red, hard to tell). It must be trivial for you to make an example, just take the document you have and delete everything except that...

@vlg ... one cell and then delete every package not used in that table cell and end up with a small document just consisting of a one row table.

8:46 AM

Weekend, woo!

@PauloCereda I was born for it (indeed, I was born on a Saturday, at 1:00 a.m.)!

@CarLaTeX ooh

My proposal is becoming popular: meta.stackexchange.com/questions/324372/… :)

JouleV

9:25 AM

@CarLaTeX Have you ever think of cleaning answered questions from your favorite list? I see all questions here, for instance, are now answered, and mostly solved, therefore you can unfavorite it!

@JouleV Yes, I clean the list when the answers which don't interest me are answered, but there are also my actual favorites, which I don't want to clean

9:46 AM

@PauloCereda weekend, hmm looking forward to a sunday roast dinner ???

@DavidCarlisle oh no

@DavidCarlisle you are mean

Jun 29 '17 at 16:15, by Paulo Cereda

@DavidCarlisle you are not mean :)

Drat, SE broke Psmith again...

@DavidCarlisle oh

10:22 AM

@MarcelKrüger if you have time to look at the new bug (I don't have it ...) it would be probably best to base it on the branch I made yesterday night, it contains your last fix and the newest fontloader code.

vlg

1:06 PM

@DavidCarlisle \color{ uses a predefined RGB val, however, it was the extra space between #8 and the color that did it. How wrong I was to ever believe I can leave in an unaccounted for space.

@vlg and next time show the code not an image of the output.

3:34 PM

@subhamsoni The center of the arc (i.e. the center of the circle/ellipse the arc is a seqment of) is determined by the start angle and radius you specify. If (X) is your start point, the center is at (X) - (start angle:radius) where (start angle:radius) is in polar coordinates.

@egreg -- Ah, yes. That makes sense now. A lot changed going from 2.09 to 2e, and it wasn't just AMS-LaTeX.

is dot below used for anything other than arabic transcriptions, and how many of these are commonly used?

1E04;LATIN CAPITAL LETTER B WITH DOT BELOW
1E05;LATIN SMALL LETTER B WITH DOT BELOW
1E0C;LATIN CAPITAL LETTER D WITH DOT BELOW
1E0D;LATIN SMALL LETTER D WITH DOT BELOW
1E24;LATIN CAPITAL LETTER H WITH DOT BELOW
1E25;LATIN SMALL LETTER H WITH DOT BELOW
1E32;LATIN CAPITAL LETTER K WITH DOT BELOW
1E33;LATIN SMALL LETTER K WITH DOT BELOW
1E36;LATIN CAPITAL LETTER L WITH DOT BELOW
1E37;LATIN SMALL LETTER L WITH DOT BELOW
1E38;LATIN CAPITAL LETTER L WITH DOT BELOW
1E39;LATIN SMALL LETTER L WITH DOT BELOW

@barbarabeeton @JosephWright (anyone:-) I'm wondering if we should predefine them...

@DavidCarlisle -- Roast carrots can be very tasty!

for pdflatex the same would work but you need to add \DeclareUnicodeCharacter{1E25}{\d{h}} in current releases. — David Carlisle 57 mins ago

@barbarabeeton go well with orange sauce?

@DavidCarlisle -- Actually, depending on the herbs used, they may go very well with orange sauce. (And the colors are complementary.)

3:45 PM

@barbarabeeton never trust Americans with the English language first they lose the u and now c goes as well....

grr my internal clock not accurate enough I thought you'd be out of edit time:-)

@DavidCarlisle that would never happen to English people, they prefer to go safe and write commit messages like limiitiiationss, ;-)

@DavidCarlisle -- There are some letters in transcriptions of Sanskrit that have underdots. (Sorry, I'm not fluent in Sanskrit ... but in any case, it isn't Arabic.)

@UlrikeFischer I can't imagine what you mean:-)

@DavidCarlisle -- I never claimed to be a flawless typist. And the keyboard I've got available isn't helping. Gotta get that situation improved ...

@barbarabeeton I could try claiming that but I doubt anyone would believe me.

@barbarabeeton if you are buying a new keyboard, I recommend one with a shift key.

3:53 PM

@DavidCarlisle do you want a screenshot?

@UlrikeFischer wait till you are at a tug meeting and have a bigger screen. Who would be so mean....

@DavidCarlisle -- You may have noticed that, as of February 9, a shift key magically appeared.

@DavidCarlisle There's info about usage at en.wikipedia.org/wiki/Dot_(diacritic)

@barbarabeeton probably faulty after years of neglect. Anyway hope you are starting to engoy your new freedom.

@egreg thanks, can we afford 44 extra declareunicodecharacter....?

@DavidCarlisle That's just a one-time job.

4:00 PM

@barbarabeeton when people use em-dashes like---this is an example---this, do they expect that there are possible line break points?

@egreg oh it's no effort to add them, just should we, with it being pre-loaded these days I suppose there is no run time cost at all really

A question about an edit that recently appeared on the tex.meta site. A title and corresponding text in the question Welcome to TeX.SE was changed trivially from TeX.SX to TeX.SE by a relative newbie (with association rep at 101). I've been using the notation "tex.sx" since I joined this site. What is the practice of other old-timers? I've noticed both forms, though I personally still prefer "tex.sx".

@barbarabeeton I always use TeX.SX

@barbarabeeton Reminds “The Joy of TeX”. ;-)

@barbarabeeton .se is the normal network abbreviation but there were reasonable objections from Sweden (I an sure there is an existing meta.tex.sx question about that) and the recommendation here was to use sx

@DavidCarlisle Should we cover all diacritics for Latin letters?

A: Which is more common: "TeX.SE" or "TeX.SX"?

4:03 PM

20

The 'official' position across the network is to use 'SE', so that is what I use when talking to people on other sites. However, 'SE' is the TLD for Sweden, and 'SX' (or 'sx') fits nicely with 'TeX'. So on the site I and many others use a variation of 'TeX-sx' (TeX-SX, TeX.SX, etc.).

Joseph Wright

@barbarabeeton I've pinged the editor to ask them

@UlrikeFischer -- In the U.S., line breaks are assumed to be possible after em-dashes. However, there are situations where em-dashes are used in place of parentheses, and then a line break after the first doesn't make sense. A careful editor (human) would have to make a judgment call in that case. This is something I should look up before pontificating.

@egreg I tried (with the extension in 2015 or so) to cover all the single-accent diacritics that are"easily" covered by T1 encoding and predefined accent constructs, but obviously missed this whole block of dot under

@DavidCarlisle I just had to implement a work around for someone as the used (unicode) font neither had a \d{h} glyph nor the combining accent, but I have no idea how they used it (but I doubt that it was arabic).

@UlrikeFischer It's a problem with several OT fonts.

4:05 PM

@DavidCarlisle -- Then I think I will roll back the update on that basis. (But maybe I should wait for the editors to respond to @JosephWright?)

@egreg I know (and the work-around is from you ;-))

@UlrikeFischer :-) Now with TU it's a bit easier, IIRC

@egreg ? I used a definition based on \o@lign. Is there something better now?

@UlrikeFischer The \Undeclare... part

@egreg Ah.

4:12 PM

@DavidCarlisle SX is the country code for the Dutch part of Sint Maarten/Saint Martin island (the French part has code MX).

@barbarabeeton I have the problem, that the fontloader code of luatex "eats" an hyphen and so gives an en-dash instead of a em-dash dashes---eaten. The suggested fix to set \automatichyphenmode=1 works, but has the side effect to suppress line breaks after the em-dash (but also after the hyphen if it is at the begin of a word). I'm wondering if we will get lots of complains if I change this.

@UlrikeFischer -- Eek! Seems to me there's something wrong there. Does it also eat the first hyphen of an en-dash, giving just a hyphen? I've certainly seen em-dashes at the ends of lines where a line break would be both appropriate and beneficial. I suspect that making the suggested fix would serve to confuse people; I don't like putting forced breaks into source text, and the only alternative I can think of to avoid that is to insert \hspace{0pt} after an unruly em-dash.

4:29 PM

@barbarabeeton no it eats only from em-dashes (and only if input as ---, Hans comment about it is, "it's about time that texies start using the proper unicode symbols instead of these funny ligatures". I must say, from a german view \automatichyphenmode=1 is the better default: we don't have em-dashes without spaces, but often words starting with an hyphen.

@UlrikeFischer -- Then it sounds like the \automatichyphenmode=1 is the better approach. Do you know if that would prevent a line break after the hyphen in expressions like $p$-adic? That does sometimes get an unwanted line break unless something is done to prevent it.

@UlrikeFischer I could understand not doing the ligature at all but what is the logic for just losing one of the three?

@UlrikeFischer oh I see, the first one is treated as \- ?

@DavidCarlisle well the first is a discretionary and as Hans writes "because <disc -> followed by -- is pretty obscure".

@UlrikeFischer he could have said "it would have been obscure except for 30 or 40 years of legacy use of --- " ...

@UlrikeFischer bizarre values for that parameter 0= true and 1 = false?

@DavidCarlisle No the possible values are 0, 1 and 2. 0 breaks a lot, 2 doesn't, 1 is in the middle.

@barbarabeeton yes, it would prevent a break:

4:46 PM

@UlrikeFischer still would seem more natural ordered the other way, also mixing break points choice with turning - into a discretionary seems strange, but it's only a name....

@DavidCarlisle pints reminds me of cider ;-)

@UlrikeFischer @barbarabeeton is not the only one who can change history

5:02 PM

@DavidCarlisle ;-) I could have waited ...

Harald Hanche-Olsen

@DavidCarlisle I think I like the notion of a break pint.

5:32 PM

@HaraldHanche-Olsen Check your vowels and bring a pint also to me!

6:09 PM

My cousin's daughter boyfriend has just converted her thesis from Word to LaTeX. She didn't want to write in LaTeX directly, but now she is enthusiastic of the result: a new LaTeX fan in the world!

5

6:28 PM

@CarLaTeX I guess it is because of the tikzducks package, right? ;-)

Dr. Manuel Kuehner

6:41 PM

@samcarter Online?

@samcarter Just wanted to ask why your account says "This account is temporarily suspended for rule violations." :).

@Dr.ManuelKuehner as has been mentioned here before, best not to discuss such things in open chat.

Dr. Manuel Kuehner

@DavidCarlisle Ok, I see. Thanks for replying.

@DavidCarlisle I did not find an email address in his/her manuals.

yo'

@Dr.ManuelKuehner Her, quite certainly :-)

Dr. Manuel Kuehner

@yo' Ok, didn't know/remember that.

7:13 PM

@marmot The thesis is on the traditional Chinese medicine. So it is better to avoid mentioning tikzducks with something Chinese...

@CarLaTeX Yes, \duck[Peking] will probably never get implemented.

@marmot :)

@CarLaTeX We need a Panda!!!

@marmot Yesssss

7:27 PM

@DavidCarlisle do you know why pdftex doesn't support gif?

Joseph Wright

8:03 PM

@UlrikeFischer It's a tar-pit, I think

@JosephWright ? what does tar-pit mean?

Joseph Wright

@UlrikeFischer At the time of active pdfTeX development, there were security issues in the library support for GIF, plus the whole legal worries

@UlrikeFischer Somewhere one does not want to go!

@JosephWright Ah, just found the translation. I suspected something like this, but couldn't find any source.

@JosephWright at first I was wondering if you really wanted to answer me, or if this was a comment to the dark panda ...

9:24 PM

Hello!!

Is it correct to use the following code to create a new math environment?:

\documentclass{article}
\usepackage{amsmath}

\newenvironment{system}{\[\left\lbrace\aligned}{\endaligned\right.\]}

\begin{document}
Example:
\begin{system}
&x+y=2\\&x=2
\end{system}

\end{document}

If yes, wouldn't be more "correct" if the \[...\] delimiters go to the system environment i.e. \[\begin{system}...\end{system}\] instead of newenvironment?

Thanks!!

I think this new environment should work as align i.e. without \[...\] but I am not sure

@UlrikeFischer quite possibly same reason that gif support was removed from ghostscript (and dvips if I recall correctly) stupid patent rule around that time...

@manooooh looks Ok to me (you might want to put \ignorespacesafterend after the \]

@DavidCarlisle thanks!! However, when I compare it with cases it is shifted to the right :( :

\documentclass{article}
\usepackage{amsmath}
\usepackage[showframe]{geometry}

\newenvironment{system}{\[\left\lbrace\aligned}{\endaligned\right.\]\ignorespacesafterend}

\begin{document}
Example:
\begin{system}
&x+y=2\\&x=2
\end{system}

\[\begin{cases}x+y=2\\x=2\end{cases}\]

\end{document}

^^^^^ Why system is being shifted to the right?

@manooooh because you were misusing cases, example coming up

\documentclass{article}
\usepackage{amsmath}

\newenvironment{system}{A\[\left\lbrace\aligned}{\endaligned\right.B\]}

\begin{document}
Example:
\begin{system}
&x+y=2\\&x=2
\end{system}
Example:
\[
A\begin{cases}
x+y=2\\x=2
\end{cases}B
\]

\end{document}

@manooooh ^^ see the cases env is taking up invisible space for the second column that you were not using, so when centred it appears further to the left

@DavidCarlisle oh, you are right. Could you explain to me how I can get rid of the bad habit of not using & in cases? :/

@DavidCarlisle yes yes, even system is perfectly centered, thanks again!

9:40 PM

Oh I put A in the wrong place in the first example, but you saw the point anyway:-)

Yes, no problem

@DavidCarlisle I would like to be just as much with you as with @egreg: I liked your proposal better than the one that egreg answered here: tex.stackexchange.com/a/476123/152550 Could you post an answer so I can accept it, please? You have had the idea, I did not even know of the existence of aligned

10:08 PM

Another question:

We know that the natural way of writing delimiters is as it appears. For example, we should write $\sin 2$ and $\sin(2)$ :

However, the spacing is not the same if we use absolute value bars that when we add parentheses: \sin|2|:

My question is, should we respect this notation, or should we write $\sin\mathopen|2\mathclose|$ ?:

I mean, I am asking if we have to respect the same space using (), ||, [], etc. or not

@manooooh you can/should use \lvert and \rvert rather than |

@DavidCarlisle OMG yes! Thank you!

I knew that | occupies less horizontal space than any other symbol, but, of course, I was wrong when writing |: I should write \lvert :)

Thanks again

@manooooh \lvert, \mid and \rvert are all just | but with different math class

@DavidCarlisle ok. But when we are writing a math operator and then a delimiter we must use \lvert and not |. I was wrong

@egreg ok. Thank you!!

10:20 PM

@manooooh yes that's the idea, or define something like \abs to be \lvert#1\rvert then use \abs{2}

@DavidCarlisle tex.stackexchange.com/a/476392/152550 you forgot \ignorespacesafterend

@manooooh who cares, I took the tick of @egreg, that's the main thing:-)

LOL

@DavidCarlisle I like that idea. How can we do that? Using \def, newcommand, ..?

@manooooh `(re)newcommand is more idiomatic latex.

@DavidCarlisle what command do we have to redefine?

Oh I found this:

204

Q: Absolute Value Symbols

What is the "best LaTeX practices" for writing absolute value symbols? Are there any packages which provide good methods? Some options include |x| and \mid x \mid, but I'm not sure which is best...

Problem solved

10:36 PM

@manooooh Oh wow, 28 votes for the physics package! Now I can understand why election results can be surprising.

Hahaha

11:03 PM

@manooooh Be sure not to follow Peter Grill's advice of changing the behavior of *.

@egreg yes yes... I just went crazy because sometimes I need the command to inline math and sometimes to display

Because, as you said, it is wrong to always use \left and \right in inline math

But if I only use `\abs` then if I have something like `vmatrix` it is not the same size :( So I have to add `\makeatletter
\let\oldabs\abs
\def\abs{\@ifstar{\oldabs}{\oldabs*}}
%
\let\oldnorm\norm
\def\norm{\@ifstar{\oldnorm}{\oldnorm*}}
\makeatother`

@manooooh Don't be the cause of your own disgrace. Doing that is silly.

@egreg ok, thanks. Do you have an alternative?

Watch this:

\documentclass{article}
\usepackage[english]{babel}
\usepackage[utf8]{inputenc}
\usepackage[T1]{fontenc}
\usepackage{mathtools}

\DeclarePairedDelimiter\abs{\lvert}{\rvert}%
\DeclarePairedDelimiter\norm{\lVert}{\rVert}%
\makeatletter
\let\oldabs\abs
\def\abs{\@ifstar{\oldabs}{\oldabs*}}
%
\let\oldnorm\norm
\def\norm{\@ifstar{\oldnorm}{\oldnorm*}}
\makeatother

\begin{document}

$\abs*{\dfrac12}$

\[\abs{\det\begin{pmatrix}\sin\theta\cos\varphi&\sin\theta\sin\varphi&\cos\theta\\\rho\cos\theta\cos\varphi&\rho\cos\theta\sin\varphi&-\rho\sin\theta\\-\rho\sin\theta\sin\varphi&\rho\sin\theta\cos\…

@manooooh Use \abs* when you need extensible delimiters. What's the problem?

Do I have to insert \! manually?

11:15 PM

@manooooh \usepackage{mleftright}\mleftright

@egreg uhm... At least in Overleaf it has not improved:

\documentclass{article}
\usepackage[english]{babel}
\usepackage[utf8]{inputenc}
\usepackage[T1]{fontenc}
\usepackage{mathtools}

\usepackage{mleftright}\mleftright

\DeclarePairedDelimiter\abs{\lvert}{\rvert}%
\DeclarePairedDelimiter\norm{\lVert}{\rVert}%
\makeatletter
\let\oldabs\abs
\def\abs{\@ifstar{\oldabs}{\oldabs*}}
%
\let\oldnorm\norm
\def\norm{\@ifstar{\oldnorm}{\oldnorm*}}
\makeatother

\begin{document}

$\abs*{\dfrac12}$

\[\abs{\det\begin{pmatrix}\sin\theta\cos\varphi&\sin\theta\sin\varphi&\cos\theta\\\rho\cos\theta\cos\varphi&\rho\cos\theta\sin\varphi&-\rho\sin\theta\\-\rho\sin\…

@manooooh I won't talk to you any more if you don't remove that \oldabs code

@manooooh This is perfectly normal. No space to remove.

@manooooh Compare the spacing between det and the left parentheses before and after.

@egreg oh, I am sorry. I thought it was needed to use \abs* but it works when deleting \makeatletter...\makeatother. Sorry, I already deleted it

@egreg yes, the space between ) on the right and the | on the right

@egreg change is very small, it has not removed that space

We are counting det as the beginning and the ) as the last as the first and last expressions, respectively

@manooooh The parentheses have side bearings.

@egreg ok. It still seems to me that there is plenty of space at the end, but you are right: as the parentheses are larger there are more side bearings

11:34 PM

@manooooh Don't worry. That space is good.

@manooooh And, personally, I don't find \abs useful in that context. Too big material in one pair of braces makes the thing difficult to parse.

@egreg I am using change of variables, so I have to find the jacobian

I can not do anything else, the Jacobian is the absolute value of the determinant of the change of variables

@manooooh I'd use explicit \left and \right:

\documentclass{article}
\usepackage{mathtools}
\usepackage{mleftright}
\mleftright

%\DeclarePairedDelimiter{\abs}{\lvert}{\rvert}

\begin{document}

\[
\left|
  \det
  \begin{pmatrix}
  \sin\theta\cos\varphi&\sin\theta\sin\varphi&\cos\theta\\
  \rho\cos\theta\cos\varphi&\rho\cos\theta\sin\varphi&-\rho\sin\theta\\
  -\rho\sin\theta\sin\varphi&\rho\sin\theta\cos\varphi&0
  \end{pmatrix}
\right|
\]

\end{document}

@manooooh I find it much clearer ^^^^^

For small pieces of text, \abs is more convenient, though.

\[
\left|
  \det
  \begin{pmatrix}
  \sin\theta\cos\varphi      & \sin\theta\sin\varphi     & \cos\theta \\
  \rho\cos\theta\cos\varphi  & \rho\cos\theta\sin\varphi & -\rho\sin\theta \\
  -\rho\sin\theta\sin\varphi & \rho\sin\theta\cos\varphi & 0
  \end{pmatrix}
\right|
\]

@manooooh Better yet ^^^^

@manooooh But also

\[
\abs*{
  \det
  \begin{pmatrix}
  \sin\theta\cos\varphi      & \sin\theta\sin\varphi     & \cos\theta \\
  \rho\cos\theta\cos\varphi  & \rho\cos\theta\sin\varphi & -\rho\sin\theta \\
  -\rho\sin\theta\sin\varphi & \rho\sin\theta\cos\varphi & 0
  \end{pmatrix}
}
\]

@egreg I find it much more advisable to use \abs* since it alerts the user who is reading the code that "Hey, now comes an absolute value". In contrast, with \left|\right| it is not very clear. Why do you say "For small pieces of text, \abs is more convenient, though"?

@manooooh Habits, perhaps. I find $\abs{x}$ better than $\lvert x\rvert$ . But if the code in the argument of \abs is large, it becomes difficult to find the closing brace, unless the code is laid out properly as shown above.

11:49 PM

@egreg in both Overleaf and TeXnicCenter you can know the pair of symbols by just passing the mouse over any of them, so the problem of not finding an opening or closing symbol no longer exists (at least for me):

Of course if you are programming with plain code you do not have this features, but meh