and the noto sans bengali example in @UlrikeFischer's answer works as far as I can tell:-)

@DavidCarlisle I knew that of course, but when I was initially designing it, I had in my mind that #1 was always a single token. Then, when I expanded the concept to handle grouping, that one slipped through the cracks.

@StevenB.Segletes আপনি ক্ষমা করা হয়!

@DavidCarlisle ধন্যবাদ, জ্ঞানী মাস্টার।

@DavidCarlisle nice ;-)

@egreg I have a description in my answer on the main site.

@DavidCarlisle Yay!

Hello

Has anyone built a handwriting to latex converter

Not very recent, but may be something of interest there.

Yeah, ive gone through thtat post. The thing is, I plan on working on a ocr latex converter. Just wondering if it will be worth it.

@NoLand'sMan mathpix is quite impressive on images of typeset math their docs say "try us on hand written notes" so might be worth uploading a few scans mathpix.com/ocr

\begin{talking} hello

@T.D. 'ello

@PauloCereda \end{duck}

@DavidCarlisle oh no

@UlrikeFischer Thanks, I'll try ASAP.

@egreg it worked for me:-)

@DavidCarlisle Also for me. Thanks @UlrikeFischer

@DavidCarlisle ooh

@PauloCereda something of interest to you here bbc.co.uk/news/uk-england-northamptonshire-48466347

@DavidCarlisle ooh

oh

Northampton

Oh cobblers

@T.D. I suspect the reference will be lost on most here:-)

;D

Can anyone help me on which package I should use to graph functions

Tikz v. pgfplots

?

@T.D. At concrete functions like y=f(x) pgfplots is easier.

:50517792they both use pgf, pgfplots is written specifically for the purpose so possibly that is most natural,

however tikz is more flexible right?

@T.D. there was a time people used mathematica or matlab and just input it...

@T.D. not necessarily

mathematica

@T.D. You can use TikZ as well. A main difference is: You have to draw your coordinate sytem by your own.

oh ok

any good resslources out there you can recommand

?

@T.D. pgfplots manual?

I am a math student/ teacher and my knowledge is quite limited

I saw that comming

coming*

@T.D. you can edit the history (for 2 minutes)

well thanks for you time

watch ^^

gotta go real quick

thanks ! good ol' irc habit

@T.D. if you prefer examples rather than a manual, there are thousands of them on this site

ok

well I have to afk see you in 2hrs

@T.D. Wait... ;)

pgfplots

\documentclass{article}
\usepackage{pgfplots}
\pgfplotsset{compat=newest}
\begin{document}
\begin{tikzpicture}
\begin{axis}[axis lines=middle, grid,
title={$y=x-\lfloor x \rfloor$},
title style={blue},
]
\foreach[evaluate={\M=\m+1}] \m in {-2,...,2}{
\addplot[blue, domain=\m:\M]{x-floor(x)};
\addplot[blue, mark=*] coordinates {(\m,0)};
\addplot[blue, mark=*, mark options={fill=white}] coordinates {(\M,1)};
}
\end{axis}
\end{tikzpicture}
\end{document}

TikZ

\documentclass{article}
\usepackage{tikz}
\begin{document}
\begin{tikzpicture}[>=latex,
y=3.5cm,  x=1.3cm,
Kurve/.style={blue}
]
% Gitternetzlinien
\draw[ystep=0.25, xstep=1, help lines] (-2.125,0) grid (3.125,1.125);

% KoSy
\draw[->] (-2.5,0) -- (3.5,0) node[right, below]{$x$};
\draw[->] (0,-2pt) -- (0,1.2) node[left]{$y$};

\foreach \x in {-2,...,3}{%
\if\x0\node[below] at (0,-2pt){$0$};%
\else\draw[shift={(\x,0)}] (0,2pt) -- (0,-2pt) node[below]{\x};\fi}

\foreach \y in {0,0.25,...,1}{%
\if\y0\else\draw[shift={(0,\y)}] (2pt,0) -- (-2pt,0) node[left]{\y};\fi}

Oh dear

pgf/100

Same "curve" in pgfplots and TikZ

thks

see you all

@cis \if\y0\else, \if\y0 \else would be less scary

@DavidCarlisle I think, this was a snipplet I used in a complicated example.

("I" used, not you) :()

Ah ok, th thing was: Do not print the zero on the x-axis. How would you do?

@cis without the space the following\node expands before the \if test (ie it always is expanded whether or not \y is 0. That possibly makes no difference in this context but it is relying on internals of the definition of \node.

@cis always have a space after a literal number

@DavidCarlisle Ah ok, I understand

\edef\zzz{\firstmark} \edef\qqq{\botmark} \ifx\zzz\qqq or perhaps \protected@edef if they have unknown content

@DavidCarlisle Ok, let me try... ;)

@cis but latex does \let\firstmark\botmark in \@outputpage so you have to interpret your tests with care

@cis you are "supposed" to use \leftmark and \rightmark rather than access \firstmark directly but sometimes that isn't quite the right thing

@DavidCarlisle nice, waiting for the bug reports :)

@DavidCarlisle I am not sure, I can undersatnd you. I used a commad:

% Parnums in head
\newcommand\parenumrange{%
\edef\CurrentFirstmark{\firstmarks1}%
\edef\CurrentBotmark{\botmarks1}%
\ifx\CurrentFirstmark\CurrentBotmark%
\ifnum\theparnumber>0{(\parnumstyle{\botmarks1})}\else\fi%
\else%
\ifnum\theparnumber>0{(\parnumstyle{\firstmarks1--\botmarks1})}\else\fi%
\fi}

It works fine... ;)

@cis oh they are etex firstmarks not classic firstmark, latex doesn't touch them.

@DavidCarlisle Ah ok. Very good.

@cis note that because latex doesn't touch them, \botmark1 will be empty after a float page not carried over from the previous text page (as latex does with \botmark)

@cis (unless koma script adds additional support for e-tex marks?)

@DavidCarlisle Puh, I don't know.

It gives somithing like this:

@cis yes but if a float drifts into your list do you lose the count in the header (I would guess you do) but if you don't have floats in that section then it's nothing to worry about.

actually you are probably OK (you are carrying \CurrentBotMark by hand not using \firstmark)

test \par \parnum
test \par \parnum
\begin{figure}
\rule{3cm}{7cm}
\caption{myfigure}
\end{figure}
test \par \parnum
test \par \parnum
test \par \parnum
test \par \parnum

@DavidCarlisle There was some float effect and NO problems with the "parnums" ;)

@cis well I meant "by explicit saving values in the macro layer not relying on the tex primitive \firstmark

General question for the pros here: I am aware of two packages that "fix" amsmath: mathtools and nccmath.

Are there activities to include those fixes back into amsmath?

(Is this a question for meta?)

@Dr.ManuelKuehner Probably best here, since it has nothing to do with the site operations (which is what meta is for). Maybe @barbarabeeton has some insight, although she has formally retired from the AMS.

@AlanMunn Thx for the reply.

@marmot Seriously, you want credit for \centering? This is taking the credit thing too far IMO. Using \centering instead of center inside a float is basic knowledge that any reasonably competent LaTeX user knows. It's one thing to be complaining about credit for substantive answers that get reused without attribution, but I really don't think you should be complaining about things like this.

@AlanMunn No, it is always the same user who has a long history of doing this, previously as Herbert now as Red Cloud.

@marmot Well I'll stay out of it, but in this case especially it's just provocation.

@AlanMunn Well, this question is about centering, isn't it? You criticize me, rightfully, for defining a diagram environment that solves the problem and the next user who does the same gets upvoted?

@marmot The issue isn't about who gets upvoted, it's about the fact that wanting to be credited for something like \centering (especially if Red-Cloud is who you say they are) is really stretching the 'I want credit' thing too far. Anyway, I've said my piece and I'll stop now.

@AlanMunn I understand what you are saying but he has been picking on me for ages and continuously. I am just sick of it.

@marmot He hasn't been 'picking on you'; he's an equal opportunity re-purposer. :) I have had the same thing happen to me too before you joined the site. I never thought it was personal.

@AlanMunn Yes, maybe. I also agree that \centering is well known, but it does not seem to be known to the OP. It is all not that important. (Yet I do not understand why his answer is not subject to the same criticism as mine. He also (re-)defines the diagram environment, even in a way that is not consistent with the OP's syntax.)

I'm gonna turn bananas!

Good evening, fine fellows!

It finally happened, I have had my first time.

it is longer than it usually is

@thymaro So that they get more sun? :) (You mean 'go bananas' )

I don't know, English is not my native language.

Anyways... the longer-than-usual thing is a table

@thymaro I know. Just trying to help out on the idiom front. :)

it spans more than one page

haha thanks for that

@thymaro So are you using longtable?

I don't know if I should and that's precisely what's making the bananas go bad.

I am confused by the chronology of table

tabular

tabularx

longtable

@thymaro Well either you split the table into two tables, or you use longtable.

xlttable

supertabular

ui... I shouldn't write too many messages in quick succession

I get from your reaction, that longtable is the way to go in 2019.

@thymaro tabularx is to have columns that fit the width of a table. It has nothing to do with breaking tables across pages. tabular is just the plain tabular environment. longtable and supertabular both split tables across pages equally well. There may be some corner cases in which one works better than the other.

I think most people use longtable.

right now, the table I want to span pages is a very simple one: \begin{tabular}{ll}. it doesn't even want to be recognized as a table and it shouldn't.

@thymaro So longtable should be completely fine.

I was merely trying to find out what, if I happened to need to make a longer table (with caption and fancy rows) what package I should be starting to study.

longtable it is, then.

@thymaro A reliable choice for sure.

thanks a lot for confirming that.

I can feel my blood pressure going down already

@thymaro And (as I think you know) it doesn't get wrapped in the table environment, since it is not a float. But if you provide a \caption to it, the caption will be listed in the list of tables. So if you don't want that, don't use \caption.

@AlanMunn indeed

one more question. If I happened to need to use tabularx's functionality in a longtable, is ltxtable still a valid choice or have tabularx and longtable evolved past the point where ltxtable is able to take good care of their combination?

@AlanMunn is that so? I know tex far less than you guys, and so I search for tex code often. my impression was the tabularx was the most common

@Lacey They do completely different things.

oh well then, please continue ignoring my silly statement :p

wait, aren't they both used to create tables?

tabularx has 1.3k questions while longtable has 1.6k, so im wrong regardless :D

ok let me ask this: should i switch my existing code that uses tabularx to longtable? will i have any benefits by doing so?

@Lacey So is Excel :P

@thymaro Yes, you still have to use ltxtable and the longtable itself has to be in a separate file.

@Lacey No. longtable is for tables that span multiple pages. tabularx is for expandable columns that fit a fixed width table.

@AlanMunn you mean like "it works fine for longer tables, as well as short tables", or "it is only meaningful for longer tables"?

@Lacey Well it won't fail if the table is short, but since it has other restrictions (it doesn't float, and as I was just discussing with @thymaro it can't be used in conjunction with tabularx out of the box) it's not advisable to use it unless you know the table will span multiple pages.

@AlanMunn so for example, what do you use most of the time, for tables?

@Lacey I mainly use tabular, and tabularx and always with the array package (for custom columns) and booktabs (for nice spacing and rules).

@Lacey If I'm typesetting lots of numerical data, then I also usually use siunitx as well, which provides an S column for formatting numerical data of various sorts.

@Lacey It really depends a lot on the particular use case. There's no 'one size fits all' solution.

@AlanMunn I was actually thinking of creating my own few macros like \table3{a}{b}{c}{1}{2}{3}{q}{w}{e}... for my own style of table use

but the adventure didn't go on for long because newdocumentcommand has a limit of 9 arguments

what would you suggest to go over that argument limit?

@Lacey Well table3 is an invalid name to begin with... And it's never a good idea to reinvent wheels. :)

tablethree, then :D

@Lacey If you need >9 arguments you're doing something wrong. :)

@AlanMunn i just mess up the table syntax easily, and it isn't intuitive for me

@AlanMunn well each argument would be one item of the table, so for tables it wouldn't be so rare to have over 9 items

@Lacey Right, but then you should use a comma delimited list for example.

@Dr.ManuelKuehner mathtools probably yes (it's also accepted by the ams)

@AlanMunn so that's the go-to solution when one wants more arguments, then?

@Lacey Again, depending on the use case. Another way is to use a key/value interface. it really depends on what you're doing.

@Lacey For example, I use TeXShop on the Mac which allows you to copy cells from a spreadsheet into your LaTeX source and they get formatted as LaTeX. I don't use it all the time, but for a quick an dirty table, it's as fast as a cut and paste.

@Lacey eek why would you do that?

@DavidCarlisle well because the table syntax is easy to mess up for me

consider that I have lived for a shorter amount of time than the time you have been using your own libraries :p

@Lacey that would be much harder, how could you tell how many rows or columns it has, or where it ends, and if you replace a by more typical content containing {} etc nested it would be impossibly difficult to tell the cell content by eye,

@DavidCarlisle it would only support my own specific kind of table, and tablethree would be the one that has 3 items on each row. likewise there would be a tabletwo and tablefour.

tex tables are powerful, but they aren't easy. i could do fine with a very limited version of them.

@Lacey and how would they end? also real tables always need a column specification, centred or align on decimal point or ...

@DavidCarlisle a specific document would only use the same type of table(so i wouldn't center some rows while left alignin others), so there'd be a setting. this macro would end up producing your tabularx code, basically.

but all the tables in the same document would look the same, that's one of the reasons i wanted to write the macro.

@Lacey you avoided saying how the table ends, it would still be a bad idea I'd say as are most uses of tabularx if it is numeric data, tabularx is all about _line breaking) within a cell and relatively few tables have paragraphs of text in the cells, so tabularx should be used less than it is.

@Lacey But you can do that without wrapping everything in a clunky macro.

@DavidCarlisle and you should stop putting the dagger next to your name ftp.cc.uoc.gr/mirrors/CTAN/macros/latex/required/tools/… i thought you died :p

@AlanMunn i have to keep around/remember a long tabularx use

it isn't the easiest thing to memorize

@Lacey Then make a macro in your editor.

@Lacey standard latex \thanks

@DavidCarlisle why would they use the symbol used to indicate someone has died :D

@AlanMunn i mean would you be against it if i defined a macro in c that i could have used the editor for as well?

@Lacey it's usually a bad idea to hide the environment syntax, \begin.. \end is the fundamental syntactic construct in latex, and you get many benefits from using it, editors can give syntax support and navigate by jumping over or commenting out environments etc, if you make up som custom syntax then you get no editor support and it is harder for you or anyone else looking at the code to see the structure

what's the point of a macro language if i can't shape it for my own specific taste

@DavidCarlisle eh I already have no editor support

@Lacey define new macros or new environments yes, but don't break the basic structure of the language just because it is flexible enough to allow you to do that

@Lacey what editor do you use?

@DavidCarlisle Thx for the reply!

@DavidCarlisle sublime text 3. im sure there is a latex package for it, i just never installed one

@Lacey I divide editors into those that are emacs and those that are not, and that falls into the not emacs category.

@Lacey Not sure I understand. What @DavidCarlisle and I are saying is that you shouldn't be writing macros to redefine core functionality. If you have trouble with the syntax, use your editor to help you remember.

@DavidCarlisle :) @Lacey But Sublime Text does have a TeX plugin I think.

@DavidCarlisle i would use emacs if it had decent font rendering :p

@Lacey github.com/SublimeText/LaTeXTools

@Lacey that's tex's job

@DavidCarlisle i mean you stare at the code all day -- or a lot of hours, anyway. it helps to see something that looks slightly better

@Lacey I have used emacs more or less every day since 1987 I have survived so far.

@DavidCarlisle oh absolutely, im not suggesting sublime text over emacs.

just saying that while it is inferior in a programming sense, i prefer the visual tradeoff, as i use not many features of any editor

@Lacey but that's a digression the same things apply to syntax highlighting on this site and other things, the environment syntax is what makes latex latex.

@DavidCarlisle i really don't think the snytax is the reason people use latex. it is characteristic for sure, though

@Lacey but you never said how you wanted to specify the number of rows in your \tablethree or do you mean that it is really a fixed format with a ficed number of rows and columns?

@DavidCarlisle I was thinking I would detect it by the parameter count, using the feature in the newdocumentcommand

@Lacey no they use it because it has a coherent (more or less) ecosystem of extensions and packages that more or less play well together and allow arbitrary extended functionality, and that ecosystem relies on macro authors not breaking too many rules

IfValueT of xparse perhaps?

@Lacey there are no technical difficulties excpet that your sketch neither specified number of rows not had a terminator which is why I asked if you mean that the macro only makes a fixed table size with a pre-determined number of rows and columns?

@DavidCarlisle pre-determined number of columns yes, rows no

these two would be equally valid

\tablethree{a}{b}{c}{1}{2}{3}
\tablethree{a}{b}{c}{1}{2}{3}{q}{w}{e}

@Lacey so I ask again, how do you terminate the thing if you do not use `\begin{mytable} \end{mytable}

@DavidCarlisle inside this macro, it would produce that

@Lacey but the 7th argument of teh first table is \tablethree which, if you want a nested table seems perfectly bvalid how do you terminate it

@Lacey no that should be the syntax in the document

@DavidCarlisle this would be very limited. no nested tables.

who even uses nested tables? :D

this macro I define would not be more powerful than xtabular itself. it would be very limited.

and that would help with keeping the exact same style across the document

@Lacey but remember that {} groups are optional so your input is the same as is \tablethree abc123\tablethree abc124qwe how is a human ever going to work out where the table ends even if you specify some arcane rule for tex?

@DavidCarlisle if they wanted to go against the way I intend the code to be used, they could take another shortcut and misspell the command instead. I wouldn't care for anything other than the ways I intend it to be used

@Lacey no, it is perfectly reasonable to hide a fixed table spec so \begin{mytable} is \begin{tabular}{ccp{4cm}|}` or whatever and so get a consistent table style just specified in one place, but the rest of the syntax is just making things hard

@Lacey you can't tell the difference from inside tex, they are the same thing. You have still not specified any implementable rule on how to stop the table.

@DavidCarlisle the macro itself would include the ending part.

so the use would be

\tabletwo
{animal}{age}
{bear}{5}
{parrot}{3}

it isn't like the ending part is something to be computed, it is always the same thing. trivial for a macro to include

I give up:-) that has no useable termination

@DavidCarlisle what do you mean with that? i have defined similar commands before. the command wouldn't be its own specific command, it would just an alias of the xtabular in a way

@Lacey and what if the thing after the table is {\bfseries zzz}{foo} how do you tell that is not part of the table?

@DavidCarlisle i think you wouldn't. isn't that a tex issue though?

i mean the same applies for every command i define no?

@Lacey you need an end marker or you need to specify the number of rows, that is nothing to do with tex, it is just a fact of life.

@Lacey no it does not apply to any command defined by standard latex constructs.

@DavidCarlisle i have defined similar commands before though, i just never had an use case for that example. i acknowledge it as a valid issue, but i just don't see anyone writing that on purpose

to be clear, i wouldn't be trying to convert the tex community to use my specific tables. this would be for my own use and the people who edit the document. very limited in use.

@Lacey I can not even guess how you intend to stop the table and distinguish the text following the table from the last row of the table, so even for your own use, I could not suggest any implementation

@Lacey why do you not want something simple like \end{mytable} to mark where the table ends????

@DavidCarlisle i would use it if i need it. i just havent. i think i may for lists separated by commas?

a primitive version of what I had in mind:

this can't have more than 9 parameters though

@Lacey as I say I give up, you clearly have something in mind but I really can not guess. But if the syntax is that hard to specify it is not likely to be easily understood by anyone reading the document.

g is a really bad idea the xparse documents do warn against its use

@Lacey you have no following text presumably "end of file" is not the intended marker to end all tables.

@DavidCarlisle I mean \underline{text} doesn't specify when the underline would end and that doesn't look hard to recognize at least to me

a table can end just like everything else can

@Lacey ??? the underline just applies to the argument, that is completely different. no I really do give up I suggest we drop this,

@DavidCarlisle sure

@Lacey What's the advantage over, say

\tabletwo{
animal & age \\
bear & 5 \\
parrot & 3
}

I don't see any advantage in your code.

@egreg to be clear, im not advocating for the code sample I put out. I posted that to get the point across that I have been using this without the "end" part just fine.

@egreg how would I implement this?

I literally don't know how to implement a command that uses that syntax

@egreg I don't see an advantage over it either, I would be fine with both honestly

actually, I think I can only use that syntax because of the 9 parameter limit :p

@Lacey \newcommand{\tabletwo}[1]{\begin{tabular}{cc}#1\end{tabular}}

@egreg @Lacey Or slightly more flexible: \newcommand{\tabletwo}[2][cc]{\begin{tabular}{#1}#2\end{tabular}} which would allow you to change the cc specification in the optional argument if needed.