9

How do I make the following integral symbols in Latex?

Enter image description here Enter image description here

I know how to make an integral and the limit. I don't know how to make it with the lines through it.

Peter Mortensen
  • 1,028
RisayaKinan
  • 93
  • Have you checked detexify? – dustin Oct 04 '14 at 17:46
  • Yes I just have. Didn't know about it. It only finds an integral with a line through it in the middle. – RisayaKinan Oct 04 '14 at 17:50
  • Welcome to TeX.SX. Do both symbols mean something different (i.e. upper/lower limts or something like that)? If not, it might just be a choice of the font used. – Johannes_B Oct 04 '14 at 17:54
  • 2
    Those symbols rather look a giant f letter than a true integral symbol –  Oct 04 '14 at 18:02
  • The are used by my professor in relation to the riemann integral. One is defined as infimum and one as supremum from the Darboux upper or lower sum. – RisayaKinan Oct 04 '14 at 18:03
  • The most similar I know is \fint and \fintop of package esint. Perhaps esint.sty could be a good starting point to make another custom integral symbol. – Fran Oct 04 '14 at 18:36
  • after a fair amount of research trying to find examples of these integral symbols, it seems to me that a more common notation uses the shapes of the unicode characters U+2A1B and U+2A1C, with horizontal bars below and above the integral. if someone can provide a reference to published examples of these "variants", i will be happy to propose them to the unicode committee as recognized variants. – barbara beeton Jan 15 '15 at 17:15

2 Answers2

12

Using the code from the entry The Principal Value Integral symbol (which defines the macro \dashint) in the TeX FAQ as a starting point, it is reasonably straightforward to define two new macros, \lowdashint and \highdashint, that place a "dash" symbol -- actually, a "minus" symbol -- a bit lower and a bit higher, respectively, than \dashint does.

In the code below, the macros \lowdashint and \highdashint are set up only for display-style and text-style math modes. (I can't imagine they'll occur in expressions in first-level, let alone second-level, subscripts and superscripts. However, please tell me if this assumption is invalid.)

You should, of course, feel free to change the vertical positions of the dashes -- cf the arguments of the \lower and \raise commands -- to suit your stylistic preferences.

enter image description here

\documentclass{article}
\usepackage{booktabs,amsmath}
\def\Xint#1{\mathchoice
    {\XXint\displaystyle\textstyle{#1}}%
    {\XXint\textstyle\scriptstyle{#1}}%
    {\XXint\scriptstyle\scriptscriptstyle{#1}}%
    {\XXint\scriptscriptstyle\scriptscriptstyle{#1}}%
      \!\int}
\def\XXint#1#2#3{{\setbox0=\hbox{$#1{#2#3}{\int}$}
    \vcenter{\hbox{$#2#3$}}\kern-.5\wd0}}
\def\dashint{\Xint-}

\def\Yint#1{\mathchoice
    {\YYint\displaystyle\textstyle{#1}}%
    {\YYYint\textstyle\scriptscriptstyle{#1}}%
    {}{}%
    !\int}
\def\YYint#1#2#3{{\setbox0=\hbox{$#1{#2#3}{\int}$}
    \lower1ex\hbox{$#2#3$}\kern-.46\wd0}}
\def\YYYint#1#2#3{{\setbox0=\hbox{$#1{#2#3}{\int}$}
    \lower0.35ex\hbox{$#2#3$}\kern-.48\wd0}}
\def\lowdashint{\Yint-}


\def\Zint#1{\mathchoice
    {\ZZint\displaystyle\textstyle{#1}}%
    {\ZZZint\textstyle\scriptscriptstyle{#1}}%
    {}{}%
    !\int}
\def\ZZint#1#2#3{{\setbox0=\hbox{$#1{#2#3}{\int}$}
    \raise1.15ex\hbox{$#2#3$}\kern-.57\wd0}}
\def\ZZZint#1#2#3{{\setbox0=\hbox{$#1{#2#3}{\int}$}
    \raise0.85ex\hbox{$#2#3$}\kern-.53\wd0}}
\def\highdashint{\Zint-}


\begin{document}
$\begin{array}{@{}lccc@{}}
\toprule
\text{Math mode} & \multicolumn{3}{c@{}}{\text{Integral symbol}}\
\cmidrule(l){2-4}
& \texttt{\string\lowdashint} 
& \texttt{\string\highdashint}
& \texttt{\string\dashint} \
\midrule
\texttt{\string\displaystyle} 
& \displaystyle \lowdashint_M f
& \displaystyle \highdashint_M f
& \displaystyle \dashint_M f \[4ex]
\texttt{\string\textstyle} 
& \lowdashint_M f
& \highdashint_M f
& \dashint_M f \
\bottomrule
\end{array}$
\end{document}
barbara beeton
  • 88,848
Mico
  • 506,678
1

One way of placing bars through math characters is to use \ooalign. In the two commands I've defined below, the first number controls how high the bar is on the integral sign, the second controls the length of the bar, and the third controls the thickness of the bar.

\documentclass{article}

\newcommand{\stI}{%
\ooalign{\hidewidth $\int$\hidewidth\cr\rule[1.1ex]
{1ex}{.4pt}}}

\newcommand{\stII}{%
\ooalign{\hidewidth $\int$\hidewidth\cr\rule[-0.1ex]
{1ex}{.4pt}}}

\begin{document}
\[\stI_m f\]
\[\stII_m f(x)\]
\end{document}

To get these looking nice, you might also want to read this answer: Big integral sign

Community
  • 1
Steven Gardiner
  • 346
  • With this setup, the m symbol is placed in the "ordinary" subscript position, rather than much lower near the lower edge of the integral symbol. – Mico Oct 04 '14 at 20:22
  • One fix is to wrap the ooalign with \mathop{}. That being said, I think your solution is better. – Steven Gardiner Oct 04 '14 at 20:48