2

I want to force the content of one of the cells in my table to overflow the cell width by a specified amount. With c columns, negative kerning does the trick. But I also want vertical alignment, and negative kerning no longer works with m columns.

Minimum working example:

\documentclass{article}

\usepackage{array}
\newcolumntype{M}[1]{>{\centering\arraybackslash}m{#1}}

\begin{document}

%\begin{tabular}[h]{|M{1in}|M{1in}|M{1in}|}
\begin{tabular}[h]{|c|c|c|}
  \hline
  \rule{0.2in}{1in} &
  \rule{1in}{0.2in} &
  \rule{0.2in}{1in} \\
  \hline
  \rule{1in}{0.2in} &
  \rule{0.2in}{1in} &
  \rule{1in}{0.2in} \\
  \hline
  \rule{0.2in}{1in} &
  \kern-0.5in\rule{1.5in}{0.2in} &
  \rule{0.2in}{1in} \\
  \hline
\end{tabular}

\end{document}

enter image description here

The bottom middle cell shows what I mean. Toggle the commenting on the \begin{tabular} lines to see it fail with m columns. How can I get it to work?

Motivation: I have a column containing a bunch of images with similar, aligned content. Most of them are the same width, but one is wider because the extra space contains some additional information like labels. I want the extra space to be ignored so that the images remain well aligned in my document. I don't want to simply use left- or right-alignment, because that wastes the extra space in the other cells.

1 Answer 1

2

Use \makebox instead of the \kern:

\documentclass{article}

\usepackage{array}
\newcolumntype{M}[1]{>{\centering\arraybackslash}m{#1}}

\begin{document}

\begin{tabular}[h]{|M{1in}|M{1in}|M{1in}|}
%\begin{tabular}[h]{|c|c|c|}
  \hline
  \rule{0.2in}{1in} &
  \rule{1in}{0.2in} &
  \rule{0.2in}{1in} \\
  \hline
  \rule{1in}{0.2in} &
  \rule{0.2in}{1in} &
  \rule{1in}{0.2in} \\
  \hline
  \rule{0.2in}{1in} &
  \makebox[1in][r]{\rule{1.5in}{0.2in}} &
  \rule{0.2in}{1in} \\
  \hline
\end{tabular}

\end{document}

Explanation: \makebox[1in][r]{\rule{1.5in}{0.2in}} puts the rule in a box of width 1in and right-aligns it in the box.

enter image description here

You must log in to answer this question.