Apostrophe breaks comments in lstlisting

Question

I defined a command to simplify the writing of code listings of (pseudo-) C code:

\lstdefinestyle{c}{
language=C,
basicstyle=\small\sffamily,
numbers=left,
numberstyle=\tiny,
frame=L,
columns=fullflexible,
showstringspaces=false
}

\lstnewenvironment{ccode}{
\mathligsoff\lstset{style=c}}{}

If I use the environment ccode everything works fine unless apostrophes are used:

\begin{ccode}
  int x';
  // use x
  x = 3;
\end{coode}

In this listing the comment is not printed in italics but in 'straight' characters. If I remove ' or add another ' the italics are restored. This is the complete .tex code:

\documentclass[]{report}
\usepackage{listings}
\usepackage{semantic}
\lstdefinestyle{c}{
  language=C,
  basicstyle=\small\sffamily,
  numbers=left,
  numberstyle=\tiny,
  frame=L,
  columns=fullflexible,
  showstringspaces=false}
\lstnewenvironment{ccode}{
\mathligsoff\lstset{style=c}}{}

\begin{document}
\begin{ccode}
  int x';
  // hi there
  x = 3;
\end{ccode}
\end{document}

EDIT: The accepted answer works flawlessly for me. Just a hint for other readers: If no strings or characters need to be displayed -- and thus no code like 'a' or 'string' -- \lstset{literate={'}{{'}}1} will also do the trick.

Answer 1

As Jubobs commented while I was working on this answer, the problem is that ' is treated as a string delimiter in C, so everything until the next ' is considered part of a string. You can see this by setting stringstyle=\itshape or similar; everything following the ' is shown as a string.

Discussion in the comments notwithstanding (I would recommend something better suited for pseudo-code instead of listings as well), if you must stick with using listings one solution is to define your own language pseudoc as a copy of c as defined in lstdrvrs, and change the ' character from a b-type delimiter to a m-type (MATLAB-style) delimiter.

\documentclass{article}
\usepackage{listings}

% modified ANSI C definition; copied from lstdrvrs.dtx
\lstdefinelanguage{pseudoC}%
  {morekeywords={auto,break,case,char,const,continue,default,do,double,%
      else,enum,extern,float,for,goto,if,int,long,register,return,%
      short,signed,sizeof,static,struct,switch,typedef,union,unsigned,%
      void,volatile,while},%
  sensitive,%
  morecomment=[s]{/*}{*/},%
  morecomment=[l]//,% nonstandard
  morestring=[b]",%
  morestring=[m]',% changed from `b' to `m'
  moredelim=*[directive]\#,%
  moredirectives={define,elif,else,endif,error,if,ifdef,ifndef,line,%
  include,pragma,undef,warning}%
}[keywords,comments,strings,directives]%

\lstdefinestyle{pseudoc}{
language=pseudoC,
basicstyle=\small\sffamily,
numbers=left,
numberstyle=\tiny,
frame=L,
columns=fullflexible,
showstringspaces=false,
}

\lstnewenvironment{ccode}{
\lstset{style=pseudoc}}{}

\begin{document}
\begin{ccode}
  int x';
  // use x
  x = 3;
  y = 'a string'; // aside: this would be a compiler error :-)
\end{ccode}
\end{document} 

I added the last line of the listing to make sure normal string behavior is preserved: