Pandoc: Markdown to PDF, without cutting off code block lines that are too long

Question

I'm trying to convert a Markdown file into a PDF file using Pandoc.

I have a lot of code blocks in my Markdown, and if I have a line of code that is too long, it will just be cut off in the PDF. How can I make Pandoc wrap lines that are too long?

Here's the line I'm using right now:

pandoc input.md --highlight-style tango -o output.pdf;

Answer 1

(See Update at the end, to get proper colors)

By default pandoc uses its own highlighting engine, but it can be changed to use package listings instead. Simply add --listings option to your pandoc command line.

However, the format used by listings does not break long lines either, but you can prepare all the required options in a separate tex file, for example:

% Contents of listings-setup.tex
\usepackage{xcolor}

\lstset{
    basicstyle=\ttfamily,
    numbers=left,
    numberstyle=\footnotesize,
    stepnumber=2,
    numbersep=5pt,
    backgroundcolor=\color{black!10},
    showspaces=false,
    showstringspaces=false,
    showtabs=false,
    tabsize=2,
    captionpos=b,
    breaklines=true,
    breakatwhitespace=true,
    breakautoindent=true,
    linewidth=\textwidth
}

And then compile with the line:

pandoc test.md --listings -H listings-setup.tex -o output.pdf

Unfortunately, this solution cannot use the --highlight-style option too. If you want to get the tango colorscheme, you have to code it yourself as options for the listing packages, and include them in the \lstset above.

Given the file test.md as follows:

% Title
% Author

# First section

Some code with long lines:

```java
public class Test {
    static Set<Thread> updateThreads = new HashSet<Thread>();

    public static void main(String[] args) {
        ConcurrentMap<Integer, String> concurrentMap = new ConcurrentHashMap<Integer, String>();
        for (int i = 0; i < 1000; i++) {
            startUpdateThread(i, concurrentMap);
        }
        for (Map.Entry<Integer, String> entry : concurrentMap.entrySet()) {
            System.out.println("Key :" + entry.getKey() + " Value:" + entry.getValue());
        }
        for (Thread thread : updateThreads) {
            thread.interrupt();
        }
    }
}
```

That's it.

The result generated by the previous pandoc command line is:

Update

In order to mimic closely the result obtained with --highlight-style tango option (and no listings package), I ran pandoc asking for the intermediate .tex file to discover how the colors were defined for that case.

I wrote the following setup using listings options, which use the same colors (only where applicable, for example listings has no mean to set a specific color for numeric constants or other syntactic elements). In addition I reduced the spacing among chars, which were too far apart with default setup. This is the new setup:

% Contents of listings-setup.tex
\usepackage{xcolor}

\lstset{
    basicstyle=\ttfamily,
    numbers=left,
    keywordstyle=\color[rgb]{0.13,0.29,0.53}\bfseries,
    stringstyle=\color[rgb]{0.31,0.60,0.02},
    commentstyle=\color[rgb]{0.56,0.35,0.01}\itshape,
    numberstyle=\footnotesize,
    stepnumber=1,
    numbersep=5pt,
    backgroundcolor=\color[RGB]{248,248,248},
    showspaces=false,
    showstringspaces=false,
    showtabs=false,
    tabsize=2,
    captionpos=b,
    breaklines=true,
    breakatwhitespace=true,
    breakautoindent=true,
    escapeinside={\%*}{*)},
    linewidth=\textwidth,
    basewidth=0.5em,
}

Using those settings, i.e. running pandoc with options:

pandoc test.md --listings -H listings-setup.tex -o output.pdf

the following result is obtained:

For reference, compare it with the result without --listings but with --highlight-syntax tango instead:

As you can see, the style is very close (except for numeric constants, which are blue using tango, but black using listings). Note also how the lines were wrapped with listings. I also added line numbers which refer to the original lines (before wrapping).

Answer 2

There is a really simple solution, using fvextra, recently suggested by jannick0.

Modify your YAML header options to include

\usepackage{fvextra}
\DefineVerbatimEnvironment{Highlighting}{Verbatim}{breaklines,commandchars=\\\{\}}

and compile with xelatex.

For instance,

---
header-includes:
 - \usepackage{fvextra}
 - \DefineVerbatimEnvironment{Highlighting}{Verbatim}{breaklines,commandchars=\\\{\}}
---

~~~~~{.java}
this is a very long long long long long long long long long long long long long line which is broken
~~~~~~

when compiled with

pandoc input.md -o output.pdf

gives

Answer 3

One does not need a listings-setup.tex in order for this to work.

Just add this to your YAML-Header include:

header-includes:
    - \usepackage{xcolor}
    - \lstset{breaklines=true}
    - \lstset{language=[Motorola68k]Assembler}
    - \lstset{basicstyle=\small\ttfamily}
    - \lstset{extendedchars=true}
    - \lstset{tabsize=2}
    - \lstset{columns=fixed}
    - \lstset{showstringspaces=false}
    - \lstset{frame=trbl}
    - \lstset{frameround=tttt}
    - \lstset{framesep=4pt}
    - \lstset{numbers=left}
    - \lstset{numberstyle=\tiny\ttfamily}
    - \lstset{postbreak=\raisebox{0ex}[0ex][0ex]{\ensuremath{\color{red}\hookrightarrow\space}}}

and compile with --listings. If there are compile errors try to add --pdf-engine=xelatex or --pdf-engine=lualatex.

Answer 4

This is only an apostille to Clement's (ihmo best) solution that cannot be illustrated in a comment. The highlight can be fixed in the YAML header. On the other hand, the lines of code can be numbered passing the option .numberLines to the code chunk:

Example stealing the code of JLDiaz's answer:

---

---

---
classoption: twocolumn # for more code line breaks
header-includes:
- \usepackage{fvextra}
- \DefineVerbatimEnvironment{Highlighting}{Verbatim}{breaklines,commandchars=\\\{\}}
output:
  pdf_document: 
    highlight: tango
---

# First section

Some code with long lines:


```{.java .numberLines}

public class Test {
    static Set<Thread> updateThreads = new HashSet<Thread>();

    public static void main(String[] args) {
        ConcurrentMap<Integer, String> concurrentMap = new ConcurrentHashMap<Integer, String>();
        for (int i = 0; i < 1000; i++) {
            startUpdateThread(i, concurrentMap);
        }
        for (Map.Entry<Integer, String> entry : concurrentMap.entrySet()) {
            System.out.println("Key :" + entry.getKey() + " Value:" + entry.getValue());
        }
        for (Thread thread : updateThreads) {
            thread.interrupt();
        }
    }
}

```
That's it.

Answer 5

If you only want to add line breaks and not modify the rest of the default code blocks layout, you can use hda's method in a simplified version, and modified to keep the original font spacing thanks to decrease monotype letter spacing in lstlisting.

TL;DR: Add this on top of your .md file:

---
header-includes:
    - \lstset{basicstyle=\small\ttfamily, basewidth=0.5em}
    - \lstset{breaklines=true}
---

and compile it with:

pandoc test.md -o test.pdf --listings