A Pandoc filter for including code from source files.
You get to:
- Keep your examples and documentation compiled and in sync
- Include small snippets from larger source files without needing to keep track of line numbers
- Dedent included snippets
The filter recognizes code blocks with the include
attribute present. It
swaps the content of the code block with contents from a file.
The simplest way to use this filter is to include an entire file:
```{include=docs/MyFile.hs}
```
You can still use other attributes, and classes, to control the code blocks:
```{.purescript include=docs/MyFile.purs}
```
There is support for delimited snippets. Use a line comment of
whatever kind you want, and enclose the snippet between start snippet <name>
and end snippet <name>
.
-- start snippet cool-thingy
main =
putStrLn "I explain some cool concept in Haskell code."
-- end snippet cool-thingy
Or why not some C code:
// start snippet wow
int main() {
printf("such performance");
}
// end snippet wow
NOTE: There can only be whitespace and a newline after the snippet name. This means that multi-line comments in C, Java, etc, will not work. Only single-line comments will.
Then, in your code block, specify the snippet name:
```{include=docs/MyFile.hs snippet=cool-thingy}
```
If you want to include a specific range of lines, use startLine
and endLine
:
```{include=docs/MyFile.hs startLine=35 endLine=80}
```
Using the dedent
attribute, you can have whitespaces removed on each line,
where possible (non-whitespace character will not be removed even if they occur
in the dedent area).
```{include=docs/MyFile.hs dedent=4}
```
"Snippet mode" and "range mode" cannot be used together.
If you include the numberLines
class in your code block, and use include
,
the startFrom
attribute will be added with respect to the included code's
location in the source file.
```{include=docs/MyFile.hs startLine=35 endLine=80 .numberLines}
```
It is possible to add hyperlinks to the original source code file specified in the include
attribute by adding the .includeLink
class in your code block.
```{include=docs/MyFile.hs .includeLink}
```
A base url will be appended to all relative paths specified in the
include
attribute of each CodeBlock
. It does not affect paths beginning with file:
, C:
, \
, /
,.... This can be done
two option:
Specify a base key along with the base url as the attribute in the YAML header:
---
title: All About Wonderland
author: Alice
date: November 2020
base: http://localhost:8000/
---
Add the base as a metavalue -M base=<base url>
or
--metavalue base=<base url>
in the command line when
calling pandoc.
pandoc --filter pandoc-include-code -M base=http://localhost:8000/ in.md -o out.html
NOTE: If the base url is specified in the metadata block, then by specifying a different base in the command line, it will override the original base.
Both of these options will add a hyperlink to the filepath defined
in the include
attribute linking to
http://localhost:8000/source/sample.hs
:
Adding a base attribute in the metadata block or the command line
will affect all relative links in the CodeBlocks
. To add an alternative base for a
specific link, add the base as an attribute base=https....
to
theCodeBlock
:
```{.haskell .includeLink include=source/sample.hs snippet=animals base=<path>}
```
This adds a hyperlink to the filepath specified in the
include
attribute linking to ../source/sample.hs
:
- The blog post Automating the Build of Your Technical Presentation shows practical examples of how to use this filter.
Executables for Linux and macOS are available in the Releases page.
You can use Homebrew to install this filter:
brew install pandoc-include-code
If you'd rather install using cabal
or stack
, you can use the following
command:
cabal install pandoc-include-code
The package is available at Hackage.
Requirements:
To install from sources, run:
git clone git@github.com:owickstrom/pandoc-include-code.git
cd pandoc-include-code
cabal configure
cabal install
If you have installed from sources, and you have ~/.local/bin
on your
PATH
, you can use the filter with Pandoc like so:
pandoc --filter pandoc-include-code input.md output.html
If you are using the Hakyll static site generator, you can use the filter by importing it as a library and using the snippet below.
Add pandoc
, pandoc-types
, and pandoc-include-code
to your project dependencies, then define a custom Hakyll compiler using a Pandoc transform:
import Text.Pandoc (Format (..), Pandoc)
import Text.Pandoc.Walk (walkM)
import Text.Pandoc.Filter.IncludeCode (includeCode)
includeCodeTransform :: Pandoc -> IO Pandoc
includeCodeTransform = walkM (includeCode (Just (Format "html5")))
includeCodePandocCompiler :: Compiler (Item String)
includeCodePandocCompiler =
pandocCompilerWithTransformM
defaultHakyllReaderOptions
defaultHakyllWriterOptions
(unsafeCompiler . includeCodeTransform)
You can now use includeCodePandocCompiler
instead of the default pandocCompiler
in your Hakyll rules:
match "*.md" $ do
route $ setExtension "html"
compile $ includeCodePandocCompiler
>>= loadAndApplyTemplate "templates/default.html" defaultContext
>>= relativizeUrls