The Markdown Preprocessor is a Python module designed to add extended features on top of the excellent Markdown syntax defined by John Gruber. These additions are mainly focused on creating larger technical documents without needing to use something as heavy and syntactically complex as Docbook.
MarkdownPP uses a set of selectable modules to apply a series of transforms to the original document, with the end goal of generating a new Markdown document that contains sections or features that would be laborious to generate or maintain by hand.
Documents designed to be preprocessed by MarkdownPP should try to follow the convention of naming files with a .mdpp extension, so that MarkdownPP can generate a document with the same name, but with the standard .md extension. As an example, this document in raw format is named "readme.mdpp", and the generated document from MarkdownPP is named "readme.md" so that GitHub can find and process that document when viewing the repository.
1. Installation and Usage
2. Modules
2.1. Includes
2.2. IncludeURLs
2.3. IncludeCode
2.4. Table of Contents
2.5. Reference
2.6. LaTeX Rendering
2.7. YouTube Embeds
3. Examples
4. Support
5. References
Currently, you'll need to download the source code from GitHub or clone the repository, and the run the installation script manually.
pip install MarkdownPP
There are two components to the project: a Python module, MarkdownPP, and a
Python script that acts as a simple command line interface to the module,
markdown-pp.
Assuming you have a file named foo.mdpp, you can generate the preprocessed
file foo.md by running the following command:
$ markdown-pp foo.mdpp -o foo.md
If you do not specify an output file name, the results will be printed to stdout, enabling them to be piped to another command.
By default, all available modules are enabled. You can specify a list of modules to exclude:
$ markdown-pp foo.mdpp -o foo.md -e latexrender,youtubembed
To see usage instructions, including a list of enabled modules, supply the -h or --help arguments:
$ markdown-pp --help
In order to facilitate large documentation projects, MarkdownPP has an Include
module that will replace a line of the form !INCLUDE "path/to/filename" with
the contents of that file, recursively including other files as needed.
File foo.mdpp:
Hello
File bar.mdpp:
World!
File index.mdpp:
!INCLUDE "foo.mdpp"
!INCLUDE "bar.mdpp"
Compiling index.mdpp with the Include module will produce the following:
Hello
World!
Furthermore, the Include module supports the shifting of headers in the file to be included. For example,
File foo.mdpp:
# Foo
## Bar
File index.mdpp:
# Title
## Subtitle
!INCLUDE "foo.mdpp", 2
Compiling index.mdpp with the Include module and using 2 as shift
parameter will yield:
# Title
## Subtitle
### Foo
#### Bar
Facilitates the inclusion of remote files, such as files kept in a subversion
or GitHub repository. Like Include, the IncludeURL module can replace a line of
the form !INCLUDEURL "http://your.domain/path/to/filename" with the contents
returned from that url, recursively including additional remote urls as needed.
IncludeURL runs immediately after the Include module finishes executing. This means that is it possible to include local files that then require remote files, but impossible parse !INCLUDE statements found in remote files. This is prevent ambiguity as to where the file would be located.
Remote file http://your.domain/foo.mdpp:
Hello
Remote file http://your.domain/bar.mdpp:
Remote World!
Local file index.mdpp:
!INCLUDEURL "http://your.domain/foo.mdpp"
!INCLUDEURL "http://your.domain/bar.mdpp"
Compiling index.mdpp with the IncludeURL module will produce the following:
Hello
Remote World!
Facilitates the inclusion of local code files. GFM fences will be added around the included code.
Local code file hello.py:
def main():
print "Hello World"
if __name__ == '__main__':
main()
Local file index.mdpp:
# My Code
!INCLUDECODE "hello.py"
Easy as that!
Compiling index.mdpp with IncludeCode module wil produce the following:
# My Code
```
def main():
print "Hello World"
if __name__ == '__main__':
main()
```
Easy as that!
Furthermore the IncludeCode module supports line extraction and language specification. The line extraction is like python list slicing (e.g. 3:6; lines three to six). Please note that line counting starts at one, not at zero.
Local file index.mdpp:
# My Code
!INCLUDECODE "hello.py" (python), 1:2
Easy as that!
Compiling index.mdpp with IncludeCode module will produce the following:
# My Code
```python
def main():
print "Hello World"
```
Easy as that!
The biggest feature provided by MarkdownPP is the generation of a table of
contents for a document, with each item linked to the appropriate section of the
markup. The table is inserted into the document wherever the preprocessor finds
!TOC at the beginning of a line. Named <a> tags are inserted above each
Markdown header, and the headings are numbered hierarchically based on the
heading tag that Markdown would generate.
Similarly, MarkdownPP can generate a list of references that follow Markdown's
alternate link syntax, eg [name]: <url> "Title". A list of links will be
inserted wherever the preprocessor finds a line beginning with !REF. The
generated reference list follows the same alternate linking method to ensure
consistency in your document, but the link need not be referenced anywhere in
the document to be included in the list.
Lines and blocks of lines beginning and ending with $ are rendered as LaTeX, using QuickLaTeX.
For example,
$\displaystyle \int x^2 = \frac{x^3}{3} + C$
becomes
As GitHub-flavored Markdown does not allow embed tags, each line of the form
!VIDEO "[youtube url]" is converted into a screenshot that links to the video,
roughly simulating the look of an embedded video player.
For example,
!VIDEO "http://www.youtube.com/embed/7aEYoP5-duY"
becomes
Example file.mdpp:
# Document Title
!TOC
## Header 1
### Header 1.a
## Header 2
!REF
[github]: http://github.com "GitHub"
The preprocessor would generate the following Markdown-ready document file.md:
# Document Title
1\. [Header 1](#header1)
1.1\. [Header 1.a](#header1a)
2\. [Header 2](#header2)
<a name="header1"></a>
## Header 1
<a name="header1a"></a>
### Header 1.a
<a name="header2"></a>
## Header 2
* [GitHub][github]
[github]: http://github.com "GitHub"
If you find any problems with MarkdownPP, or have any feature requests, please report them to GitHub, and I will respond when possible. Code contributions are always welcome, and ideas for new modules, or additions to existing modules, are also appreciated.