Options

MultiMarkdown supports a number of options to its processor which control the behaviour of the output document.

These options can be supplied to the constructor, on in a hash with the individual calls to the markdown method. See the synopsis for examples of both of the above styles.

The options for the processor are:

bibliography_title

The title of the generated bibliography, defaults to 'Bibliography'.

disable_bibliography

If true, this disables the MultiMarkdown bibliography/citation handling.

disable_definition_lists

If true, this disables the MultiMarkdown definition list handling.

disable_footnotes

If true, this disables the MultiMarkdown footnotes handling.

disable_tables

If true, this disables the MultiMarkdown table handling.

empty_element_suffix

This option can be used to generate normal HTML output. By default, it is "/>", which is xHTML, change to ">" for normal HTML.

heading_ids

Controls if "hX" tags generated have an id attribute. Defaults to true. Turn off for compatibility with the original markdown.

heading_ids_spaces_to_dash

Controls whether spaces in headings should be rendered as "-" characters in the heading ids (for compatibility with GitHub markdown, and others)

img_ids

Controls if "img" tags generated have an id attribute. Defaults to true. Turn off for compatibility with the original markdown.

strip_metadata

If true, any metadata in the input document is removed from the output document (note - does not take effect in complete document format).

tab_width

Controls indent width in the generated markup, defaults to 4

transliterated_ids

In markdown label values, change accented and other non-ASCII letter characters with Text::Unidecode. If that module is not available, this issues a warning and does nothing. When "unicode_ids" is true, this is ignored. The default is false.

unicode_ids

In markdown label values, allow any Unicode letter character along with the allowed ASCII symbol characters. This overrules "transliterated_ids" when true. The default is false.

use_metadata

Controls the metadata options below.

Metadata

MultiMarkdown supports the concept of 'metadata', which allows you to specify a number of formatting options within the document itself. Metadata should be placed in the top few lines of a file, on value per line as colon separated key/value pairs. The metadata should be separated from the document with a blank line.

Most metadata keys are also supported as options to the constructor, or options to the markdown method itself. (Note, as metadata, keys contain space, whereas options the keys are underscore separated.)

You can attach arbitrary metadata to a document, which is output in HTML "<META>" tags if unknown, see t/11-document_format.t for an example.

These are the known metadata keys:

document_format

If set to 'complete', MultiMarkdown will render an entire xHTML page, otherwise it will render a document fragment

use wikilinks

If set to '1' or 'on', causes links that are WikiWords to automatically be processed into links.

Class methods

new

A simple constructor, see the SYNTAX and OPTIONS sections for more information.

Instance methods

markdown( MARKDOWN_TEXT [, HASHREF] )

This is the legacy interface to this module, but it does too much and is a poor name. For the function form, use "multimarkdown_to_html" instead. At the moment that's just a wrapper for "markdown" in the functional form. For the object-oriented forms, use "to_html" instead. That's also just a wrapper for this, but will later change to enforce object-orientedness (i.e. exclude the functional form).

And now the legacy stuff.

This works as either a class method, instance method, or exportable function:

        my $html = Text::MultiMarkdown->markdown( $text );
        my $mm = Text::MultiMarkdown->new;
        my $html = $mm->markdown($text);
        use Text::MultiMarkdown qw(markdown);
        my $html = markdown( $text );
    

Any of these forms take an optional HASH_REF argument for options. These are the options for this module or the parent class Text::Markdown:

        my $html = Text::MultiMarkdown->markdown( $text, { ... } );
        my $mm = Text::MultiMarkdown->new;
        my $html = $mm->markdown($text, { ... });
        use Text::MultiMarkdown qw(markdown);
        my $html = markdown( $text, { ... } );
    

To make this work in all these cases, since this was the legacy design, various unsavory things have to happen.

When called as a class method, a new object is constructed. We guess that it's a class method by looking at the first argument and seeing that it looks like a Perl package name. In prior versions this was documented to not work, but there was also a TODO test for it to work. So, now it works. This might fail if the entire markdown text is exactly a valid Perl package name.

If the first argument is a blessed reference, we guess that this is an instance method. With the optional HASH_REF argument this constructs a new argument with all of the settings of the original object and the stuff in HASH_REF. This might fail if you have some weird case where you call this as a function but pass as the TEXT argument an object that has overloaded stringification .

There are these situations:

        CLASS->markdown( TEXT );
        CLASS->markdown( TEXT, HASHREF );
        OBJ->markdown( TEXT );
        OBJ->markdown( TEXT, HASHREF );
        markdown( TEXT );
        markdown( TEXT, HASHREF );
    

These are really:

        markdown( CLASS, TEXT )
        markdown( CLASS, TEXT, HASHREF )
        markdown( OBJ, TEXT )
        markdown( OBJ, TEXT, HASHREF )
        markdown( TEXT );
        markdown( TEXT, HASHREF );
    

Which breaks down to these groups:

        1) markdown( TEXT );
        2.1) markdown( TEXT, HASHREF );
        2.2) markdown( CLASS, TEXT )
        2.3) markdown( OBJ, TEXT )
        3.1) markdown( CLASS, TEXT, HASHREF )
        3.2) markdown( OBJ, TEXT, HASHREF )
    

In 1), 2.2), and 3.1), we should make a new object and then do our thing.

In 3.1), the previous version specifically said that we can't call this as a class method.

In 3.2), we need to merge the options in the existing object with the new options. This was never a documented feature though.

Part of the tickyness is that interface for Text::Markdown. We need to pass the HASHREF to _CleanUpRunData in the SUPER class

multimarkdown_to_html

For the functional interface, you should use this instead of "markdown" because it's a better name. At the moment it's the same as calling "markdown", but eventually this will diverge from the object-oriented form "to_html", which is also a better name.

to_html

As a class or instance method, you should use this instead of "markdown" because it's a better name. At the moment it's the same as calling "markdown", but eventually this will diverge from the functional form "multimarkdown_to_html", which is also a better name.