root/trunk/lib/markdown.php

<?php

#
# Markdown  -  A text-to-HTML conversion tool for web writers
#
# Copyright (c) 2004-2005 John Gruber  
# <http://daringfireball.net/projects/markdown/>
#
# Copyright (c) 2004-2005 Michel Fortin - PHP Port  
# <http://www.michelf.com/projects/php-markdown/>
#


global    $MarkdownPHPVersion, $MarkdownSyntaxVersion,
        $md_empty_element_suffix, $md_tab_width,
        $md_nested_brackets_depth, $md_nested_brackets, 
        $md_escape_table, $md_backslash_escape_table, 
        $md_list_level;

$MarkdownPHPVersion    = '1.0.1c'; # Fri 9 Dec 2005
$MarkdownSyntaxVersion = '1.0.1';  # Sun 12 Dec 2004


#
# Global default settings:
#
$md_empty_element_suffix = " />";     # Change to ">" for HTML output
$md_tab_width = 4;

#
# WordPress settings:
#
$md_wp_posts    = true;  # Set to false to remove Markdown from posts.
$md_wp_comments = true;  # Set to false to remove Markdown from comments.


# -- WordPress Plugin Interface -----------------------------------------------
/*
Plugin Name: Markdown
Plugin URI: http://www.michelf.com/projects/php-markdown/
Description: <a href="http://daringfireball.net/projects/markdown/syntax">Markdown syntax</a> allows you to write using an easy-to-read, easy-to-write plain text format. Based on the original Perl version by <a href="http://daringfireball.net/">John Gruber</a>. <a href="http://www.michelf.com/projects/php-markdown/">More...</a>
Version: 1.0.1c
Author: Michel Fortin
Author URI: http://www.michelf.com/
*/
if (isset($wp_version)) {
    # More details about how it works here:
    # <http://www.michelf.com/weblog/2005/wordpress-text-flow-vs-markdown/>
    
    # Post content and excerpts
    if ($md_wp_posts) {
        remove_filter('the_content',  'wpautop');
        remove_filter('the_excerpt',  'wpautop');
        add_filter('the_content',     'Markdown', 6);
        add_filter('get_the_excerpt', 'Markdown', 6);
        add_filter('get_the_excerpt', 'trim', 7);
        add_filter('the_excerpt',     'md_add_p');
        add_filter('the_excerpt_rss', 'md_strip_p');
        
        remove_filter('content_save_pre',  'balanceTags', 50);
        remove_filter('excerpt_save_pre',  'balanceTags', 50);
        add_filter('the_content',        'balanceTags', 50);
        add_filter('get_the_excerpt', 'balanceTags', 9);
        
        function md_add_p($text) {
            if (strlen($text) == 0) return;
            if (strcasecmp(substr($text, -3), '<p>') == 0) return $text;
            return '<p>'.$text.'</p>';
        }
        function md_strip_p($t) { return preg_replace('{</?[pP]>}', '', $t); }
    }
    
    # Comments
    if ($md_wp_comments) {
        remove_filter('comment_text', 'wpautop');
        remove_filter('comment_text', 'make_clickable');
        add_filter('pre_comment_content', 'Markdown', 6);
        add_filter('pre_comment_content', 'md_hide_tags', 8);
        add_filter('pre_comment_content', 'md_show_tags', 12);
        add_filter('get_comment_text',    'Markdown', 6);
        add_filter('get_comment_excerpt', 'Markdown', 6);
        add_filter('get_comment_excerpt', 'md_strip_p', 7);
    
        global $md_hidden_tags;
        $md_hidden_tags = array(
            '<p>'    => md5('<p>'),        '</p>'    => md5('</p>'),
            '<pre>'    => md5('<pre>'),    '</pre>'=> md5('</pre>'),
            '<ol>'    => md5('<ol>'),        '</ol>'    => md5('</ol>'),
            '<ul>'    => md5('<ul>'),        '</ul>'    => md5('</ul>'),
            '<li>'    => md5('<li>'),        '</li>'    => md5('</li>'),
            );
        
        function md_hide_tags($text) {
            global $md_hidden_tags;
            return str_replace(array_keys($md_hidden_tags), 
                                array_values($md_hidden_tags), $text);
        }
        function md_show_tags($text) {
            global $md_hidden_tags;
            return str_replace(array_values($md_hidden_tags), 
                                array_keys($md_hidden_tags), $text);
        }
    }
}


# -- bBlog Plugin Info --------------------------------------------------------
function identify_modifier_markdown() {
    global $MarkdownPHPVersion;
    return array(
        'name'            => 'markdown',
        'type'            => 'modifier',
        'nicename'        => 'Markdown',
        'description'    => 'A text-to-HTML conversion tool for web writers',
        'authors'        => 'Michel Fortin and John Gruber',
        'licence'        => 'GPL',
        'version'        => $MarkdownPHPVersion,
        'help'            => '<a href="http://daringfireball.net/projects/markdown/syntax">Markdown syntax</a> allows you to write using an easy-to-read, easy-to-write plain text format. Based on the original Perl version by <a href="http://daringfireball.net/">John Gruber</a>. <a href="http://www.michelf.com/projects/php-markdown/">More...</a>'
    );
}

# -- Smarty Modifier Interface ------------------------------------------------
function smarty_modifier_markdown($text) {
    return Markdown($text);
}

# -- Textile Compatibility Mode -----------------------------------------------
# Rename this file to "classTextile.php" and it can replace Textile anywhere.
if (strcasecmp(substr(__FILE__, -16), "classTextile.php") == 0) {
    # Try to include PHP SmartyPants. Should be in the same directory.
    @include_once 'smartypants.php';
    # Fake Textile class. It calls Markdown instead.
    class Textile {
        function TextileThis($text, $lite='', $encode='', $noimage='', $strict='') {
            if ($lite == '' && $encode == '')   $text = Markdown($text);
            if (function_exists('SmartyPants')) $text = SmartyPants($text);
            return $text;
        }
    }
}



#
# Globals:
#

# Regex to match balanced [brackets].
# Needed to insert a maximum bracked depth while converting to PHP.
$md_nested_brackets_depth = 6;
$md_nested_brackets = 
    str_repeat('(?>[^\[\]]+|\[', $md_nested_brackets_depth).
    str_repeat('\])*', $md_nested_brackets_depth);

# Table of hash values for escaped characters:
$md_escape_table = array(
    "\\" => md5("\\"),
    "`" => md5("`"),
    "*" => md5("*"),
    "_" => md5("_"),
    "{" => md5("{"),
    "}" => md5("}"),
    "[" => md5("["),
    "]" => md5("]"),
    "(" => md5("("),
    ")" => md5(")"),
    ">" => md5(">"),
    "#" => md5("#"),
    "+" => md5("+"),
    "-" => md5("-"),
    "." => md5("."),
    "!" => md5("!")
);
# Create an identical table but for escaped characters.
$md_backslash_escape_table;
foreach ($md_escape_table as $key => $char)
    $md_backslash_escape_table["\\$key"] = $char;


function Markdown($text) {
#
# Main function. The order in which other subs are called here is
# essential. Link and image substitutions need to happen before
# _EscapeSpecialCharsWithinTagAttributes(), so that any *'s or _'s in the <a>
# and <img> tags get encoded.
#
    # Clear the global hashes. If we don't clear these, you get conflicts
    # from other articles when generating a page which contains more than
    # one article (e.g. an index page that shows the N most recent
    # articles):
    global $md_urls, $md_titles, $md_html_blocks;
    $md_urls = array();
    $md_titles = array();
    $md_html_blocks = array();

    # Standardize line endings:
    #   DOS to Unix and Mac to Unix
    $text = str_replace(array("\r\n", "\r"), "\n", $text);

    # Make sure $text ends with a couple of newlines:
    $text .= "\n\n";

    # Convert all tabs to spaces.
    $text = _Detab($text);

    # Strip any lines consisting only of spaces and tabs.
    # This makes subsequent regexen easier to write, because we can
    # match consecutive blank lines with /\n+/ instead of something
    # contorted like /[ \t]*\n+/ .
    $text = preg_replace('/^[ \t]+$/m', '', $text);

    # Turn block-level HTML blocks into hash entries
    $text = _HashHTMLBlocks($text);

    # Strip link definitions, store in hashes.
    $text = _StripLinkDefinitions($text);

    $text = _RunBlockGamut($text);

    $text = _UnescapeSpecialChars($text);

    return $text . "\n";
}


function _StripLinkDefinitions($text) {
#
# Strips link definitions from text, stores the URLs and titles in
# hash references.
#
    global $md_tab_width;
    $less_than_tab = $md_tab_width - 1;

    # Link defs are in the form: ^[id]: url "optional title"
    $text = preg_replace_callback('{
                        ^[ ]{0,'.$less_than_tab.'}\[(.+)\]:    # id = $1
                          [ \t]*
                          \n?                # maybe *one* newline
                          [ \t]*
                        <?(\S+?)>?            # url = $2
                          [ \t]*
                          \n?                # maybe one newline
                          [ \t]*
                        (?:
                            (?<=\s)            # lookbehind for whitespace
                            ["(]
                            (.+?)            # title = $3
                            [")]
                            [ \t]*
                        )?    # title is optional
                        (?:\n+|\Z)
        }xm',
        '_StripLinkDefinitions_callback',
        $text);
    return $text;
}
function _StripLinkDefinitions_callback($matches) {
    global $md_urls, $md_titles;
    $link_id = strtolower($matches[1]);
    $md_urls[$link_id] = _EncodeAmpsAndAngles($matches[2]);
    if (isset($matches[3]))
        $md_titles[$link_id] = str_replace('"', '&quot;', $matches[3]);
    return ''; # String that will replace the block
}


function _HashHTMLBlocks($text) {
    global $md_tab_width;
    $less_than_tab = $md_tab_width - 1;

    # Hashify HTML blocks:
    # We only want to do this for block-level HTML tags, such as headers,
    # lists, and tables. That's because we still want to wrap <p>s around
    # "paragraphs" that are wrapped in non-block-level tags, such as anchors,
    # phrase emphasis, and spans. The list of tags we're looking for is
    # hard-coded:
    $block_tags_a = 'p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|'.
                    'script|noscript|form|fieldset|iframe|math|ins|del';
    $block_tags_b = 'p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|'.
                    'script|noscript|form|fieldset|iframe|math';

    # First, look for nested blocks, e.g.:
    #     <div>
    #         <div>
    #         tags for inner block must be indented.
    #         </div>
    #     </div>
    #
    # The outermost tags must start at the left margin for this to match, and
    # the inner nested divs must be indented.
    # We need to do this before the next, more liberal match, because the next
    # match will start at the first `<div>` and stop at the first `</div>`.
    $text = preg_replace_callback("{
                (                        # save in $1
                    ^                    # start of line  (with /m)
                    <($block_tags_a)    # start tag = $2
                    \\b                    # word break
                    (.*\\n)*?            # any number of lines, minimally matching
                    </\\2>                # the matching end tag
                    [ \\t]*                # trailing spaces/tabs
                    (?=\\n+|\\Z)    # followed by a newline or end of document
                )
        }xm",
        '_HashHTMLBlocks_callback',
        $text);

    #
    # Now match more liberally, simply from `\n<tag>` to `</tag>\n`
    #
    $text = preg_replace_callback("{
                (                        # save in $1
                    ^                    # start of line  (with /m)
                    <($block_tags_b)    # start tag = $2
                    \\b                    # word break
                    (.*\\n)*?            # any number of lines, minimally matching
                    .*</\\2>                # the matching end tag
                    [ \\t]*                # trailing spaces/tabs
                    (?=\\n+|\\Z)    # followed by a newline or end of document
                )
        }xm",
        '_HashHTMLBlocks_callback',
        $text);

    # Special case just for <hr />. It was easier to make a special case than
    # to make the other regex more complicated.
    $text = preg_replace_callback('{
                (?:
                    (?<=\n\n)        # Starting after a blank line
                    |                # or
                    \A\n?            # the beginning of the doc
                )
                (                        # save in $1
                    [ ]{0,'.$less_than_tab.'}
                    <(hr)                # start tag = $2
                    \b                    # word break
                    ([^<>])*?            # 
                    /?>                    # the matching end tag
                    [ \t]*
                    (?=\n{2,}|\Z)        # followed by a blank line or end of document
                )
        }x',
        '_HashHTMLBlocks_callback',
        $text);

    # Special case for standalone HTML comments:
    $text = preg_replace_callback('{
                (?:
                    (?<=\n\n)        # Starting after a blank line
                    |                # or
                    \A\n?            # the beginning of the doc
                )
                (                        # save in $1
                    [ ]{0,'.$less_than_tab.'}
                    (?s:
                        <!
                        (--.*?--\s*)+
                        >
                    )
                    [ \t]*
                    (?=\n{2,}|\Z)        # followed by a blank line or end of document
                )
            }x',
            '_HashHTMLBlocks_callback',
            $text);

    return $text;
}
function _HashHTMLBlocks_callback($matches) {
    global $md_html_blocks;
    $text = $matches[1];
    $key = md5($text);
    $md_html_blocks[$key] = $text;
    return "\n\n$key\n\n"; # String that will replace the block
}


function _RunBlockGamut($text) {
#
# These are all the transformations that form block-level
# tags like paragraphs, headers, and list items.
#
    global $md_empty_element_suffix;

    $text = _DoHeaders($text);

    # Do Horizontal Rules:
    $text = preg_replace(
        array('{^[ ]{0,2}([ ]?\*[ ]?){3,}[ \t]*$}mx',
              '{^[ ]{0,2}([ ]? -[ ]?){3,}[ \t]*$}mx',
              '{^[ ]{0,2}([ ]? _[ ]?){3,}[ \t]*$}mx'),
        "\n<hr$md_empty_element_suffix\n", 
        $text);

    $text = _DoLists($text);
    $text = _DoCodeBlocks($text);
    $text = _DoBlockQuotes($text);

    # We already ran _HashHTMLBlocks() before, in Markdown(), but that
    # was to escape raw HTML in the original Markdown source. This time,
    # we're escaping the markup we've just created, so that we don't wrap
    # <p> tags around block-level tags.
    $text = _HashHTMLBlocks($text);
    $text = _FormParagraphs($text);

    return $text;
}


function _RunSpanGamut($text) {
#
# These are all the transformations that occur *within* block-level
# tags like paragraphs, headers, and list items.
#
    global $md_empty_element_suffix;

    $text = _DoCodeSpans($text);

    $text = _EscapeSpecialChars($text);

    # Process anchor and image tags. Images must come first,
    # because ![foo][f] looks like an anchor.
    $text = _DoImages($text);
    $text = _DoAnchors($text);

    # Make links out of things like `<http://example.com/>`
    # Must come after _DoAnchors(), because you can use < and >
    # delimiters in inline links like [this](<url>).
    $text = _DoAutoLinks($text);
    $text = _EncodeAmpsAndAngles($text);
    $text = _DoItalicsAndBold($text);

    # Do hard breaks:
    $text = preg_replace('/ {2,}\n/', "<br$md_empty_element_suffix\n", $text);

    return $text;
}


function _EscapeSpecialChars($text) {
    global $md_escape_table;
    $tokens = _TokenizeHTML($text);

    $text = '';   # rebuild $text from the tokens
#    $in_pre = 0;  # Keep track of when we're inside <pre> or <code> tags.
#    $tags_to_skip = "!<(/?)(?:pre|code|kbd|script|math)[\s>]!";

    foreach ($tokens as $cur_token) {
        if ($cur_token[0] == 'tag') {
            # Within tags, encode * and _ so they don't conflict
            # with their use in Markdown for italics and strong.
            # We're replacing each such character with its
            # corresponding MD5 checksum value; this is likely
            # overkill, but it should prevent us from colliding
            # with the escape values by accident.
            $cur_token[1] = str_replace(array('*', '_'),
                array($md_escape_table['*'], $md_escape_table['_']),
                $cur_token[1]);
            $text .= $cur_token[1];
        } else {
            $t = $cur_token[1];
            $t = _EncodeBackslashEscapes($t);
            $text .= $t;
        }
    }
    return $text;
}


function _DoAnchors($text) {
#
# Turn Markdown link shortcuts into XHTML <a> tags.
#
    global $md_nested_brackets;
    #
    # First, handle reference-style links: [link text] [id]
    #
    $text = preg_replace_callback("{
        (                    # wrap whole match in $1
          \\[
            ($md_nested_brackets)    # link text = $2
          \\]

          [ ]?                # one optional space
          (?:\\n[ ]*)?        # one optional newline followed by spaces

          \\[
            (.*?)        # id = $3
          \\]
        )
        }xs",
        '_DoAnchors_reference_callback', $text);

    #
    # Next, inline-style links: [link text](url "optional title")
    #
    $text = preg_replace_callback("{
        (                # wrap whole match in $1
          \\[
            ($md_nested_brackets)    # link text = $2
          \\]
          \\(            # literal paren
            [ \\t]*
            <?(.*?)>?    # href = $3
            [ \\t]*
            (            # $4
              (['\"])    # quote char = $5
              (.*?)        # Title = $6
              \\5        # matching quote
            )?            # title is optional
          \\)
        )
        }xs",
        '_DoAnchors_inline_callback', $text);

    return $text;
}
function _DoAnchors_reference_callback($matches) {
    global $md_urls, $md_titles, $md_escape_table;
    $whole_match = $matches[1];
    $link_text   = $matches[2];
    $link_id     = strtolower($matches[3]);

    if ($link_id == "") {
        $link_id = strtolower($link_text); # for shortcut links like [this][].
    }

    if (isset($md_urls[$link_id])) {
        $url = $md_urls[$link_id];
        # We've got to encode these to avoid conflicting with italics/bold.
        $url = str_replace(array('*', '_'),
                           array($md_escape_table['*'], $md_escape_table['_']),
                           $url);
        $result = "<a href=\"$url\"";
        if ( isset( $md_titles[$link_id] ) ) {
            $title = $md_titles[$link_id];
            $title = str_replace(array('*',     '_'),
                                 array($md_escape_table['*'], 
                                       $md_escape_table['_']), $title);
            $result .=  " title=\"$title\"";
        }
        $result .= ">$link_text</a>";
    }
    else {
        $result = $whole_match;
    }
    return $result;
}
function _DoAnchors_inline_callback($matches) {
    global $md_escape_table;
    $whole_match    = $matches[1];
    $link_text        = $matches[2];
    $url            = $matches[3];
    $title            =& $matches[6];

    # We've got to encode these to avoid conflicting with italics/bold.
    $url = str_replace(array('*', '_'),
                       array($md_escape_table['*'], $md_escape_table['_']), 
                       $url);
    $result