Friendica Communications Platform (please note that this is a clone of the repository at github, issues are handled there) https://friendi.ca
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

3043 lines
84 KiB

11 years ago
<?php
#
# Markdown Extra - A text-to-HTML conversion tool for web writers
#
# PHP Markdown & Extra
# Copyright (c) 2004-2008 Michel Fortin
# <http://www.michelf.com/projects/php-markdown/>
#
# Original Markdown
# Copyright (c) 2004-2006 John Gruber
# <http://daringfireball.net/projects/markdown/>
#
define( 'MARKDOWN_VERSION', "1.0.1m" ); # Sat 21 Jun 2008
define( 'MARKDOWNEXTRA_VERSION', "1.2.3" ); # Wed 31 Dec 2008
#
# Global default settings:
#
# Change to ">" for HTML output
@define( 'MARKDOWN_EMPTY_ELEMENT_SUFFIX', " />");
# Define the width of a tab for code blocks.
@define( 'MARKDOWN_TAB_WIDTH', 4 );
# Optional title attribute for footnote links and backlinks.
@define( 'MARKDOWN_FN_LINK_TITLE', "" );
@define( 'MARKDOWN_FN_BACKLINK_TITLE', "" );
# Optional class attribute for footnote links and backlinks.
@define( 'MARKDOWN_FN_LINK_CLASS', "" );
@define( 'MARKDOWN_FN_BACKLINK_CLASS', "" );
# Enables special handling for links pointing outside of the current domain.
@define( 'MARKDOWN_EL_ENABLE', true); # Use this feature at all?
@define( 'MARKDOWN_EL_LOCAL_DOMAIN', null); # Leave as null to autodetect
@define( 'MARKDOWN_EL_NEW_WINDOW', true); # Open link in a new browser?
@define( 'MARKDOWN_EL_CSS_CLASS', 'external'); # Leave as null for no class
# Enables header auto-self-linking.
@define( 'MARKDOWN_HA_ENABLE', true ); # Use this feature at all?
@define( 'MARKDOWN_HA_CLASS', 'hidden-selflink' ); # Leave as null for no class
@define( 'MARKDOWN_HA_TEXT', '&larr;' ); # The text to use as the link
#
# WordPress settings:
#
# Change to false to remove Markdown from posts and/or comments.
@define( 'MARKDOWN_WP_POSTS', true );
@define( 'MARKDOWN_WP_COMMENTS', true );
### Standard Function Interface ###
@define( 'MARKDOWN_PARSER_CLASS', 'MarkdownExtra_Parser' );
function Markdown($text) {
#
# Initialize the parser and return the result of its transform method.
#
# Setup static parser variable.
static $parser;
if (!isset($parser)) {
$parser_class = MARKDOWN_PARSER_CLASS;
$parser = new $parser_class;
}
# Transform text using parser.
return $parser->transform($text);
}
### WordPress Plugin Interface ###
/*
Plugin Name: Markdown Extra
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: 1.2.2
Author: Michel Fortin
Author URI: http://www.michelf.com/
*/
if (isset($wp_version)) {
# More details about how it works here:
# <http://www.michelf.com/weblog/2005/wordpress-text-flow-vs-markdown/>
# Post content and excerpts
# - Remove WordPress paragraph generator.
# - Run Markdown on excerpt, then remove all tags.
# - Add paragraph tag around the excerpt, but remove it for the excerpt rss.
if (MARKDOWN_WP_POSTS) {
remove_filter('the_content', 'wpautop');
remove_filter('the_content_rss', 'wpautop');
remove_filter('the_excerpt', 'wpautop');
add_filter('the_content', 'mdwp_MarkdownPost', 6);
add_filter('the_content_rss', 'mdwp_MarkdownPost', 6);
add_filter('get_the_excerpt', 'mdwp_MarkdownPost', 6);
add_filter('get_the_excerpt', 'trim', 7);
add_filter('the_excerpt', 'mdwp_add_p');
add_filter('the_excerpt_rss', 'mdwp_strip_p');
remove_filter('content_save_pre', 'balanceTags', 50);
remove_filter('excerpt_save_pre', 'balanceTags', 50);
add_filter('the_content', 'balanceTags', 50);
add_filter('get_the_excerpt', 'balanceTags', 9);
}
# Add a footnote id prefix to posts when inside a loop.
function mdwp_MarkdownPost($text) {
static $parser;
if (!$parser) {
$parser_class = MARKDOWN_PARSER_CLASS;
$parser = new $parser_class;
}
if (is_single() || is_page() || is_feed()) {
$parser->fn_id_prefix = "";
} else {
$parser->fn_id_prefix = get_the_ID() . ".";
}
return $parser->transform($text);
}
# Comments
# - Remove WordPress paragraph generator.
# - Remove WordPress auto-link generator.
# - Scramble important tags before passing them to the kses filter.
# - Run Markdown on excerpt then remove paragraph tags.
if (MARKDOWN_WP_COMMENTS) {
remove_filter('comment_text', 'wpautop', 30);
remove_filter('comment_text', 'make_clickable');
add_filter('pre_comment_content', 'Markdown', 6);
add_filter('pre_comment_content', 'mdwp_hide_tags', 8);
add_filter('pre_comment_content', 'mdwp_show_tags', 12);
add_filter('get_comment_text', 'Markdown', 6);
add_filter('get_comment_excerpt', 'Markdown', 6);
add_filter('get_comment_excerpt', 'mdwp_strip_p', 7);
global $mdwp_hidden_tags, $mdwp_placeholders;
$mdwp_hidden_tags = explode(' ',
'<p> </p> <pre> </pre> <ol> </ol> <ul> </ul> <li> </li>');
$mdwp_placeholders = explode(' ', str_rot13(
'pEj07ZbbBZ U1kqgh4w4p pre2zmeN6K QTi31t9pre ol0MP1jzJR '.
'ML5IjmbRol ulANi1NsGY J7zRLJqPul liA8ctl16T K9nhooUHli'));
}
function mdwp_add_p($text) {
if (!preg_match('{^$|^<(p|ul|ol|dl|pre|blockquote)>}i', $text)) {
$text = '<p>'.$text.'</p>';
$text = preg_replace('{\n{2,}}', "</p>\n\n<p>", $text);
}
return $text;
}
function mdwp_strip_p($t) { return preg_replace('{</?p>}i', '', $t); }
function mdwp_hide_tags($text) {
global $mdwp_hidden_tags, $mdwp_placeholders;
return str_replace($mdwp_hidden_tags, $mdwp_placeholders, $text);
}
function mdwp_show_tags($text) {
global $mdwp_hidden_tags, $mdwp_placeholders;
return str_replace($mdwp_placeholders, $mdwp_hidden_tags, $text);
}
}
### bBlog Plugin Info ###
function identify_modifier_markdown() {
return array(
'name' => 'markdown',
'type' => 'modifier',
'nicename' => 'PHP Markdown Extra',
'description' => 'A text-to-HTML conversion tool for web writers',
'authors' => 'Michel Fortin and John Gruber',
'licence' => 'GPL',
'version' => MARKDOWNEXTRA_VERSION,
'help' => '<a href="http://daringfireball.net/projects/markdown/syntax">Markdown syntax</a> allows you to write using an easy-to-read, easy-to-write plain text format. Based on the original Perl version by <a href="http://daringfireball.net/">John Gruber</a>. <a href="http://www.michelf.com/projects/php-markdown/">More...</a>',
);
}
### Smarty Modifier Interface ###
function smarty_modifier_markdown($text) {
return Markdown($text);
}
### Textile Compatibility Mode ###
# Rename this file to "classTextile.php" and it can replace Textile everywhere.
if (strcasecmp(substr(__FILE__, -16), "classTextile.php") == 0) {
# Try to include PHP SmartyPants. Should be in the same directory.
@include_once 'smartypants.php';
# Fake Textile class. It calls Markdown instead.
class Textile {
function TextileThis($text, $lite='', $encode='') {
if ($lite == '' && $encode == '') $text = Markdown($text);
if (function_exists('SmartyPants')) $text = SmartyPants($text);
return $text;
}
# Fake restricted version: restrictions are not supported for now.
function TextileRestricted($text, $lite='', $noimage='') {
return $this->TextileThis($text, $lite);
}
# Workaround to ensure compatibility with TextPattern 4.0.3.
function blockLite($text) { return $text; }
}
}
#
# Markdown Parser Class
#
class Markdown_Parser {
# Regex to match balanced [brackets].
# Needed to insert a maximum bracked depth while converting to PHP.
var $nested_brackets_depth = 6;
var $nested_brackets_re;
var $nested_url_parenthesis_depth = 4;
var $nested_url_parenthesis_re;
# Table of hash values for escaped characters:
var $escape_chars = '\`*_{}[]()>#+-.!';
var $escape_chars_re;
# Change to ">" for HTML output.
var $empty_element_suffix = MARKDOWN_EMPTY_ELEMENT_SUFFIX;
var $tab_width = MARKDOWN_TAB_WIDTH;
# Change to `true` to disallow markup or entities.
var $no_markup = false;
var $no_entities = false;
# Predefined urls and titles for reference links and images.
var $predef_urls = array();
var $predef_titles = array();
function Markdown_Parser() {
#
# Constructor function. Initialize appropriate member variables.
#
$this->_initDetab();
$this->prepareItalicsAndBold();
$this->nested_brackets_re =
str_repeat('(?>[^\[\]]+|\[', $this->nested_brackets_depth).
str_repeat('\])*', $this->nested_brackets_depth);
$this->nested_url_parenthesis_re =
str_repeat('(?>[^()\s]+|\(', $this->nested_url_parenthesis_depth).
str_repeat('(?>\)))*', $this->nested_url_parenthesis_depth);
$this->escape_chars_re = '['.preg_quote($this->escape_chars).']';
# Sort document, block, and span gamut in ascendent priority order.
asort($this->document_gamut);
asort($this->block_gamut);
asort($this->span_gamut);
}
# Internal hashes used during transformation.
var $urls = array();
var $titles = array();
var $html_hashes = array();
# Status flag to avoid invalid nesting.
var $in_anchor = false;
function setup() {
#
# Called before the transformation process starts to setup parser
# states.
#
# Clear global hashes.
$this->urls = $this->predef_urls;
$this->titles = $this->predef_titles;
$this->html_hashes = array();
$in_anchor = false;
}
function teardown() {
#
# Called after the transformation process to clear any variable
# which may be taking up memory unnecessarly.
#
$this->urls = array();
$this->titles = array();
$this->html_hashes = array();
}
function transform($text) {
#
# Main function. Performs some preprocessing on the input text
# and pass it through the document gamut.
#
$this->setup();
# Remove UTF-8 BOM and marker character in input, if present.
$text = preg_replace('{^\xEF\xBB\xBF|\x1A}', '', $text);
# Standardize line endings:
# DOS to Unix and Mac to Unix
$text = preg_replace('{\r\n?}', "\n", $text);
# Make sure $text ends with a couple of newlines:
$text .= "\n\n";
# Convert all tabs to spaces.
$text = $this->detab($text);
# Turn block-level HTML blocks into hash entries
$text = $this->hashHTMLBlocks($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 /[ ]*\n+/ .
$text = preg_replace('/^[ ]+$/m', '', $text);
# Run document gamut methods.
foreach ($this->document_gamut as $method => $priority) {
$text = $this->$method($text);
}
$this->teardown();
return $text . "\n";
}
var $document_gamut = array(
# Strip link definitions, store in hashes.
"stripLinkDefinitions" => 20,
"runBasicBlockGamut" => 30,
);
function stripLinkDefinitions($text) {
#
# Strips link definitions from text, stores the URLs and titles in
# hash references.
#
$less_than_tab = $this->tab_width - 1;
# Link defs are in the form: ^[id]: url "optional title"
$text = preg_replace_callback('{
^[ ]{0,'.$less_than_tab.'}\[(.+)\][ ]?: # id = $1
[ ]*
\n? # maybe *one* newline
[ ]*
<?(\S+?)>? # url = $2
[ ]*
\n? # maybe one newline
[ ]*
(?:
(?<=\s) # lookbehind for whitespace
["(]
(.*?) # title = $3
[")]
[ ]*
)? # title is optional
(?:\n+|\Z)
}xm',
array(&$this, '_stripLinkDefinitions_callback'),
$text);
return $text;
}
function _stripLinkDefinitions_callback($matches) {
$link_id = strtolower($matches[1]);
$this->urls[$link_id] = $matches[2];
$this->titles[$link_id] =& $matches[3];
return ''; # String that will replace the block
}
function hashHTMLBlocks($text) {
if ($this->no_markup) return $text;
$less_than_tab = $this->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:
#
# * List "a" is made of tags which can be both inline or block-level.
# These will be treated block-level when the start tag is alone on
# its line, otherwise they're not matched here and will be taken as
# inline later.
# * List "b" is made of tags which are always block-level;
#
$block_tags_a_re = 'ins|del';
$block_tags_b_re = 'p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|address|'.
'script|noscript|form|fieldset|iframe|math|textarea';
# Regular expression for the content of a block tag.
$nested_tags_level = 4;
$attr = '
(?> # optional tag attributes
\s # starts with whitespace
(?>
[^>"/]+ # text outside quotes
|
/+(?!>) # slash not followed by ">"
|
"[^"]*" # text inside double quotes (tolerate ">")
|
\'[^\']*\' # text inside single quotes (tolerate ">")
)*
)?
';
$content =
str_repeat('
(?>
[^<]+ # content without tag
|
<\2 # nested opening tag
'.$attr.' # attributes
(?>
/>
|
>', $nested_tags_level). # end of opening tag
'.*?'. # last level nested tag content
str_repeat('
</\2\s*> # closing nested tag
)
|
<(?!/\2\s*> # other tags with a different name
)
)*',
$nested_tags_level);
$content2 = str_replace('\2', '\3', $content);
# 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('{(?>
(?>
(?<=\n\n) # Starting after a blank line
| # or
\A\n? # the beginning of the doc
)
( # save in $1
# Match from `\n<tag>` to `</tag>\n`, handling nested tags
# in between.
[ ]{0,'.$less_than_tab.'}
<('.$block_tags_b_re.')# start tag = $2
'.$attr.'> # attributes followed by > and \n
'.$content.' # content, support nesting
</\2> # the matching end tag
[ ]* # trailing spaces/tabs
(?=\n+|\Z) # followed by a newline or end of document
| # Special version for tags of group a.
[ ]{0,'.$less_than_tab.'}
<('.$block_tags_a_re.')# start tag = $3
'.$attr.'>[ ]*\n # attributes followed by >
'.$content2.' # content, support nesting
</\3> # the matching end tag
[ ]* # trailing spaces/tabs
(?=\n+|\Z) # followed by a newline or end of document
| # Special case just for <hr />. It was easier to make a special
# case than to make the other regex more complicated.
[ ]{0,'.$less_than_tab.'}
<(hr) # start tag = $2
'.$attr.' # attributes
/?> # the matching end tag
[ ]*
(?=\n{2,}|\Z) # followed by a blank line or end of document
| # Special case for standalone HTML comments:
[ ]{0,'.$less_than_tab.'}
(?s:
<!-- .*? -->
)
[ ]*
(?=\n{2,}|\Z) # followed by a blank line or end of document
| # PHP and ASP-style processor instructions (<? and <%)
[ ]{0,'.$less_than_tab.'}
(?s:
<([?%]) # $2
.*?
\2>
)
[ ]*
(?=\n{2,}|\Z) # followed by a blank line or end of document
)
)}Sxmi',
array(&$this, '_hashHTMLBlocks_callback'),
$text);
return $text;
}
function _hashHTMLBlocks_callback($matches) {
$text = $matches[1];
$key = $this->hashBlock($text);
return "\n\n$key\n\n";
}
function hashPart($text, $boundary = 'X') {
#
# Called whenever a tag must be hashed when a function insert an atomic
# element in the text stream. Passing $text to through this function gives
# a unique text-token which will be reverted back when calling unhash.
#
# The $boundary argument specify what character should be used to surround
# the token. By convension, "B" is used for block elements that needs not
# to be wrapped into paragraph tags at the end, ":" is used for elements
# that are word separators and "X" is used in the general case.
#
# Swap back any tag hash found in $text so we do not have to `unhash`
# multiple times at the end.
$text = $this->unhash($text);
# Then hash the block.
static $i = 0;
$key = "$boundary\x1A" . ++$i . $boundary;
$this->html_hashes[$key] = $text;
return $key; # String that will replace the tag.
}
function hashBlock($text) {
#
# Shortcut function for hashPart with block-level boundaries.
#
return $this->hashPart($text, 'B');
}
var $block_gamut = array(
#
# These are all the transformations that form block-level
# tags like paragraphs, headers, and list items.
#
"doHeaders" => 10,
"doHorizontalRules" => 20,
"doLists" => 40,
"doCodeBlocks" => 50,
"doBlockQuotes" => 60,
);
function runBlockGamut($text) {
#
# Run block gamut tranformations.
#
# We need to escape raw HTML in Markdown source before doing anything
# else. This need to be done for each block, and not only at the
# begining in the Markdown function since hashed blocks can be part of
# list items and could have been indented. Indented blocks would have
# been seen as a code block in a previous pass of hashHTMLBlocks.
$text = $this->hashHTMLBlocks($text);
return $this->runBasicBlockGamut($text);
}
function runBasicBlockGamut($text) {
#
# Run block gamut tranformations, without hashing HTML blocks. This is
# useful when HTML blocks are known to be already hashed, like in the first
# whole-document pass.
#
foreach ($this->block_gamut as $method => $priority) {
$text = $this->$method($text);
}
# Finally form paragraph and restore hashed blocks.
$text = $this->formParagraphs($text);
return $text;
}
function doHorizontalRules($text) {
# Do Horizontal Rules:
return preg_replace(
'{
^[ ]{0,3} # Leading space
([-*_]) # $1: First marker
(?> # Repeated marker group
[ ]{0,2} # Zero, one, or two spaces.
\1 # Marker character
){2,} # Group repeated at least twice
[ ]* # Tailing spaces
$ # End of line.
}mx',
"\n".$this->hashBlock("<hr$this->empty_element_suffix")."\n",
$text);
}
var $span_gamut = array(
#
# These are all the transformations that occur *within* block-level
# tags like paragraphs, headers, and list items.
#
# Process character escapes, code spans, and inline HTML
# in one shot.
"parseSpan" => -30,
# Process anchor and image tags. Images must come first,
# because ![foo][f] looks like an anchor.
"doImages" => 10,
"doAnchors" => 20,
# Make links out of things like `<http://example.com/>`
# Must come after doAnchors, because you can use < and >
# delimiters in inline links like [this](<url>).
"doAutoLinks" => 30,
"encodeAmpsAndAngles" => 40,
"doItalicsAndBold" => 50,
"doHardBreaks" => 60,
);
function runSpanGamut($text) {
#
# Run span gamut tranformations.
#
foreach ($this->span_gamut as $method => $priority) {
$text = $this->$method($text);
}
return $text;
}
function doHardBreaks($text) {
# Do hard breaks:
return preg_replace_callback('/ {2,}\n/',
array(&$this, '_doHardBreaks_callback'), $text);
}
function _doHardBreaks_callback($matches) {
return $this->hashPart("<br$this->empty_element_suffix\n");
}
function doAnchors($text) {
#
# Turn Markdown link shortcuts into XHTML <a> tags.
#
if ($this->in_anchor) return $text;
$this->in_anchor = true;
#
# First, handle reference-style links: [link text] [id]
#
$text = preg_replace_callback('{
( # wrap whole match in $1
\[
('.$this->nested_brackets_re.') # link text = $2
\]
[ ]? # one optional space
(?:\n[ ]*)? # one optional newline followed by spaces
\[
(.*?) # id = $3
\]
)
}xs',
array(&$this, '_doAnchors_reference_callback'), $text);
#
# Next, inline-style links: [link text](url "optional title")
#
$text = preg_replace_callback('{
( # wrap whole match in $1
\[
('.$this->nested_brackets_re.') # link text = $2
\]
\( # literal paren
[ ]*
(?:
<(\S*)> # href = $3
|
('.$this->nested_url_parenthesis_re.') # href = $4
)
[ ]*
( # $5
([\'"]) # quote char = $6
(.*?) # Title = $7
\6 # matching quote
[ ]* # ignore any spaces/tabs between closing quote and )
)? # title is optional
\)
)
}xs',
array(&$this, '_DoAnchors_inline_callback'), $text);
#
# Last, handle reference-style shortcuts: [link text]
# These must come last in case you've also got [link test][1]
# or [link test](/foo)
#
// $text = preg_replace_callback('{
// ( # wrap whole match in $1
// \[
// ([^\[\]]+) # link text = $2; can\'t contain [ or ]
// \]
// )
// }xs',
// array(&$this, '_doAnchors_reference_callback'), $text);
$this->in_anchor = false;
return $text;
}
function _doAnchors_reference_callback($matches) {
$whole_match = $matches[1];
$link_text = $matches[2];
$link_id =& $matches[3];
if ($link_id == "") {
# for shortcut links like [this][] or [this].
$link_id = $link_text;
}
# lower-case and turn embedded newlines into spaces
$link_id = strtolower($link_id);
$link_id = preg_replace('{[ ]?\n}', ' ', $link_id);
if (isset($this->urls[$link_id])) {
$url = $this->urls[$link_id];
$url = $this->encodeAttribute($url);
$result = "<a href=\"$url\"";
if ( isset( $this->titles[$link_id] ) ) {
$title = $this->titles[$link_id];
$title = $this->encodeAttribute($title);
$result .= " title=\"$title\"";
}
$link_text = $this->runSpanGamut($link_text);
$result .= ">$link_text</a>";
$result = $this->hashPart($result);
}
else {
$result = $whole_match;
}
return $result;
}
function _doAnchors_inline_callback($matches) {
$whole_match = $matches[1];
$link_text = $this->runSpanGamut($matches[2]);
$url = $matches[3] == '' ? $matches[4] : $matches[3];
$title =& $matches[7];
$url = $this->encodeAttribute($url);
$result = "<a href=\"$url\"";
if (isset($title)) {
$title = $this->encodeAttribute($title);
$result .= " title=\"$title\"";
}
$link_text = $this->runSpanGamut($link_text);
$result .= ">$link_text</a>";
return $this->hashPart($result);
}
function doImages($text) {
#
# Turn Markdown image shortcuts into <img> tags.
#
#
# First, handle reference-style labeled images: ![alt text][id]
#
$text = preg_replace_callback('{
( # wrap whole match in $1
!\[
('.$this->nested_brackets_re.') # alt text = $2
\]
[ ]? # one optional space
(?:\n[ ]*)? # one optional newline followed by spaces
\[
(.*?) # id = $3
\]
)
}xs',
array(&$this, '_doImages_reference_callback'), $text);
#
# Next, handle inline images: ![alt text](url "optional title")
# Don't forget: encode * and _
#
$text = preg_replace_callback('{
( # wrap whole match in $1
!\[
('.$this->nested_brackets_re.') # alt text = $2
\]
\s? # One optional whitespace character
\( # literal paren
[ ]*
(?:
<(\S*)> # src url = $3
|
('.$this->nested_url_parenthesis_re.') # src url = $4
)
[ ]*
( # $5
([\'"]) # quote char = $6
(.*?) # title = $7
\6 # matching quote
[ ]*
)? # title is optional
\)
)
}xs',
array(&$this, '_doImages_inline_callback'), $text);
return $text;
}
function _doImages_reference_callback($matches) {
$whole_match = $matches[1];
$alt_text = $matches[2];
$link_id = strtolower($matches[3]);
if ($link_id == "") {
$link_id = strtolower($alt_text); # for shortcut links like ![this][].
}
$alt_text = $this->encodeAttribute($alt_text);
if (isset($this->urls[$link_id])) {
$url = $this->encodeAttribute($this->urls[$link_id]);
$result = "<img src=\"$url\" alt=\"$alt_text\"";
if (isset($this->titles[$link_id])) {
$title = $this->titles[$link_id];
$title = $this->encodeAttribute($title);
$result .= " title=\"$title\"";
}
$result .= $this->empty_element_suffix;
$result = $this->hashPart($result);
}
else {
# If there's no such link ID, leave intact:
$result = $whole_match;
}
return $result;
}
function _doImages_inline_callback($matches) {
$whole_match = $matches[1];
$alt_text = $matches[2];
$url = $matches[3] == '' ? $matches[4] : $matches[3];
$title =& $matches[7];
$alt_text = $this->encodeAttribute($alt_text);
$url = $this->encodeAttribute($url);
$result = "<img src=\"$url\" alt=\"$alt_text\"";
if (isset($title)) {
$title = $this->encodeAttribute($title);
$result .= " title=\"$title\""; # $title already quoted
}
$result .= $this->empty_element_suffix;
return $this->hashPart($result);
}
function doHeaders($text) {
# Setext-style headers:
# Header 1
# ========
#
# Header 2
# --------
#
$text = preg_replace_callback('{ ^(.+?)[ ]*\n(=+|-+)[ ]*\n+ }mx',
array(&$this, '_doHeaders_callback_setext'), $text);
# atx-style headers:
# # Header 1
# ## Header 2
# ## Header 2 with closing hashes ##
# ...
# ###### Header 6
#
$text = preg_replace_callback('{
^(\#{1,6}) # $1 = string of #\'s
[ ]*
(.+?) # $2 = Header text
[ ]*
\#* # optional closing #\'s (not counted)
\n+
}xm',
array(&$this, '_doHeaders_callback_atx'), $text);
return $text;
}
function _doHeaders_callback_setext($matches) {
# Terrible hack to check we haven't found an empty list item.
if ($matches[2] == '-' && preg_match('{^-(?: |$)}', $matches[1]))
return $matches[0];
$level = $matches[2]{0} == '=' ? 1 : 2;
$block = "<h$level>".$this->runSpanGamut($matches[1])."</h$level>";
return "\n" . $this->hashBlock($block) . "\n\n";
}
function _doHeaders_callback_atx($matches) {
$level = strlen($matches[1]);
$block = "<h$level>".$this->runSpanGamut($matches[2])."</h$level>";
return "\n" . $this->hashBlock($block) . "\n\n";
}
function doLists($text) {
#
# Form HTML ordered (numbered) and unordered (bulleted) lists.
#
$less_than_tab = $this->tab_width - 1;
# Re-usable patterns to match list item bullets and number markers:
$marker_ul_re = '[*+-]';
$marker_ol_re = '\d+[.]';
$marker_any_re = "(?:$marker_ul_re|$marker_ol_re)";
$markers_relist = array($marker_ul_re, $marker_ol_re);
foreach ($markers_relist as $marker_re) {
# Re-usable pattern to match any entirel ul or ol list:
$whole_list_re = '
( # $1 = whole list
( # $2
[ ]{0,'.$less_than_tab.'}
('.$marker_re.') # $3 = first list item marker
[ ]+
)
(?s:.+?)
( # $4
\z
|
\n{2,}
(?=\S)
(?! # Negative lookahead for another list item marker
[ ]*
'.$marker_re.'[ ]+
)
)
)
'; // mx
# We use a different prefix before nested lists than top-level lists.
# See extended comment in _ProcessListItems().
if ($this->list_level) {
$text = preg_replace_callback('{
^
'.$whole_list_re.'
}mx',
array(&$this, '_doLists_callback'), $text);
}
else {
$text = preg_replace_callback('{
(?:(?<=\n)\n|\A\n?) # Must eat the newline
'.$whole_list_re.'
}mx',
array(&$this, '_doLists_callback'), $text);
}
}
return $text;
}
function _doLists_callback($matches) {
# Re-usable patterns to match list item bullets and number markers:
$marker_ul_re = '[*+-]';
$marker_ol_re = '\d+[.]';
$marker_any_re = "(?:$marker_ul_re|$marker_ol_re)";
$list = $matches[1];
$list_type = preg_match("/$marker_ul_re/", $matches[3]) ? "ul" : "ol";
$marker_any_re = ( $list_type == "ul" ? $marker_ul_re : $marker_ol_re );
$list .= "\n";
$result = $this->processListItems($list, $marker_any_re);
$result = $this->hashBlock("<$list_type>\n" . $result . "</$list_type>");
return "\n". $result ."\n\n";
}
var $list_level = 0;
function processListItems($list_str, $marker_any_re) {
#
# Process the contents of a single ordered or unordered list, splitting it
# into individual list items.
#
# The $this->list_level global keeps track of when we're inside a list.
# Each time we enter a list, we increment it; when we leave a list,
# we decrement. If it's zero, we're not in a list anymore.
#
# We do this because when we're not inside a list, we want to treat
# something like this:
#
# I recommend upgrading to version
# 8. Oops, now this line is treated
# as a sub-list.
#
# As a single paragraph, despite the fact that the second line starts
# with a digit-period-space sequence.
#
# Whereas when we're inside a list (or sub-list), that line will be
# treated as the start of a sub-list. What a kludge, huh? This is
# an aspect of Markdown's syntax that's hard to parse perfectly
# without resorting to mind-reading. Perhaps the solution is to
# change the syntax rules such that sub-lists must start with a
# starting cardinal number; e.g. "1." or "a.".
$this->list_level++;
# trim trailing blank lines:
$list_str = preg_replace("/\n{2,}\\z/", "\n", $list_str);
$list_str = preg_replace_callback('{
(\n)? # leading line = $1
(^[ ]*) # leading whitespace = $2
('.$marker_any_re.' # list marker and space = $3
(?:[ ]+|(?=\n)) # space only required if item is not empty
)
((?s:.*?)) # list item text = $4
(?:(\n+(?=\n))|\n) # tailing blank line = $5
(?= \n* (\z | \2 ('.$marker_any_re.') (?:[ ]+|(?=\n))))
}xm',
array(&$this, '_processListItems_callback'), $list_str);
$this->list_level--;
return $list_str;
}
function _processListItems_callback($matches) {
$item = $matches[4];
$leading_line =& $matches[1];
$leading_space =& $matches[2];
$marker_space = $matches[3];
$tailing_blank_line =& $matches[5];
if ($leading_line || $tailing_blank_line ||
preg_match('/\n{2,}/', $item))
{
# Replace marker with the appropriate whitespace indentation
$item = $leading_space . str_repeat(' ', strlen($marker_space)) . $item;
$item = $this->runBlockGamut($this->outdent($item)."\n");
}
else {
# Recursion for sub-lists:
$item = $this->doLists($this->outdent($item));
$item = preg_replace('/\n+$/', '', $item);
$item = $this->runSpanGamut($item);
}
return "<li>" . $item . "</li>\n";
}
function doCodeBlocks($text) {
#
# Process Markdown `<pre><code>` blocks.
#
$text = preg_replace_callback('{
(?:\n\n|\A\n?)
( # $1 = the code block -- one or more lines, starting with a space/tab
(?>
[ ]{'.$this->tab_width.'} # Lines must start with a tab or a tab-width of spaces
.*\n+
)+
)
((?=^[ ]{0,'.$this->tab_width.'}\S)|\Z) # Lookahead for non-space at line-start, or end of doc
}xm',
array(&$this, '_doCodeBlocks_callback'), $text);
return $text;
}
function _doCodeBlocks_callback($matches) {
$codeblock = $matches[1];
$codeblock = $this->outdent($codeblock);
$codeblock = htmlspecialchars($codeblock, ENT_NOQUOTES);
# trim leading newlines and trailing newlines
$codeblock = preg_replace('/\A\n+|\n+\z/', '', $codeblock);
$codeblock = "<pre><code>$codeblock\n</code></pre>";
return "\n\n".$this->hashBlock($codeblock)."\n\n";
}
function makeCodeSpan($code) {
#
# Create a code span markup for $code. Called from handleSpanToken.
#
$code = htmlspecialchars(trim($code), ENT_NOQUOTES);
return $this->hashPart("<code>$code</code>");
}
var $em_relist = array(
'' => '(?:(?<!\*)\*(?!\*)|(?<!_)_(?!_))(?=\S)(?![.,:;]\s)',
'*' => '(?<=\S)(?<!\*)\*(?!\*)',
'_' => '(?<=\S)(?<!_)_(?!_)',
);
var $strong_relist = array(
'' => '(?:(?<!\*)\*\*(?!\*)|(?<!_)__(?!_))(?=\S)(?![.,:;]\s)',
'**' => '(?<=\S)(?<!\*)\*\*(?!\*)',
'__' => '(?<=\S)(?<!_)__(?!_)',
);
var $em_strong_relist = array(
'' => '(?:(?<!\*)\*\*\*(?!\*)|(?<!_)___(?!_))(?=\S)(?![.,:;]\s)',
'***' => '(?<=\S)(?<!\*)\*\*\*(?!\*)',
'___' => '(?<=\S)(?<!_)___(?!_)',
);
var $em_strong_prepared_relist;
function prepareItalicsAndBold() {
#
# Prepare regular expressions for seraching emphasis tokens in any
# context.
#
foreach ($this->em_relist as $em => $em_re) {
foreach ($this->strong_relist as $strong => $strong_re) {
# Construct list of allowed token expressions.
$token_relist = array();
if (isset($this->em_strong_relist["$em$strong"])) {
$token_relist[] = $this->em_strong_relist["$em$strong"];
}
$token_relist[] = $em_re;
$token_relist[] = $strong_re;
# Construct master expression from list.
$token_re = '{('. implode('|', $token_relist) .')}';
$this->em_strong_prepared_relist["$em$strong"] = $token_re;
}
}
}
function doItalicsAndBold($text) {
$token_stack = array('');
$text_stack = array('');
$em = '';
$strong = '';
$tree_char_em = false;
while (1) {
#
# Get prepared regular expression for seraching emphasis tokens
# in current context.
#
$token_re = $this->em_strong_prepared_relist["$em$strong"];
#
# Each loop iteration seach for the next emphasis token.
# Each token is then passed to handleSpanToken.
#
$parts = preg_split($token_re, $text, 2, PREG_SPLIT_DELIM_CAPTURE);
$text_stack[0] .= $parts[0];
$token =& $parts[1];
$text =& $parts[2];
if (empty($token)) {
# Reached end of text span: empty stack without emitting.
# any more emphasis.
while ($token_stack[0]) {
$text_stack[1] .= array_shift($token_stack);
$text_stack[0] .= array_shift($text_stack);
}
break;
}
$token_len = strlen($token);
if ($tree_char_em) {
# Reached closing marker while inside a three-char emphasis.
if ($token_len == 3) {
# Three-char closing marker, close em and strong.
array_shift($token_stack);
$span = array_shift($text_stack);
$span = $this->runSpanGamut($span);
$span = "<strong><em>$span</em></strong>";
$text_stack[0] .= $this->hashPart($span);
$em = '';
$strong = '';
} else {
# Other closing marker: close one em or strong and
# change current token state to match the other
$token_stack[0] = str_repeat($token{0}, 3-$token_len);
$tag = $token_len == 2 ? "strong" : "em";
$span = $text_stack[0];
$span = $this->runSpanGamut($span);
$span = "<$tag>$span</$tag>";
$text_stack[0] = $this->hashPart($span);
$$tag = ''; # $$tag stands for $em or $strong
}
$tree_char_em = false;
} else if ($token_len == 3) {
if ($em) {
# Reached closing marker for both em and strong.
# Closing strong marker:
for ($i = 0; $i < 2; ++$i) {
$shifted_token = array_shift($token_stack);
$tag = strlen($shifted_token) == 2 ? "strong" : "em";
$span = array_shift($text_stack);
$span = $this->runSpanGamut($span);
$span = "<$tag>$span</$tag>";
$text_stack[0] .= $this->hashPart($span);
$$tag = ''; # $$tag stands for $em or $strong
}
} else {
# Reached opening three-char emphasis marker. Push on token
# stack; will be handled by the special condition above.
$em = $token{0};
$strong = "$em$em";
array_unshift($token_stack, $token);
array_unshift($text_stack, '');
$tree_char_em = true;
}
} else if ($token_len == 2) {
if ($strong) {
# Unwind any dangling emphasis marker:
if (strlen($token_stack[0]) == 1) {
$text_stack[1] .= array_shift($token_stack);
$text_stack[0] .= array_shift($text_stack);
}
# Closing strong marker:
array_shift($token_stack);
$span = array_shift($text_stack);
$span = $this->runSpanGamut($span);
$span = "<strong>$span</strong>";
$text_stack[0] .= $this->hashPart($span);
$strong = '';
} else {
array_unshift($token_stack, $token);
array_unshift($text_stack, '');
$strong = $token;
}
} else {
# Here $token_len == 1
if ($em) {
if (strlen($token_stack[0]) == 1) {
# Closing emphasis marker:
array_shift($token_stack);
$span = array_shift($text_stack);
$span = $this->runSpanGamut($span);
$span = "<em>$span</em>";
$text_stack[0] .= $this->hashPart($span);
$em = '';
} else {
$text_stack[0] .= $token;
}
} else {
array_unshift($token_stack, $token);
array_unshift($text_stack, '');
$em = $token;
}
}
}
return $text_stack[0];
}
function doBlockQuotes($text) {
$text = preg_replace_callback('/
( # Wrap whole match in $1
(?>
^[ ]*>[ ]? # ">" at the start of a line
.+\n # rest of the first line
(.+\n)* # subsequent consecutive lines
\n* # blanks
)+
)
/xm',
array(&$this, '_doBlockQuotes_callback'), $text);
return $text;
}
function _doBlockQuotes_callback($matches) {
$bq = $matches[1];
# trim one level of quoting - trim whitespace-only lines
$bq = preg_replace('/^[ ]*>[ ]?|^[ ]+$/m', '', $bq);
$bq = $this->runBlockGamut($bq); # recurse
$bq = preg_replace('/^/m', " ", $bq);
# These leading spaces cause problem with <pre> content,
# so we need to fix that:
$bq = preg_replace_callback('{(\s*<pre>.+?</pre>)}sx',
array(&$this, '_DoBlockQuotes_callback2'), $bq);
return "\n". $this->hashBlock("<blockquote>\n$bq\n</blockquote>")."\n\n";
}
function _doBlockQuotes_callback2($matches) {
$pre = $matches[1];
$pre = preg_replace('/^ /m', '', $pre);
return $pre;
}
function formParagraphs($text) {
#
# Params:
# $text - string to process with html <p> tags
#
# Strip leading and trailing lines:
$text = preg_replace('/\A\n+|\n+\z/', '', $text);
$grafs = preg_split('/\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY);
#
# Wrap <p> tags