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|address|'. 'script|noscript|form|fieldset|iframe|math|ins|del'; $block_tags_b = 'p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|address|'. 'script|noscript|form|fieldset|iframe|math'; # 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(' # closing nested tag ) | <(?!/\2\s*> # other tags with a different name ) )*', $nested_tags_level); # First, look for nested blocks, e.g.: #
#
#
# 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 `
# tags for inner block must be indented.
#
# ` and stop at the first `
`.
$text = preg_replace_callback('{
( # save in $1
^ # start of line (with /m)
<('.$block_tags_a.')# start tag = $2
'.$attr.'>\n # attributes followed by > and \n
'.$content.' # content, support nesting
# the matching end tag
[ ]* # trailing spaces/tabs
(?=\n+|\Z) # followed by a newline or end of document
)
}xmi',
array(&$this, '_hashHTMLBlocks_callback'),
$text);
#
# Match from `\n. 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 = $2 \b # word break ([^<>])*? # /?> # the matching end tag [ ]* (?=\n{2,}|\Z) # followed by a blank line or end of document ) }xi', array(&$this, '_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: ) [ ]* (?=\n{2,}|\Z) # followed by a blank line or end of document ) }x', array(&$this, '_hashHTMLBlocks_callback'), $text); # PHP and ASP-style processor instructions ( ) [ ]* (?=\n{2,}|\Z) # followed by a blank line or end of document ) }x', array(&$this, '_hashHTMLBlocks_callback'), $text); return $text; } function _hashHTMLBlocks_callback($matches) { $text = $matches[1]; $key = $this->hashBlock($text); return "\n\n$key\n\n"; } function hashBlock($text) { # # Called whenever a tag must be hashed when a function insert a block-level # tag in $text, it pass through this function and is automaticaly escaped, # which remove the need to call _HashHTMLBlocks at every step. # # 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. $key = "B\x1A". md5($text); $this->html_hashes[$key] = $text; $this->html_blocks[$key] = $text; return $key; # String that will replace the tag. } function hashSpan($text, $word_separator = false) { # # Called whenever a tag must be hashed when a function insert a span-level # element in $text, it pass through this function and is automaticaly # escaped, blocking invalid nested overlap. If optional argument # $word_separator is true, surround the hash value by spaces. # # 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 span. $key = "S\x1A". md5($text); if ($word_separator) $key = ":$key:"; $this->html_hashes[$key] = $text; return $key; # String that will replace the span tag. } 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( array('{^[ ]{0,2}([ ]?\*[ ]?){3,}[ ]*$}mx', '{^[ ]{0,2}([ ]? -[ ]?){3,}[ ]*$}mx', '{^[ ]{0,2}([ ]? _[ ]?){3,}[ ]*$}mx'), "\n".$this->hashBlock("
".$this->runSpanGamut($matches[1])."
"; return "\n" . $this->hashBlock($block) . "\n\n"; } function _doHeaders_callback_setext_h2($matches) { $block = "".$this->runSpanGamut($matches[1])."
"; return "\n" . $this->hashBlock($block) . "\n\n"; } function _doHeaders_callback_atx($matches) { $level = strlen($matches[1]); $block = "` blocks.
#
$text = preg_replace_callback('{
(?:\n\n|\A)
( # $1 = the code block -- one or more lines, starting with a space/tab
(?:
(?:[ ]{'.$this->tab_width.'} | \t) # 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 = $this->encodeCode($this->outdent($codeblock));
// $codeblock = $this->detab($codeblock);
# trim leading newlines and trailing whitespace
$codeblock = preg_replace(array('/\A\n+/', '/\n+\z/'), '', $codeblock);
$result = "\n\n".$this->hashBlock("" . $codeblock . "\n
")."\n\n";
return $result;
}
function doCodeSpans($text) {
#
# * Backtick quotes are used for Just type foo `bar` baz at the prompt.
#
# There's no arbitrary limit to the number of backticks you
# can use as delimters. If you need three consecutive backticks
# in your code, use four for delimiters, etc.
#
# * You can use spaces to get literal backticks at the edges:
#
# ... type `` `bar` `` ...
#
# Turns to:
#
# ... type `bar` ...
#
$text = preg_replace_callback('@
(?encodeCode($c);
return $this->hashSpan("$c");
}
function encodeCode($_) {
#
# Encode/escape certain characters inside Markdown code runs.
# The point is that in code, these characters are literals,
# and lose their special Markdown meanings.
#
# Encode all ampersands; HTML entities are not
# entities within a Markdown code span.
$_ = str_replace('&', '&', $_);
# Do the angle bracket song and dance:
$_ = str_replace(array('<', '>'),
array('<', '>'), $_);
# Now, escape characters that are magic in Markdown:
// $_ = str_replace(array_keys($this->escape_table),
// array_values($this->escape_table), $_);
return $_;
}
function doItalicsAndBold($text) {
# must go first:
$text = preg_replace_callback('{
( # $1: Marker
(?
[^*_]+? # Anthing not em markers.
|
# Balence any regular emphasis inside.
\1 (?=\S) .+? (?<=\S) \1
|
. # Allow unbalenced * and _.
)+?
)
(?<=\S) \1\1 # End mark not preceded by whitespace.
}sx',
array(&$this, '_doItalicAndBold_strong_callback'), $text);
# Then :
$text = preg_replace_callback(
'{ ( (?runSpanGamut($text);
return $this->hashSpan("$text");
}
function _doItalicAndBold_strong_callback($matches) {
$text = $matches[2];
$text = $this->runSpanGamut($text);
return $this->hashSpan("$text");
}
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
$bq = preg_replace(array('/^[ ]*>[ ]?/m', '/^[ ]+$/m'), '', $bq);
$bq = $this->runBlockGamut($bq); # recurse
$bq = preg_replace('/^/m', " ", $bq);
# These leading spaces cause problem with content,
# so we need to fix that:
$bq = preg_replace_callback('{(\s*.+?
)}sx',
array(&$this, '_DoBlockQuotes_callback2'), $bq);
return "\n". $this->hashBlock("\n$bq\n
")."\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 tags
#
# Strip leading and trailing lines:
$text = preg_replace(array('/\A\n+/', '/\n+\z/'), '', $text);
$grafs = preg_split('/\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY);
#
# Wrap
tags.
#
foreach ($grafs as $key => $value) {
if (!isset( $this->html_blocks[$value] )) {
$value = $this->runSpanGamut($value);
$value = preg_replace('/^([ ]*)/', "
", $value);
$value .= "
";
$grafs[$key] = $this->unhash($value);
}
}
#
# Unhashify HTML blocks
#
foreach ($grafs as $key => $graf) {
# Modify elements of @grafs in-place...
if (isset($this->html_blocks[$graf])) {
$block = $this->html_blocks[$graf];
$graf = $block;
// if (preg_match('{
// \A
// ( # $1 = tag
// ]*
// \b
// markdown\s*=\s* ([\'"]) # $2 = attr quote char
// 1
// \2
// [^>]*
// >
// )
// ( # $3 = contents
// .*
// )
// () # $4 = closing tag
// \z
// }xs', $block, $matches))
// {
// list(, $div_open, , $div_content, $div_close) = $matches;
//
// # We can't call Markdown(), because that resets the hash;
// # that initialization code should be pulled into its own sub, though.
// $div_content = $this->hashHTMLBlocks($div_content);
//
// # Run document gamut methods on the content.
// foreach ($this->document_gamut as $method => $priority) {
// $div_content = $this->$method($div_content);
// }
//
// $div_open = preg_replace(
// '{\smarkdown\s*=\s*([\'"]).+?\1}', '', $div_open);
//
// $graf = $div_open . "\n" . $div_content . "\n" . $div_close;
// }
$grafs[$key] = $graf;
}
}
return implode("\n\n", $grafs);
}
function encodeAmpsAndAngles($text) {
# Smart processing for ampersands and angle brackets that need to be encoded.
# Ampersand-encoding based entirely on Nat Irons's Amputator MT plugin:
# http://bumppo.net/projects/amputator/
$text = preg_replace('/&(?!#?[xX]?(?:[0-9a-fA-F]+|\w+);)/',
'&', $text);;
# Encode naked <'s
$text = preg_replace('{<(?![a-z/?\$!%])}i', '<', $text);
return $text;
}
function encodeBackslashEscapes($text) {
#
# Parameter: String.
# Returns: The string, with after processing the following backslash
# escape sequences.
#
# Must process escaped backslashes first.
return str_replace(array_keys($this->backslash_escape_table),
array_values($this->backslash_escape_table), $text);
}
function doAutoLinks($text) {
$text = preg_replace_callback('{<((https?|ftp|dict):[^\'">\s]+)>}',
array(&$this, '_doAutoLinks_url_callback'), $text);
# Email addresses:
$text = preg_replace_callback('{
<
(?:mailto:)?
(
[-.\w\x80-\xFF]+
\@
[-a-z0-9\x80-\xFF]+(\.[-a-z0-9\x80-\xFF]+)*\.[a-z]+
)
>
}xi',
array(&$this, '_doAutoLinks_email_callback'), $text);
return $text;
}
function _doAutoLinks_url_callback($matches) {
$url = $this->encodeAmpsAndAngles($matches[1]);
$link = "$url";
return $this->hashSpan($link);
}
function _doAutoLinks_email_callback($matches) {
$address = $matches[1];
$link = $this->encodeEmailAddress($address);
return $this->hashSpan($link);
}
function encodeEmailAddress($addr) {
#
# Input: an email address, e.g. "foo@example.com"
#
# Output: the email address as a mailto link, with each character
# of the address encoded as either a decimal or hex entity, in
# the hopes of foiling most address harvesting spam bots. E.g.:
#
#
#
# Based by a filter by Matthew Wickline, posted to BBEdit-Talk.
# With some optimizations by Milian Wolff.
#
$addr = "mailto:" . $addr;
$chars = preg_split('/(? $char) {
$ord = ord($char);
# Ignore non-ascii chars.
if ($ord < 128) {
$r = ($seed * (1 + $key)) % 100; # Pseudo-random function.
# roughly 10% raw, 45% hex, 45% dec
# '@' *must* be encoded. I insist.
if ($r > 90 && $char != '@') /* do nothing */;
else if ($r < 45) $chars[$key] = '&#x'.dechex($ord).';';
else $chars[$key] = '&#'.$ord.';';
}
}
$addr = implode('', $chars);
$text = implode('', array_slice($chars, 7)); # text without `mailto:`
$addr = "$text";
return $addr;
}
function tokenizeHTML($str) {
#
# Parameter: String containing HTML + Markdown markup.
# Returns: An array of the tokens comprising the input
# string. Each token is either a tag or a run of text
# between tags. Each element of the array is a
# two-element array; the first is either 'tag' or 'text';
# the second is the actual value.
# Note: Markdown code spans are taken into account: no tag token is
# generated within a code span.
#
$tokens = array();
while ($str != "") {
#
# Each loop iteration seach for either the next tag or the next
# openning code span marker. If a code span marker is found, the
# code span is extracted in entierty and will result in an extra
# text token.
#
$parts = preg_split('{
(
(? # comment
|
<\?.*?\?> | <%.*?%> # processing instruction
|
<[/!$]?[-a-zA-Z0-9:]+ # regular tags
(?:
\s
(?>[^"\'>]+|"[^"]*"|\'[^\']*\')*
)?
>
)
}xs', $str, 2, PREG_SPLIT_DELIM_CAPTURE);
# Create token from text preceding tag.
if ($parts[0] != "") {
$tokens[] = array('text', $parts[0]);
}
# Check if we reach the end.
if (count($parts) < 3) {
break;
}
# Create token from tag or code span.
if ($parts[1]{0} == "`") {
$tokens[] = array('text', $parts[1]);
$str = $parts[2];
# Skip the whole code span, pass as text token.
if (preg_match('/^(.*(?tab_width})/m", "", $text);
}
# String length function for detab. `_initDetab` will create a function to
# hanlde UTF-8 if the default function does not exist.
var $utf8_strlen = 'mb_strlen';
function detab($text) {
#
# Replace tabs with the appropriate amount of space.
#
# For each line we separate the line in blocks delemited by
# tab characters. Then we reconstruct every line by adding the
# appropriate number of space between each blocks.
$strlen = $this->utf8_strlen; # strlen function for UTF-8.
$lines = explode("\n", $text);
$text = "";
foreach ($lines as $line) {
# Split in blocks.
$blocks = explode("\t", $line);
# Add each blocks to the line.
$line = $blocks[0];
unset($blocks[0]); # Do not add first block twice.
foreach ($blocks as $block) {
# Calculate amount of space, insert spaces, insert block.
$amount = $this->tab_width -
$strlen($line, 'UTF-8') % $this->tab_width;
$line .= str_repeat(" ", $amount) . $block;
}
$text .= "$line\n";
}
return $text;
}
function _initDetab() {
#
# Check for the availability of the function in the `utf8_strlen` property
# (initially `mb_strlen`). If the function is not available, create a
# function that will loosely count the number of UTF-8 characters with a
# regular expression.
#
if (function_exists($this->utf8_strlen)) return;
$this->utf8_strlen = create_function('$text', 'return preg_match_all(
"/[\\\\x00-\\\\xBF]|[\\\\xC0-\\\\xFF][\\\\x80-\\\\xBF]*/",
$text, $m);');
}
function unhash($text) {
#
# Swap back in all the tags hashed by _HashHTMLBlocks.
#
return str_replace(array_keys($this->html_hashes),
array_values($this->html_hashes), $text);
}
}
/*
PHP Markdown
============
Description
-----------
This is a PHP translation of the original Markdown formatter written in
Perl by John Gruber.
Markdown is a text-to-HTML filter; it translates an easy-to-read /
easy-to-write structured text format into HTML. Markdown's text format
is most similar to that of plain text email, and supports features such
as headers, *emphasis*, code blocks, blockquotes, and links.
Markdown's syntax is designed not as a generic markup language, but
specifically to serve as a front-end to (X)HTML. You can use span-level
HTML tags anywhere in a Markdown document, and you can use block level
HTML tags (like and as well).
For more information about Markdown's syntax, see:
Please include with your report: (1) the example input; (2) the output you
expected; (3) the output Markdown actually produced.
Version History
---------------
See the readme file for detailed release notes for this version.
1.0.1g (3 Jul 2007)
1.0.1f (7 Feb 2007)
1.0.1e (28 Dec 2006)
1.0.1d (1 Dec 2006)
1.0.1c (9 Dec 2005)
1.0.1b (6 Jun 2005)
1.0.1a (15 Apr 2005)
1.0.1 (16 Dec 2004)
1.0 (21 Aug 2004)
Copyright and License
---------------------
PHP Markdown
Copyright (c) 2004-2007 Michel Fortin