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 = '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); $content2 = str_replace('\2', '\3', $content); # 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('{(?>
(?>
(?<=\n\n) # Starting after a blank line
| # or
\A\n? # the beginning of the doc
)
( # save in $1
# Match from `\n. It was easier to make a special # case than to make the other regex more complicated. [ ]{0,'.$less_than_tab.'} <(hr) # start tag = $2 \b # word break ([^<>])*? # /?> # the matching end tag [ ]* (?=\n{2,}|\Z) # followed by a blank line or end of document | # Special case for standalone HTML comments: [ ]{0,'.$less_than_tab.'} (?s: ) [ ]* (?=\n{2,}|\Z) # followed by a blank line or end of document | # PHP and ASP-style processor instructions ( ) [ ]* (?=\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 "S" is used for general span-level elements. # # 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("
` blocks.
#
$text = preg_replace_callback('{
(?:\n\n|\A)
( # $1 = the code block -- one or more lines, starting with a space/tab
(?>
[ ]{'.$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);
$codeblock = "$codeblock\n
";
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");
}
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 $this->hashPart("$text");
}
function _doItalicAndBold_strong_callback($matches) {
$text = $matches[2];
$text = $this->runSpanGamut($text);
return $this->hashPart("$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
$bq = preg_replace('/^[ ]*>[ ]?|^[ ]+$/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('/\A\n+|\n+\z/', '', $text);
$grafs = preg_split('/\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY);
#
# Wrap
tags and unhashify HTML blocks
#
foreach ($grafs as $key => $value) {
if (!preg_match('/^B\x1A[0-9]+B$/', $value)) {
# Is a paragraph.
$value = $this->runSpanGamut($value);
$value = preg_replace('/^([ ]*)/', "
", $value);
$value .= "
";
$grafs[$key] = $this->unhash($value);
}
else {
# Is a block.
# Modify elements of @grafs in-place...
$graf = $value;
$block = $this->html_hashes[$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.
if ($this->no_entities) {
$text = str_replace('&', '&', $text);
$text = str_replace('<', '<', $text);
return $text;
}
# 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 $text;
}
function doAutoLinks($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 $this->hashPart($link);
}
function _doAutoLinks_email_callback($matches) {
$address = $matches[1];
$link = $this->encodeEmailAddress($address);
return $this->hashPart($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 $addr;
}
function parseSpan($str) {
#
# Take the string $str and parse it into tokens, hashing embeded HTML,
# escaped characters and handling code spans.
#
$output = '';
$regex = '{
(
\\\\['.preg_quote($this->escape_chars).']
|
(?no_markup ? '' : '
|
# comment
|
<\?.*?\?> | <%.*?%> # processing instruction
|
<[/!$]?[-a-zA-Z0-9:]+ # regular tags
(?>
\s
(?>[^"\'>]+|"[^"]*"|\'[^\']*\')*
)?
>
').'
)
}xs';
while (1) {
#
# Each loop iteration seach for either the next tag, the next
# openning code span marker, or the next escaped character.
# Each token is then passed to handleSpanToken.
#
$parts = preg_split($regex, $str, 2, PREG_SPLIT_DELIM_CAPTURE);
# Create token from text preceding tag.
if ($parts[0] != "") {
$output .= $parts[0];
}
# Check if we reach the end.
if (isset($parts[1])) {
$output .= $this->handleSpanToken($parts[1], $parts[2]);
$str = $parts[2];
}
else {
break;
}
}
return $output;
}
function handleSpanToken($token, &$str) {
#
# Handle $token provided by parseSpan by determining its nature and
# returning the corresponding value that should replace it.
#
switch ($token{0}) {
case "\\":
return $this->hashPart("&#". ord($token{1}). ";");
case "`":
# Search for end marker in remaining text.
if (preg_match('/^(.*?[^`])'.$token.'(?!`)(.*)$/sm',
$str, $matches))
{
$str = $matches[2];
$codespan = $this->makeCodeSpan($matches[1]);
return $this->hashPart($codespan);
}
return $token; // return as text since no ending marker found.
default:
return $this->hashPart($token);
}
}
function outdent($text) {
#
# Remove one level of line-leading tabs or spaces
#
return preg_replace('/^(\t|[ ]{1,'.$this->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.
$text = preg_replace_callback('/^.*\t.*$/m',
array(&$this, '_detab_callback'), $text);
return $text;
}
function _detab_callback($matches) {
$line = $matches[0];
$strlen = $this->utf8_strlen; # strlen function for UTF-8.
# 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;
}
return $line;
}
function _initDetab() {
#
# Check for the availability of the function in the `utf8_strlen` property
# (initially `mb_strlen`). If the function is not available, create a
# function that will loosely count the number of UTF-8 characters with a
# regular expression.
#
if (function_exists($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 preg_replace_callback('/(.)\x1A[0-9]+\1/',
array(&$this, '_unhash_callback'), $text);
}
function _unhash_callback($matches) {
return $this->html_hashes[$matches[0]];
}
}
/*
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.
Copyright and License
---------------------
PHP Markdown
Copyright (c) 2004-2007 Michel Fortin