1
0
mirror of https://github.com/erusev/parsedown.git synced 2023-08-10 21:13:06 +03:00
parsedown/Parsedown.php

1143 lines
32 KiB
PHP
Raw Normal View History

2013-11-09 01:40:00 +04:00
<?php
2013-07-11 00:22:16 +04:00
2013-11-09 01:40:00 +04:00
#
#
2013-07-11 00:22:16 +04:00
# Parsedown
# http://parsedown.org
2013-11-09 01:40:00 +04:00
#
# (c) Emanuil Rusev
2013-07-11 00:22:16 +04:00
# http://erusev.com
2013-11-09 01:40:00 +04:00
#
2014-01-24 03:28:03 +04:00
# For the full license information, view the LICENSE file that was distributed
# with this source code.
2013-11-09 01:40:00 +04:00
#
#
2013-07-11 00:22:16 +04:00
class Parsedown
{
2014-01-25 20:47:44 +04:00
# Multiton
2014-01-24 03:28:03 +04:00
static function instance($name = 'default')
{
if (isset(self::$instances[$name]))
2014-01-24 00:34:02 +04:00
{
return self::$instances[$name];
2014-01-24 00:34:02 +04:00
}
2013-11-09 01:40:00 +04:00
$instance = new Parsedown();
2013-11-09 01:40:00 +04:00
self::$instances[$name] = $instance;
2013-11-09 01:40:00 +04:00
return $instance;
}
2013-11-09 01:40:00 +04:00
private static $instances = array();
2013-11-09 01:40:00 +04:00
#
# Setters
#
2014-01-25 20:47:44 +04:00
# Enables GFM line breaks.
2014-01-24 03:28:03 +04:00
function set_breaks_enabled($breaks_enabled)
{
$this->breaks_enabled = $breaks_enabled;
return $this;
}
2014-01-26 21:53:24 +04:00
private $breaks_enabled = false;
2013-11-09 01:40:00 +04:00
#
2014-01-29 04:14:59 +04:00
# Synopsis
2014-01-24 03:28:03 +04:00
#
2014-01-29 04:14:59 +04:00
# Markdown is intended to be easy-to-read by humans - those of us who read
# line by line, left to right, top to bottom. In order to take advantage of
# this, Parsedown tries to read in a similar way. It breaks texts into
# lines, it iterates through them and it looks at how they start and relate
# to each other.
#
# Methods
#
2013-11-09 01:40:00 +04:00
function parse($text)
{
2014-01-24 02:41:45 +04:00
# standardize line breaks
$text = str_replace("\r\n", "\n", $text);
$text = str_replace("\r", "\n", $text);
2013-11-09 01:40:00 +04:00
2014-01-24 02:41:45 +04:00
# replace tabs with spaces
$text = str_replace("\t", ' ', $text);
2013-11-09 01:40:00 +04:00
2014-01-24 02:41:45 +04:00
# remove surrounding line breaks
$text = trim($text, "\n");
2013-11-09 01:40:00 +04:00
2014-01-24 02:41:45 +04:00
# split text into lines
$lines = explode("\n", $text);
2013-11-09 01:40:00 +04:00
2014-01-24 02:41:45 +04:00
# convert lines into html
$text = $this->parse_block_elements($lines);
2013-11-09 01:40:00 +04:00
2014-01-24 02:41:45 +04:00
# remove trailing line breaks
2014-01-23 15:07:15 +04:00
$text = chop($text, "\n");
2013-11-09 01:40:00 +04:00
return $text;
}
2013-11-09 01:40:00 +04:00
#
2014-01-25 20:47:44 +04:00
# Private
2013-11-09 01:40:00 +04:00
private function parse_block_elements(array $lines, $context = '')
{
2014-01-26 15:47:56 +04:00
$blocks = array();
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
$block = array(
'type' => '',
);
2013-11-09 01:40:00 +04:00
foreach ($lines as $line)
{
2014-01-29 04:14:59 +04:00
# context
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
switch ($block['type'])
{
2014-01-28 00:21:58 +04:00
case 'fenced':
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
if ( ! isset($block['closed']))
{
2014-01-26 15:47:56 +04:00
if (preg_match('/^[ ]*'.$block['fence'][0].'{3,}[ ]*$/', $line))
{
2014-01-26 15:47:56 +04:00
$block['closed'] = true;
}
else
{
2014-01-26 15:47:56 +04:00
if ($block['text'] !== '')
{
2014-01-26 15:47:56 +04:00
$block['text'] .= "\n";
}
2014-01-26 15:47:56 +04:00
$block['text'] .= $line;
}
continue 2;
}
2013-11-09 01:40:00 +04:00
break;
2013-11-02 23:42:55 +04:00
2014-01-28 00:21:58 +04:00
case 'markup':
2014-01-26 15:47:56 +04:00
if ( ! isset($block['closed']))
{
if (stripos($line, $block['start']) !== false) # opening tag
{
2014-01-26 15:47:56 +04:00
$block['depth']++;
}
if (stripos($line, $block['end']) !== false) # closing tag
{
if ($block['depth'] > 0)
{
$block['depth']--;
}
else
{
$block['closed'] = true;
}
}
2014-01-26 15:47:56 +04:00
$block['text'] .= "\n".$line;
continue 2;
}
break;
}
2013-11-09 01:40:00 +04:00
2014-01-29 04:14:59 +04:00
# ~
2013-11-09 01:40:00 +04:00
$indentation = 0;
while(isset($line[$indentation]) and $line[$indentation] === ' ')
{
$indentation++;
}
$outdented_line = $indentation > 0 ? ltrim($line) : $line;
2014-01-17 01:43:34 +04:00
2014-01-29 04:14:59 +04:00
# blank
if ($outdented_line === '')
{
2014-01-26 15:47:56 +04:00
$block['interrupted'] = true;
2013-11-09 01:40:00 +04:00
continue;
}
2013-11-09 01:40:00 +04:00
2014-01-29 04:14:59 +04:00
# context
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
switch ($block['type'])
{
2014-01-28 00:21:58 +04:00
case 'quote':
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
if ( ! isset($block['interrupted']))
{
$line = preg_replace('/^[ ]*>[ ]?/', '', $line);
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
$block['lines'] []= $line;
2013-11-09 01:40:00 +04:00
continue 2;
}
2013-11-09 01:40:00 +04:00
break;
case 'li':
if ($block['indentation'] === $indentation and preg_match('/^'.$block['marker'].'[ ]+(.*)/', $outdented_line, $matches))
{
unset($block['last']);
$blocks []= $block;
$block['last'] = true;
2014-02-03 00:27:22 +04:00
$block['lines'] = array($matches[1]);
unset($block['first']);
unset($block['interrupted']);
continue 2;
}
2013-11-09 01:40:00 +04:00
if ( ! isset($block['interrupted']))
{
2014-02-03 00:27:22 +04:00
$line = preg_replace('/^[ ]{0,'.$block['baseline'].'}/', '', $line);
2013-11-09 01:40:00 +04:00
$block['lines'] []= $line;
continue 2;
}
elseif ($line[0] === ' ')
{
$block['lines'] []= '';
2014-02-03 00:27:22 +04:00
$line = preg_replace('/^[ ]{0,'.$block['baseline'].'}/', '', $line);
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
$block['lines'] []= $line;
2013-11-09 01:40:00 +04:00
unset($block['interrupted']);
continue 2;
}
2013-11-09 01:40:00 +04:00
break;
}
2013-11-09 01:40:00 +04:00
# indentation sensitive types
2013-11-09 01:40:00 +04:00
switch ($line[0])
{
case ' ':
2013-11-10 00:23:56 +04:00
2014-01-29 04:14:59 +04:00
# code
if ($indentation >= 4)
{
$code_line = substr($line, 4);
2014-01-17 01:43:12 +04:00
2014-01-28 00:21:58 +04:00
if ($block['type'] === 'code')
{
2014-01-26 15:47:56 +04:00
if (isset($block['interrupted']))
{
2014-01-26 15:47:56 +04:00
$block['text'] .= "\n";
2014-01-26 15:47:56 +04:00
unset($block['interrupted']);
}
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
$block['text'] .= "\n".$code_line;
}
else
{
2014-01-26 15:47:56 +04:00
$blocks []= $block;
2014-01-26 15:47:56 +04:00
$block = array(
2014-01-28 00:21:58 +04:00
'type' => 'code',
'text' => $code_line,
);
}
continue 2;
}
2013-11-09 01:40:00 +04:00
break;
2013-11-09 01:40:00 +04:00
case '#':
# atx heading (#)
2013-11-09 01:40:00 +04:00
if (isset($line[1]))
{
2014-01-26 15:47:56 +04:00
$blocks []= $block;
2013-11-09 01:40:00 +04:00
$level = 1;
2014-01-17 02:36:11 +04:00
while (isset($line[$level]) and $line[$level] === '#')
{
$level++;
}
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
$block = array(
'type' => 'heading',
'text' => trim($line, '# '),
'level' => $level,
);
2013-11-09 01:40:00 +04:00
continue 2;
}
2013-11-09 01:40:00 +04:00
break;
2013-11-09 01:40:00 +04:00
case '-':
case '=':
2013-11-09 01:40:00 +04:00
2014-01-29 04:14:59 +04:00
# setext heading (===)
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
if ($block['type'] === 'paragraph' and isset($block['interrupted']) === false)
{
2014-01-23 15:07:15 +04:00
$chopped_line = chop($line);
2013-11-09 01:40:00 +04:00
$i = 1;
2013-11-09 01:40:00 +04:00
while (isset($chopped_line[$i]))
{
if ($chopped_line[$i] !== $line[0])
{
break 2;
}
2013-11-09 01:40:00 +04:00
$i++;
}
2014-01-26 15:47:56 +04:00
$block['type'] = 'heading';
2014-01-26 15:47:56 +04:00
$block['level'] = $line[0] === '-' ? 2 : 1;
2013-11-09 01:40:00 +04:00
continue 2;
}
break;
}
2013-11-10 00:23:56 +04:00
# indentation insensitive types
2013-11-09 01:40:00 +04:00
switch ($outdented_line[0])
{
case '<':
2013-11-09 01:40:00 +04:00
$position = strpos($outdented_line, '>');
2013-11-09 01:40:00 +04:00
2014-01-27 02:10:24 +04:00
if ($position > 1)
{
$substring = substr($outdented_line, 1, $position - 1);
2013-11-09 01:40:00 +04:00
2014-01-27 02:10:24 +04:00
$substring = chop($substring);
if (substr($substring, -1) === '/')
{
2014-01-27 02:10:24 +04:00
$is_self_closing = true;
2013-11-09 01:40:00 +04:00
2014-01-27 02:10:24 +04:00
$substring = substr($substring, 0, -1);
}
2014-01-27 02:10:24 +04:00
$position = strpos($substring, ' ');
if ($position)
{
2014-01-27 02:10:24 +04:00
$name = substr($substring, 0, $position);
}
else
{
$name = $substring;
}
$name = strtolower($name);
// hr,h1,h2,h3,h4,h5,h6 cases
if ($name[0] == 'h' and strpos('r123456', $name[1]) !== false)
{
if ($name == 'hr')
$is_self_closing = 1;
}
elseif ( ! ctype_alpha($name))
{
break;
}
2014-01-27 02:10:24 +04:00
if (in_array($name, self::$text_level_elements))
{
break;
}
2014-01-26 15:47:56 +04:00
$blocks []= $block;
2014-01-27 02:10:24 +04:00
if (isset($is_self_closing))
{
2014-01-26 15:47:56 +04:00
$block = array(
'type' => 'self-closing tag',
'text' => $outdented_line,
);
2014-01-27 02:10:24 +04:00
unset($is_self_closing);
continue 2;
}
2014-01-26 15:47:56 +04:00
$block = array(
2014-01-28 00:21:58 +04:00
'type' => 'markup',
'text' => $outdented_line,
'start' => '<'.$name.'>',
'end' => '</'.$name.'>',
'depth' => 0,
);
2013-11-09 01:40:00 +04:00
if (stripos($outdented_line, $block['end']))
{
2014-01-26 15:47:56 +04:00
$block['closed'] = true;
}
continue 2;
}
2013-11-09 01:40:00 +04:00
break;
2013-11-09 01:40:00 +04:00
case '>':
2013-11-09 01:40:00 +04:00
# quote
2013-11-09 01:40:00 +04:00
if (preg_match('/^>[ ]?(.*)/', $outdented_line, $matches))
{
2014-01-26 15:47:56 +04:00
$blocks []= $block;
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
$block = array(
2014-01-28 00:21:58 +04:00
'type' => 'quote',
'lines' => array(
$matches[1],
),
);
continue 2;
}
break;
case '[':
2013-11-09 01:40:00 +04:00
# reference
2014-02-05 16:03:23 +04:00
$position = strpos($outdented_line, ']:');
if ($position)
{
2014-02-05 16:03:23 +04:00
$reference = array();
2013-11-02 23:42:55 +04:00
2014-02-05 16:03:23 +04:00
$label = substr($outdented_line, 1, $position - 1);
$label = strtolower($label);
2014-02-05 16:03:23 +04:00
$substring = substr($outdented_line, $position + 2);
$substring = trim($substring);
2014-02-06 04:36:11 +04:00
if ($substring === '')
{
break;
}
2014-02-05 16:03:23 +04:00
if ($substring[0] === '<')
{
$position = strpos($substring, '>');
if ($position === false)
{
break;
}
$reference['»'] = substr($substring, 1, $position - 1);
$substring = substr($substring, $position + 1);
}
else
{
$position = strpos($substring, ' ');
if ($position === false)
{
$reference['»'] = $substring;
2013-12-26 23:53:48 +04:00
2014-02-05 16:03:23 +04:00
$substring = false;
}
else
{
$reference['»'] = substr($substring, 0, $position);
$substring = substr($substring, $position + 1);
}
}
if ($substring !== false)
{
2014-02-06 04:36:22 +04:00
if ($substring[0] !== '"' and $substring[0] !== "'" and $substring[0] !== '(')
2014-02-05 16:03:23 +04:00
{
2014-02-06 04:36:22 +04:00
break;
}
2014-02-05 16:03:23 +04:00
2014-02-06 04:36:22 +04:00
$last_char = substr($substring, -1);
2014-02-05 16:03:23 +04:00
2014-02-06 04:36:22 +04:00
if ($last_char !== '"' and $last_char !== "'" and $last_char !== ')')
{
break;
2014-02-05 16:03:23 +04:00
}
2014-02-06 04:36:22 +04:00
$reference['#'] = substr($substring, 1, -1);
}
2013-11-09 01:40:00 +04:00
2014-02-05 16:03:23 +04:00
$this->reference_map[$label] = $reference;
continue 2;
}
break;
2013-11-09 01:40:00 +04:00
case '`':
case '~':
# fenced code block
if (preg_match('/^([`]{3,}|[~]{3,})[ ]*(\S+)?[ ]*$/', $outdented_line, $matches))
{
2014-01-26 15:47:56 +04:00
$blocks []= $block;
2014-01-26 15:47:56 +04:00
$block = array(
2014-01-28 00:21:58 +04:00
'type' => 'fenced',
'text' => '',
'fence' => $matches[1],
);
if (isset($matches[2]))
{
2014-01-26 15:47:56 +04:00
$block['language'] = $matches[2];
}
continue 2;
}
break;
case '*':
case '+':
case '-':
case '_':
2013-11-09 01:40:00 +04:00
# hr
2013-11-09 01:40:00 +04:00
if (preg_match('/^([-*_])([ ]{0,2}\1){2,}[ ]*$/', $outdented_line))
{
2014-01-26 15:47:56 +04:00
$blocks []= $block;
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
$block = array(
2014-01-28 00:21:58 +04:00
'type' => 'rule',
);
continue 2;
}
# li
if (preg_match('/^([*+-][ ]+)(.*)/', $outdented_line, $matches))
{
2014-01-26 15:47:56 +04:00
$blocks []= $block;
2014-02-03 00:27:22 +04:00
$baseline = $indentation + strlen($matches[1]);
2014-01-26 15:47:56 +04:00
$block = array(
'type' => 'li',
'indentation' => $indentation,
2014-02-03 00:27:22 +04:00
'baseline' => $baseline,
'marker' => '[*+-]',
'first' => true,
'last' => true,
'lines' => array(),
);
2014-02-03 00:27:22 +04:00
$block['lines'] []= preg_replace('/^[ ]{0,4}/', '', $matches[2]);
continue 2;
}
}
2013-11-09 01:40:00 +04:00
2014-02-03 00:27:22 +04:00
# li
2013-11-09 01:40:00 +04:00
if ($outdented_line[0] <= '9' and preg_match('/^(\d+[.][ ]+)(.*)/', $outdented_line, $matches))
{
2014-01-26 15:47:56 +04:00
$blocks []= $block;
2013-11-09 01:40:00 +04:00
2014-02-03 00:27:22 +04:00
$baseline = $indentation + strlen($matches[1]);
2014-01-26 15:47:56 +04:00
$block = array(
'type' => 'li',
'indentation' => $indentation,
2014-02-03 00:27:22 +04:00
'baseline' => $baseline,
'marker' => '\d+[.]',
'first' => true,
'last' => true,
'ordered' => true,
'lines' => array(),
);
2013-11-09 01:40:00 +04:00
2014-02-03 00:27:22 +04:00
$block['lines'] []= preg_replace('/^[ ]{0,4}/', '', $matches[2]);
continue;
}
2013-11-09 01:40:00 +04:00
# paragraph
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
if ($block['type'] === 'paragraph')
{
2014-01-26 15:47:56 +04:00
if (isset($block['interrupted']))
{
2014-01-26 15:47:56 +04:00
$blocks []= $block;
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
$block['text'] = $line;
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
unset($block['interrupted']);
}
else
{
if ($this->breaks_enabled)
{
2014-01-26 15:47:56 +04:00
$block['text'] .= ' ';
}
2014-01-20 11:26:25 +04:00
2014-01-26 15:47:56 +04:00
$block['text'] .= "\n".$line;
}
}
else
{
2014-01-26 15:47:56 +04:00
$blocks []= $block;
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
$block = array(
'type' => 'paragraph',
'text' => $line,
);
}
}
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
$blocks []= $block;
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
unset($blocks[0]);
2013-11-09 01:40:00 +04:00
2014-01-29 04:14:59 +04:00
# $blocks » HTML
2013-11-09 01:40:00 +04:00
$markup = '';
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
foreach ($blocks as $block)
{
2014-01-26 15:47:56 +04:00
switch ($block['type'])
{
case 'paragraph':
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
$text = $this->parse_span_elements($block['text']);
2013-11-09 01:40:00 +04:00
if ($context === 'li' and $markup === '')
{
2014-01-26 15:47:56 +04:00
if (isset($block['interrupted']))
{
$markup .= "\n".'<p>'.$text.'</p>'."\n";
}
else
{
$markup .= $text;
2014-01-26 15:47:56 +04:00
if (isset($blocks[2]))
{
$markup .= "\n";
}
}
}
else
{
$markup .= '<p>'.$text.'</p>'."\n";
}
2013-11-09 01:40:00 +04:00
break;
2013-11-09 01:40:00 +04:00
2014-01-28 00:21:58 +04:00
case 'quote':
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
$text = $this->parse_block_elements($block['lines']);
2013-11-09 01:40:00 +04:00
$markup .= '<blockquote>'."\n".$text.'</blockquote>'."\n";
2013-11-09 01:40:00 +04:00
break;
2013-11-09 01:40:00 +04:00
2014-01-28 00:21:58 +04:00
case 'code':
2014-01-20 01:34:20 +04:00
2014-01-26 15:47:56 +04:00
$text = htmlspecialchars($block['text'], ENT_NOQUOTES, 'UTF-8');
2014-01-20 01:34:20 +04:00
$markup .= '<pre><code>'.$text.'</code></pre>'."\n";
2014-01-20 01:34:20 +04:00
break;
2014-01-20 01:34:20 +04:00
2014-01-28 00:21:58 +04:00
case 'fenced':
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
$text = htmlspecialchars($block['text'], ENT_NOQUOTES, 'UTF-8');
2013-11-09 01:40:00 +04:00
$markup .= '<pre><code';
2014-01-20 01:34:20 +04:00
2014-01-26 15:47:56 +04:00
if (isset($block['language']))
{
2014-01-26 15:47:56 +04:00
$markup .= ' class="language-'.$block['language'].'"';
}
2013-12-03 01:26:43 +04:00
$markup .= '>'.$text.'</code></pre>'."\n";
2013-11-09 01:40:00 +04:00
break;
2013-11-09 01:40:00 +04:00
case 'heading':
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
$text = $this->parse_span_elements($block['text']);
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
$markup .= '<h'.$block['level'].'>'.$text.'</h'.$block['level'].'>'."\n";
2013-11-09 01:40:00 +04:00
break;
2013-11-09 01:40:00 +04:00
2014-01-28 00:21:58 +04:00
case 'rule':
2013-11-09 01:40:00 +04:00
$markup .= '<hr />'."\n";
2013-11-09 01:40:00 +04:00
break;
2013-11-02 23:42:55 +04:00
case 'li':
2013-11-17 00:04:26 +04:00
2014-01-26 15:47:56 +04:00
if (isset($block['first']))
{
$type = isset($block['ordered']) ? 'ol' : 'ul';
2013-11-17 00:04:26 +04:00
$markup .= '<'.$type.'>'."\n";
}
2013-11-17 00:04:26 +04:00
2014-01-26 15:47:56 +04:00
if (isset($block['interrupted']) and ! isset($block['last']))
{
2014-01-26 15:47:56 +04:00
$block['lines'] []= '';
}
2013-11-17 00:04:26 +04:00
2014-01-26 15:47:56 +04:00
$text = $this->parse_block_elements($block['lines'], 'li');
2013-11-17 00:04:26 +04:00
$markup .= '<li>'.$text.'</li>'."\n";
2013-11-17 00:04:26 +04:00
2014-01-26 15:47:56 +04:00
if (isset($block['last']))
{
$type = isset($block['ordered']) ? 'ol' : 'ul';
2014-01-22 03:03:21 +04:00
$markup .= '</'.$type.'>'."\n";
}
2013-11-17 00:04:26 +04:00
break;
2014-01-28 00:21:58 +04:00
case 'markup':
2014-01-26 15:47:56 +04:00
$markup .= $block['text']."\n";
break;
2013-11-17 00:04:26 +04:00
default:
2013-11-09 01:40:00 +04:00
2014-01-26 15:47:56 +04:00
$markup .= $block['text']."\n";
}
}
2013-11-09 01:40:00 +04:00
return $markup;
}
2013-11-09 01:40:00 +04:00
private function parse_span_elements($text, $markers = array(" \n", '![', '&', '*', '<', '[', '\\', '_', '`', 'http', '~~'))
{
if (isset($text[1]) === false or $markers === array())
{
return $text;
}
2013-11-09 01:40:00 +04:00
# ~
2013-11-09 01:40:00 +04:00
$markup = '';
2013-11-09 01:40:00 +04:00
while ($markers)
{
$closest_marker = null;
$closest_marker_index = 0;
$closest_marker_position = null;
2013-11-09 01:40:00 +04:00
foreach ($markers as $index => $marker)
{
$marker_position = strpos($text, $marker);
2013-11-09 01:40:00 +04:00
if ($marker_position === false)
{
unset($markers[$index]);
continue;
}
2013-11-09 01:40:00 +04:00
if ($closest_marker === null or $marker_position < $closest_marker_position)
{
$closest_marker = $marker;
$closest_marker_index = $index;
$closest_marker_position = $marker_position;
}
}
2013-11-09 01:40:00 +04:00
# ~
if ($closest_marker === null or isset($text[$closest_marker_position + 1]) === false)
{
$markup .= $text;
break;
}
else
{
$markup .= substr($text, 0, $closest_marker_position);
}
$text = substr($text, $closest_marker_position);
# ~
2013-11-09 01:40:00 +04:00
unset($markers[$closest_marker_index]);
2013-11-09 01:40:00 +04:00
# ~
2013-11-09 01:40:00 +04:00
switch ($closest_marker)
{
case " \n":
2014-01-20 11:26:25 +04:00
$markup .= '<br />'."\n";
2014-01-20 11:26:25 +04:00
$offset = 3;
2014-01-20 11:26:25 +04:00
break;
2014-01-20 11:26:25 +04:00
case '![':
case '[':
if (strpos($text, ']') and preg_match('/\[((?:[^][]|(?R))*)\]/', $text, $matches))
{
$element = array(
'!' => $text[0] === '!',
'a' => $matches[1],
);
$offset = strlen($matches[0]);
if ($element['!'])
{
$offset++;
}
$remaining_text = substr($text, $offset);
if ($remaining_text[0] === '(' and preg_match('/\([ ]*(.*?)(?:[ ]+[\'"](.+?)[\'"])?[ ]*\)/', $remaining_text, $matches))
{
$element['»'] = $matches[1];
2013-07-11 00:22:16 +04:00
if (isset($matches[2]))
{
$element['#'] = $matches[2];
}
2013-12-26 23:53:48 +04:00
$offset += strlen($matches[0]);
}
elseif ($this->reference_map)
{
$reference = $element['a'];
if (preg_match('/^\s*\[(.*?)\]/', $remaining_text, $matches))
{
$reference = $matches[1] ? $matches[1] : $element['a'];
$offset += strlen($matches[0]);
}
$reference = strtolower($reference);
if (isset($this->reference_map[$reference]))
{
$element['»'] = $this->reference_map[$reference]['»'];
2013-12-26 23:53:48 +04:00
if (isset($this->reference_map[$reference]['#']))
{
$element['#'] = $this->reference_map[$reference]['#'];
}
}
else
{
unset($element);
}
}
else
{
unset($element);
}
}
if (isset($element))
{
$element['»'] = str_replace('&', '&amp;', $element['»']);
$element['»'] = str_replace('<', '&lt;', $element['»']);
if ($element['!'])
{
$markup .= '<img alt="'.$element['a'].'" src="'.$element['»'].'"';
2014-01-21 00:19:23 +04:00
if (isset($element['#']))
{
$markup .= ' title="'.$element['#'].'"';
}
2014-01-21 00:19:23 +04:00
$markup .= ' />';
}
else
{
$element['a'] = $this->parse_span_elements($element['a'], $markers);
$markup .= '<a href="'.$element['»'].'"';
2014-01-21 00:49:10 +04:00
if (isset($element['#']))
{
$markup .= ' title="'.$element['#'].'"';
}
2014-01-21 00:49:10 +04:00
$markup .= '>'.$element['a'].'</a>';
}
unset($element);
}
else
{
$markup .= $closest_marker;
2013-07-11 00:22:16 +04:00
$offset = $closest_marker === '![' ? 2 : 1;
}
break;
case '&':
if (preg_match('/^&#?\w+;/', $text, $matches))
{
$markup .= $matches[0];
$offset = strlen($matches[0]);
}
else
{
$markup .= '&amp;';
$offset = 1;
}
break;
case '*':
case '_':
2014-01-27 02:58:18 +04:00
if ($text[1] === $closest_marker and preg_match(self::$strong_regex[$closest_marker], $text, $matches))
{
$markers[] = $closest_marker;
$matches[1] = $this->parse_span_elements($matches[1], $markers);
$markup .= '<strong>'.$matches[1].'</strong>';
}
2014-01-27 02:58:18 +04:00
elseif (preg_match(self::$em_regex[$closest_marker], $text, $matches))
{
$markers[] = $closest_marker;
$matches[1] = $this->parse_span_elements($matches[1], $markers);
$markup .= '<em>'.$matches[1].'</em>';
}
if (isset($matches) and $matches)
{
$offset = strlen($matches[0]);
}
else
{
$markup .= $closest_marker;
$offset = 1;
}
break;
case '<':
if (strpos($text, '>') !== false)
{
2014-01-30 03:05:05 +04:00
if ($text[1] === 'h' and preg_match('/^<(https?:[\/]{2}[^\s]+?)>/i', $text, $matches))
{
$element_url = $matches[1];
$element_url = str_replace('&', '&amp;', $element_url);
$element_url = str_replace('<', '&lt;', $element_url);
2013-12-06 02:29:51 +04:00
$markup .= '<a href="'.$element_url.'">'.$element_url.'</a>';
$offset = strlen($matches[0]);
}
elseif (strpos($text, '@') > 1 and preg_match('/<(\S+?@\S+?)>/', $text, $matches))
{
$markup .= '<a href="mailto:'.$matches[1].'">'.$matches[1].'</a>';
2014-01-21 16:23:04 +04:00
$offset = strlen($matches[0]);
}
elseif (preg_match('/^<\/?\w.*?>/', $text, $matches))
{
$markup .= $matches[0];
$offset = strlen($matches[0]);
}
else
{
$markup .= '&lt;';
$offset = 1;
}
}
else
{
$markup .= '&lt;';
$offset = 1;
}
break;
case '\\':
2014-01-20 00:49:43 +04:00
2014-01-27 02:58:18 +04:00
if (in_array($text[1], self::$special_characters))
{
$markup .= $text[1];
2014-01-20 00:49:43 +04:00
$offset = 2;
}
else
{
$markup .= '\\';
2014-01-20 00:49:43 +04:00
$offset = 1;
}
2014-01-20 00:49:43 +04:00
break;
2014-01-20 00:49:43 +04:00
case '`':
2014-02-06 16:16:14 +04:00
if (preg_match('/^(`+)[ ]*(.+?)[ ]*(?<!`)\1(?!`)/', $text, $matches))
{
$element_text = $matches[2];
$element_text = htmlspecialchars($element_text, ENT_NOQUOTES, 'UTF-8');
2013-11-09 01:40:00 +04:00
$markup .= '<code>'.$element_text.'</code>';
$offset = strlen($matches[0]);
}
else
{
$markup .= '`';
2013-11-09 01:40:00 +04:00
$offset = 1;
}
2013-11-09 01:40:00 +04:00
break;
2013-11-09 01:40:00 +04:00
case 'http':
2013-11-09 01:40:00 +04:00
2014-02-06 02:10:18 +04:00
if (preg_match('/^https?:[\/]{2}[^\s]+\b\/*/ui', $text, $matches))
{
$element_url = $matches[0];
$element_url = str_replace('&', '&amp;', $element_url);
$element_url = str_replace('<', '&lt;', $element_url);
2013-11-09 01:40:00 +04:00
$markup .= '<a href="'.$element_url.'">'.$element_url.'</a>';
2013-11-09 01:40:00 +04:00
$offset = strlen($matches[0]);
}
else
{
$markup .= 'http';
2013-11-22 02:20:45 +04:00
$offset = 4;
}
2013-11-22 02:20:45 +04:00
break;
2013-11-09 01:40:00 +04:00
case '~~':
2013-11-09 01:40:00 +04:00
if (preg_match('/^~~(?=\S)(.+?)(?<=\S)~~/', $text, $matches))
{
$matches[1] = $this->parse_span_elements($matches[1], $markers);
2013-11-09 01:40:00 +04:00
$markup .= '<del>'.$matches[1].'</del>';
2013-11-09 01:40:00 +04:00
$offset = strlen($matches[0]);
}
else
{
$markup .= '~~';
2013-11-21 15:39:00 +04:00
$offset = 2;
}
2013-11-09 01:40:00 +04:00
break;
}
if (isset($offset))
{
$text = substr($text, $offset);
}
2013-11-09 01:40:00 +04:00
$markers[$closest_marker_index] = $closest_marker;
}
2013-11-09 01:40:00 +04:00
return $markup;
}
2014-01-18 17:10:24 +04:00
#
2014-01-26 21:53:24 +04:00
# Fields
#
private $reference_map = array();
#
# Read-only
2014-01-18 17:10:24 +04:00
2014-01-27 02:58:18 +04:00
private static $strong_regex = array(
'*' => '/^[*]{2}((?:[^*]|[*][^*]*[*])+?)[*]{2}(?![*])/s',
'_' => '/^__((?:[^_]|_[^_]*_)+?)__(?!_)/us',
);
2014-01-18 17:10:24 +04:00
2014-01-27 02:58:18 +04:00
private static $em_regex = array(
'*' => '/^[*]((?:[^*]|[*][*][^*]+?[*][*])+?)[*](?![*])/s',
'_' => '/^_((?:[^_]|__[^_]*__)+?)_(?!_)\b/us',
);
2014-01-27 02:58:18 +04:00
private static $special_characters = array(
'\\', '`', '*', '_', '{', '}', '[', ']', '(', ')', '>', '#', '+', '-', '.', '!',
);
private static $text_level_elements = array(
'a', 'br', 'bdo', 'abbr', 'blink', 'nextid', 'acronym', 'basefont',
'b', 'em', 'big', 'cite', 'small', 'spacer', 'listing',
'i', 'rp', 'sub', 'code', 'strike', 'marquee',
'q', 'rt', 'sup', 'font', 'strong',
's', 'tt', 'var', 'mark',
'u', 'xm', 'wbr', 'nobr',
'ruby',
'span',
'time',
);
}