function any_wiki_towiki ( $html, $options = false )
{
global $_wiki_xhtml_parser;
$html = wiki_normalize_newlines($html);
$html = "\n$html\n\n\n";
// html block (escapes all that is within)
$html = preg_replace_callback('|<!--\[html\]-->(.*?)<!--\[/html]-->|s', '_towiki_html', $html);
// Simple markup
$html = preg_replace('!</?(i|em)>!', '//', $html);
$html = preg_replace('!</?(b|strong)>!', '**', $html);
$html = preg_replace('|<sup><a|', '<a', $html);
$html = preg_replace('|</a></sup>|', '</a>', $html);
$html = preg_replace('|</?sup>|', '^^', $html);
$html = preg_replace('|<tt>|', '{{', $html);
$html = preg_replace('|</tt>|', '}}', $html);
// coloured texts
$html = preg_replace('|<span style="color: #?([^"]*);">(.*?)</span>|s', '##\1|\2##', $html);
// mailto links
$html = preg_replace('|<!--\[mailto\]-->([^ ]*?)<!--\[/mailto\]-->|', 'mailto:\1', $html);
$html = preg_replace('|<!--\[mailto\]-->(.*?)<!--\[/mailto\]-->|', '[mailto:\1]', $html);
// images
$html = preg_replace('/<!--(\[image .*?\])-->/', '[\1]', $html);
// literal texts
$html = preg_replace('|<!--\[lit\]-->(.*?)<!--\[/lit]-->|s', '``\1``', $html);
// comments
$html = preg_replace('|<!-- (.*?) -->|s', '<x-comment>\1</x-comment>', $html);
// toc
if (strpos($html, '<!--[toc]-->') !== false)
{
// The [[toc]] tag
$html = preg_replace('|<!--\[toc\]-->.*?<!--\[/toc]-->|s', "\n[[toc]]\n", $html);
// The toc anchors in the headers
$html = preg_replace('|(<h[2-6]>)<a name=\'[0-9]+-[0-9]+\'></a>|', '\1', $html);
}
//
// Make sure that the XML parser gets legal utf8 text
$html = any_text_utf8($html);
// Parse the remaining text
if (!isset($_wiki_xhtml_parser))
{
$_wiki_xhtml_parser = new XML_Wiki_Parser();
}
$_wiki_xhtml_parser->reset();
$ret = $_wiki_xhtml_parser->parseString("<wiki>\n" . $html . "\n</wiki>", true);
if ($_wiki_xhtml_parser->isError($ret))
{
// Error in XML - just return the text... sorry...
$html = $ret->getMessage() . "\n\n\n" . strip_tags($html);
}
else
{
$html = $_wiki_xhtml_parser->wiki;
}
$_wiki_xhtml_parser->reset();
// $_wiki_xhtml_parser->free();
// Correct <del/><ins/> pairs
$html = preg_replace('|(@@---.*?)@@@@(\+\+\+.*?@@)|', '\1\2', $html);
// Correct multiple newlines
$html = trim(preg_replace("/\n\n\n*/", "\n\n", $html)) . "\n";
// Remove too many spaces at linestarts in blockquotes
$html = preg_replace("/\n(>+)[ \t]+/", "\n\\1 ", $html);
// When a line is requested, remove superfluous <p/>'s
if (!empty($options['target']) && $options['target'] == 'line')
{
$html = str_replace(array("\n", "<p>", "</p>"), array(' ', '', ''), $html);
}
return $html;
}
function wiki_tokenizer ( $s, $options = array() )
{
$s = wiki_normalize_newlines($s) . "\n\n";
$i = 0; // the offset of the scanner
$line_offs = 0; // the token offset in the current line
$len = strlen($s); // the length of the input stream
$tk = array(); // the token list returned
$tk_s = array(); // token strings
// Get the settings
$allow_wikiword = !empty($options['allow_wikiword']);
// Translate the character stream into tokens, use the ending "\n" as a buffer.
while ($i < $len-2)
{
$c = $s{$i};
$n_c = $s{$i+1};
$nn_c = $s{$i+2};
$line_offs++;
switch ($c)
{
case "\n":
if ($n_c == "\n")
{
while ($i < $len - 1 && $s{$i+1} == "\n")
{
$i++;
}
$tk[] = "p";
$tk_s[] = "\n\n";
}
else
{
$tk[] = "newline";
$tk_s[] = "\n";
}
$line_offs = 0;
break;
case ' ':
if ($n_c == '_' && $nn_c == "\n")
{
$tk[] = 'br';
$tk_s[] = " _\n";
$i += 2;
}
else
{
$tok = ' ';
while ($s{$i+1} == ' ')
{
$tok .= ' ';
$i++;
}
if ( $s{$i+1} == '_'
&& $s{$i+2} == "\n")
{
$tk[] = 'br';
$tk_s[] = " _\n";
$i += 2;
}
else
{
$tk[] = ' ';
$tk_s[] = $tok;
}
}
break;
case '`':
if ($n_c == '`')
{
$j = $i+2;
$tok = '';
while ( $j < $len - 2
&& ($s{$j} != '`' || $s{$j+1} != '`'))
{
$tok .= $s{$j};
$j++;
}
if ($s{$j} == '`' && $s{$j+1} == '`')
{
$tk[] = 'literal';
$tk_s[] = str_replace("\n", " ", $tok);
$i = $j+1;
}
else
{
$tk[] = '`';
$tk_s[] = '`';
}
}
else
{
$tk[] = '`';
$tk_s[] = '`';
}
break;
case '<':
// Check for <html> on one line
if ( $line_offs == 1
&& substr($s, $i, 7) == "<html>\n"
&& ($end = strpos($s, "\n</html>\n", $i+5)) !== false)
{
$tk[] = 'html';
$tk_s[] = substr($s, $i+7, $end - ($i+6));
$i = $end + 8;
}
// Check for <code> on one line
else if ( $line_offs == 1
&& substr($s, $i, 7) == "<code>\n"
&& ($end = strpos($s, "\n</code>\n", $i+5)) !== false)
{
$tk[] = 'code';
$tk_s[] = substr($s, $i+7, $end - ($i+6));
$i = $end + 8;
}
// Check for a <!-- ... --> block
else if ( substr($s, $i, 4) == '<!--'
&& ($end = strpos($s, '-->', $i+4)) !== false)
{
$tk[] = 'comment';
$tk_s[] = trim(substr($s, $i+4, $end - ($i+4)));
$i = $end + 2;
}
else
{
$tk[] = '<';
$tk_s[] = '<';
}
break;
case '/':
if ($n_c == '/')
{
$tk[] = "em";
$tk_s[] = "//";
$i+=1;
}
else
{
$tk[] = $c;
$tk_s[] = $c;
}
break;
case '*':
if ($n_c == '*')
{
$tk[] = "strong";
$tk_s[] = "**";
$i+=1;
}
else
{
$tk[] = $c;
$tk_s[] = $c;
}
break;
case '^':
if ($n_c == '^')
{
$tk[] = "sup";
$tk_s[] = "^^";
$i++;
}
else
{
$tk[] = $c;
$tk_s[] = $c;
}
break;
case '@':
case '#':
case '(':
case ')':
case '|':
case '[':
case ']':
case '{':
case '}':
if ($c == '[' & $n_c == '[')
{
// check for block-level [[toc]]
if ( $line_offs == 1
&& substr($s, $i, 8) == "[[toc]]\n")
{
$tk[] = 'toc';
$tk_s[] = '[[toc]]';
$i += 6;
}
else
{
$tk[] = $c.$c;
$tk_s[] = $c.$c;
$i++;
}
}
else if ($n_c == $c)
{
$tk[] = $c.$c;
$tk_s[] = $c.$c;
$i++;
}
else
{
$tk[] = $c;
$tk_s[] = $c;
}
break;
case '>':
$tok = '>';
while ($s{$i+1} == '>')
{
$tok .= '>';
$i++;
}
$tk[] = ">";
$tk_s[] = $tok;
break;
case '\'':
if ($n_c == '\'' && $nn_c == '\'')
{
$tk[] = "strong";
$tk_s[] = "'''";
$i+=2;
}
else if ($n_c == '\'')
{
$tk[] = "em";
$tk_s[] = "''";
$i+=1;
}
else
{
$tk[] = $c;
$tk_s[] = $c;
}
break;
case ':':
if ($n_c == '/' && $nn_c == '/')
{
$tk[] = '://';
$tk_s[] = '://';
$i += 2;
}
else
{
$tk[] = ':';
$tk_s[] = ':';
}
break;
default:
$class = _charclass($c);
$tok = $c;
$j = $i;
while ($class == _charclass($s{$j+1}) && $j < $len - 2)
{
$j++;
$tok .= $s{$j};
}
if ($class == 'word')
{
if ( (($tok == 'http' || $tok == 'https') && substr($s, $j+1, 3) == '://')
|| ($tok == 'mailto' && $s[$j+1] == ':'))
{
// http:// or mailto: -- fetch till whitespace or one of "])|>"
if ($tok == 'mailto')
{
$class = 'mailto';
}
else
{
$class = 'url';
}
while (strpos("\n\t |[](){}<>\"'", $s{$j+1}) === false)
{
$j++;
$tok .= $s{$j};
}
}
else if ( $allow_wikiword
&& $c >= 'A'
&& $c <= 'Z'
&& preg_match('/^[A-Z][a-z0-9_]+[A-Z][a-zA-Z0-9_]*$/', $tok))
{
$class = "wiki-word";
}
}
$tk[] = $class;
$tk_s[] = $tok;
$i = $j;
break;
}
$i++;
}
$tk[] = 'end';
$tk_s[] = '';
return array($tk, $tk_s);
}
