/**
* @param array $matches A set of results of the `transform` function
* @return string
*/
protected function _url_callback($matches)
{
$url = Lexer::runGamut(GamutLoader::TOOL_ALIAS . ':EncodeAttribute', $matches[1]);
Kernel::addConfig('urls', $url);
$block = Kernel::get('OutputFormatBag')->buildTag('link', $url, array('href' => $url));
return parent::hashPart($block);
}
/**
* Build `<pre><code>` blocks.
*
* @param array $matches A set of results of the `transform()` function
* @return string
*/
protected function _callback($matches)
{
$codeblock = Lexer::runGamut(GamutLoader::TOOL_ALIAS . ':Outdent', $matches[1]);
$codeblock = Helper::escapeCodeContent($codeblock);
# trim leading newlines and trailing newlines
$codeblock = preg_replace('/\\A\\n+|\\n+\\z/', '', $codeblock);
$codeblock = Kernel::get('OutputFormatBag')->buildTag('preformatted', $codeblock);
return "\n\n" . parent::hashBlock($codeblock) . "\n\n";
}
/**
* Build each blockquote block
*
* @param array $matches A set of results of the `transform()` function
* @return string
*/
protected function _callback($matches)
{
$blockq = $matches[1];
$cite = isset($matches[2]) ? $matches[2] : null;
// trim one level of quoting - trim whitespace-only lines
$blockq = preg_replace('/^[ ]*>[ ]?(\\((.+?)\\))?|^[ ]+$/m', '', $blockq);
$blockq = Lexer::runGamut('html_block_gamut', $blockq);
# recurse
$blockq = preg_replace('/^/m', " ", $blockq);
// These leading spaces cause problem with <pre> content,
// so we need to fix that:
$blockq = preg_replace_callback('{(\\s*<pre>.+?</pre>)}sx', array($this, '_callback_spaces'), $blockq);
$attributes = array();
if (!empty($cite)) {
$attributes['cite'] = $cite;
}
$block = Kernel::get('OutputFormatBag')->buildTag('blockquote', $blockq, $attributes);
return "\n" . parent::hashBlock($block) . "\n\n";
}
/**
* Process ATX-style headers
*
* @param array $matches The results from the `transform()` function
* @return string
*/
protected function _atx_callback($matches)
{
$level = strlen($matches[1]) + $this->_getRebasedHeaderLevel();
$domid = !empty($matches[3]) ? $matches[3] : Helper::header2Label($matches[2]);
$domid = Kernel::get('DomId')->set($domid);
$title = Lexer::runGamut('span_gamut', $matches[2]);
Kernel::addConfig('menu', array('level' => $level, 'text' => parent::unhash($title)), $domid);
$block = Kernel::get('OutputFormatBag')->buildTag('title', $title, array('level' => $level, 'id' => $domid));
$this->_setContentTitle($title);
return "\n" . parent::hashBlock($block) . "\n\n";
}
/**
* @param array $matches A set of results of the `transform` function
* @return string
*/
protected function _inline_callback($matches)
{
$alt_text = $matches[2];
$url = $matches[3] == '' ? $matches[4] : $matches[3];
$title =& $matches[7];
$attributes = array();
$attributes['alt'] = Lexer::runGamut(GamutLoader::TOOL_ALIAS . ':EncodeAttribute', $alt_text);
$attributes['src'] = Lexer::runGamut(GamutLoader::TOOL_ALIAS . ':EncodeAttribute', $url);
if (!empty($title)) {
$attributes['title'] = Lexer::runGamut(GamutLoader::TOOL_ALIAS . ':EncodeAttribute', $title);
}
$block = Kernel::get('OutputFormatBag')->buildTag('image', null, $attributes);
return parent::hashPart($block);
}
public function buildFootnoteStandardLink($text = null, array $attributes = array(), $note_type = Note::FOOTNOTE_DEFAULT)
{
$type_info = Note::getTypeInfo($note_type);
if ($this->getConfig($type_info['prefix'] . '_link_class')) {
$attributes['class'] = Helper::fillPlaceholders(Lexer::runGamut(GamutLoader::TOOL_ALIAS . ':EncodeAttribute', $this->getConfig($type_info['prefix'] . '_link_class')), $text);
}
if ($this->getConfig($type_info['prefix'] . '_link_title_mask')) {
$attributes['title'] = Helper::fillPlaceholders(Lexer::runGamut(GamutLoader::TOOL_ALIAS . ':EncodeAttribute', $this->getConfig($type_info['prefix'] . '_link_title_mask')), $text);
}
$backlink_id = $attributes['backlink_id'];
unset($attributes['backlink_id']);
unset($attributes['counter']);
$link = Kernel::get('OutputFormatBag')->buildTag('link', $text, $attributes);
return Kernel::get('OutputFormatBag')->buildTag('sup', $link, array('id' => $backlink_id));
}
/**
* Process the dd contents.
*
* @param array $matches
* @return string
*/
protected function _item_callback_dd($matches)
{
$leading_line = $matches[1];
$marker_space = $matches[2];
$def = $matches[3];
if ($leading_line || preg_match('/\\n{2,}/', $def)) {
// Replace marker with the appropriate whitespace indentation
$def = str_repeat(' ', strlen($marker_space)) . $def;
$def = Lexer::runGamut('html_block_gamut', Lexer::runGamut(GamutLoader::TOOL_ALIAS . ':Outdent', $def . "\n\n"));
// $def = "\n$def\n";
} else {
$def = rtrim($def);
$def = Lexer::runGamut('span_gamut', Lexer::runGamut(GamutLoader::TOOL_ALIAS . ':Outdent', $def));
}
return Kernel::get('OutputFormatBag')->buildTag('definition_list_item_definition', $def);
}
/**
* @param array $matches A set of results of the `transform()` function
* @return string
*/
protected function _items_callback($matches)
{
$item = $matches[4];
$leading_line =& $matches[1];
$leading_space =& $matches[2];
$marker_space = $matches[3];
$trailing_blank_line =& $matches[5];
if ($leading_line || $trailing_blank_line || preg_match('/\\n{2,}/', $item)) {
// Replace marker with the appropriate whitespace indentation
$item = $leading_space . str_repeat(' ', strlen($marker_space)) . $item;
$item = Lexer::runGamut('html_block_gamut', Lexer::runGamut(GamutLoader::TOOL_ALIAS . ':Outdent', $item) . "\n");
} else {
// Recursion for sub-lists:
$item = self::transform(Lexer::runGamut(GamutLoader::TOOL_ALIAS . ':Outdent', $item));
$item = preg_replace('/\\n+$/', '', $item);
$item = Lexer::runGamut('span_gamut', $item);
}
return Kernel::get('OutputFormatBag')->buildTag('list_item', $item);
}
/**
* Do a footnote transformation
*
* @param string $note_id
* @param string $note_content
* @param int $type
* @return void
*/
protected function _doTransformNote($note_id, $note_content, $type = self::FOOTNOTE_DEFAULT)
{
if (!empty($note_content)) {
++self::$notes_counter;
$type_info = $this->getTypeInfo($type);
$note_content .= "\n";
// Need to append newline before parsing.
$note_content = Lexer::runGamut('html_block_gamut', $note_content . "\n");
$note_content = preg_replace_callback('{F\\x1Afn:(.*?)\\x1A:}', array($this, '_append_callback'), $note_content);
$note_id = Lexer::runGamut(GamutLoader::TOOL_ALIAS . ':EncodeAttribute', $note_id);
$backlink_id = Kernel::get('DomId')->get($type_info['prefix'] . 'ref:' . $note_id);
$footlink_id = Kernel::get('DomId')->get($type_info['prefix'] . ':' . $note_id);
$attributes = array();
$attributes['rev'] = $type_info['name'];
$attributes['counter'] = self::$notes_counter;
$attributes['href'] = '#' . $backlink_id;
$note = array('count' => self::$notes_counter, 'type' => $type_info['name'], 'in-text-id' => $backlink_id, 'note-id' => $footlink_id, 'text' => Kernel::get('OutputFormatBag')->buildTag($type_info['outputformat_methods']['item'], $note_content, $attributes));
Kernel::get(Kernel::TYPE_CONTENT)->addNote($note, $note_id);
}
}
/**
* Form HTML tables: parses table contents
*
* @param array $matches
* @return string
*/
protected function _callback($matches)
{
$attributes = array();
//self::doDebug($matches);
// The head string may have a begin slash
$caption = count($matches) > 3 ? $matches[1] : null;
$head = count($matches) > 3 ? preg_replace('/^ *[|]/m', '', $matches[2]) : preg_replace('/^ *[|]/m', '', $matches[1]);
$underline = count($matches) > 3 ? $matches[3] : $matches[2];
$content = count($matches) > 3 ? preg_replace('/^ *[|]/m', '', $matches[4]) : preg_replace('/^ *[|]/m', '', $matches[3]);
// Remove any tailing pipes for each line.
$underline = preg_replace('/[|] *$/m', '', $underline);
$content = preg_replace('/[|] *$/m', '', $content);
// Reading alignement from header underline.
$separators = preg_split('/ *[|] */', $underline);
foreach ($separators as $n => $s) {
$attributes[$n] = array();
if (preg_match('/^ *-+: *$/', $s)) {
$attributes[$n]['style'] = 'text-align:right;';
} elseif (preg_match('/^ *:-+: *$/', $s)) {
$attributes[$n]['style'] = 'text-align:center;';
} elseif (preg_match('/^ *:-+ *$/', $s)) {
$attributes[$n]['style'] = 'text-align:left;';
}
}
// Split content by row.
$headers = explode("\n", trim($head, "\n"));
$text = '';
if (!empty($caption)) {
$this->table_id = Helper::header2Label($caption);
$text .= preg_replace_callback('/\\[(.*)\\]/', array($this, '_doCaption'), Lexer::runGamut('span_gamut', $caption));
}
$lines = '';
foreach ($headers as $_header) {
$line = '';
// Parsing span elements, including code spans, character escapes,
// and inline HTML tags, so that pipes inside those gets ignored.
$_header = Lexer::runGamut('filter:Span', $_header);
// Split row by cell.
$_header = preg_replace('/[|] *$/m', '', $_header);
$_headers = preg_split('/[|]/', $_header);
$col_count = count($_headers);
// Write column headers.
// we first loop for colspans
$headspans = array();
foreach ($_headers as $_i => $_cell) {
if ($_cell == '') {
if ($_i == 0) {
$headspans[1] = 2;
} else {
if (isset($headspans[$_i - 1])) {
$headspans[$_i - 1]++;
} else {
$headspans[$_i - 1] = 2;
}
}
}
}
foreach ($_headers as $n => $__header) {
if ($__header != '') {
$cell_attributes = $attributes[$n];
if (isset($headspans[$n])) {
$cell_attributes['colspan'] = $headspans[$n];
}
$line .= Kernel::get('OutputFormatBag')->buildTag('table_cell_head', Lexer::runGamut('span_gamut', trim($__header)), $cell_attributes) . "\n";
}
}
$lines .= Kernel::get('OutputFormatBag')->buildTag('table_line', $line) . "\n";
}
$text .= Kernel::get('OutputFormatBag')->buildTag('table_header', $lines);
// Split content by row.
$rows = explode("\n", trim($content, "\n"));
$lines = '';
foreach ($rows as $row) {
$line = '';
// Parsing span elements, including code spans, character escapes,
// and inline HTML tags, so that pipes inside those gets ignored.
$row = Lexer::runGamut('filter:Span', $row);
// Split row by cell.
$row_cells = preg_split('/ *[|] */', $row, $col_count);
$row_cells = array_pad($row_cells, $col_count, '');
// we first loop for colspans
$colspans = array();
foreach ($row_cells as $_i => $_cell) {
if ($_cell == '') {
if ($_i == 0) {
$colspans[1] = 2;
} else {
if (isset($colspans[$_i - 1])) {
$colspans[$_i - 1]++;
} else {
$colspans[$_i - 1] = 2;
}
}
}
}
foreach ($row_cells as $n => $cell) {
if ($cell != '') {
$cell_attributes = $attributes[$n];
if (isset($colspans[$n])) {
$cell_attributes['colspan'] = $colspans[$n];
}
$line .= Kernel::get('OutputFormatBag')->buildTag('table_cell', Lexer::runGamut('span_gamut', trim($cell)), $cell_attributes) . "\n";
}
}
$lines .= Kernel::get('OutputFormatBag')->buildTag('table_line', $line) . "\n";
}
$text .= Kernel::get('OutputFormatBag')->buildTag('table_body', $lines);
$table = Kernel::get('OutputFormatBag')->buildTag('table', $text);
return parent::hashBlock($table) . "\n";
}
/**
* Handle $token provided by parseSpan by determining its nature and
* returning the corresponding value that should replace it.
*
* @param string $token
* @param string $str
* @return string
*/
public function handleSpanToken($token, &$str)
{
switch ($token[0]) {
case "\\":
if ($token[1] == "(") {
$texend = strpos($str, '\\)');
if ($texend) {
$eqn = substr($str, 0, $texend);
$str = substr($str, $texend + 2);
$texspan = Lexer::runGamut('filter:Maths:span', $eqn);
return parent::hashPart($texspan);
} else {
return $str;
}
} else {
return parent::hashPart("&#" . ord($token[1]) . ";");
}
case "`":
// Search for end marker in remaining text.
if (preg_match('/^(.*?[^`])' . preg_quote($token) . '(?!`)(.*)$/sm', $str, $matches)) {
$str = $matches[2];
$codespan = Lexer::runGamut('filter:CodeBlock:span', $matches[1], true);
return parent::hashPart($codespan);
}
return $token;
// return as text since no ending marker found.
// return as text since no ending marker found.
default:
return parent::hashPart($token);
}
}
/**
* Process paragraphs
*
* @param string $text The text to parse
*
* @return string The text parsed
*/
public function RebuildParagraph($text)
{
// Strip leading and trailing lines:
$text = preg_replace('/\\A\\n+|\\n+\\z/', '', $text);
$grafs = preg_split('/\\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY);
// Wrap <p> tags and unhashify HTML blocks
foreach ($grafs as $key => $value) {
$value = trim(Lexer::runGamut('span_gamut', $value));
// Check if this should be enclosed in a paragraph.
// Clean tag hashes & block tag hashes are left alone.
$is_p = !preg_match('/^B\\x1A[0-9]+B|^C\\x1A[0-9]+C$/', $value);
if ($is_p) {
$value = Kernel::get('OutputFormatBag')->buildTag('paragraph', $value);
}
$grafs[$key] = $value;
}
// Join grafs in one text, then unhash HTML tags.
// $text = implode("\n\n", $grafs);
$text = implode('', $grafs);
// Finish by removing any tag hashes still present in $text.
$text = Lexer::runGamut('filter:HTML:unhash', $text, true);
return $text;
}
/**
* @param string $text
* @return string
*/
public function transform($text)
{
$token_stack = array('');
$text_stack = array('');
$italic = '';
$strong = '';
$tree_char_em = false;
while (1) {
// Get prepared regular expression for seraching emphasis tokens in current context.
$token_re = self::$em_strong_prepared["{$italic}{$strong}"];
// Each loop iteration search 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);
$span = Lexer::runGamut('span_gamut', array_shift($text_stack));
$span = Kernel::get('OutputFormatBag')->buildTag('italic', $span);
$span = Kernel::get('OutputFormatBag')->buildTag('bold', $span);
$text_stack[0] .= parent::hashPart($span);
$italic = '';
$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 ? "bold" : "italic";
$span = Lexer::runGamut('span_gamut', $text_stack[0]);
$span = Kernel::get('OutputFormatBag')->buildTag($tag, $span);
$text_stack[0] = parent::hashPart($span);
${$tag} = '';
// $$tag stands for $italic or $strong
}
$tree_char_em = false;
} elseif ($token_len == 3) {
if ($italic) {
// Reached closing marker for both em and strong.
// Closing strong marker:
for ($i = 0; $i < 2; ++$i) {
$shifted_token = array_shift($token_stack);
$tag = strlen($shifted_token) == 2 ? "bold" : "italic";
$span = Lexer::runGamut('span_gamut', array_shift($text_stack));
$span = Kernel::get('OutputFormatBag')->buildTag($tag, $span);
$text_stack[0] .= parent::hashPart($span);
${$tag} = '';
// $$tag stands for $italic or $strong
}
} else {
// Reached opening three-char emphasis marker. Push on token
// stack; will be handled by the special condition above.
$italic = $token[0];
$strong = "{$italic}{$italic}";
array_unshift($token_stack, $token);
array_unshift($text_stack, '');
$tree_char_em = true;
}
} elseif ($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);
$span = Lexer::runGamut('span_gamut', array_shift($text_stack));
$span = Kernel::get('OutputFormatBag')->buildTag('bold', $span);
$text_stack[0] .= parent::hashPart($span);
$strong = '';
} else {
array_unshift($token_stack, $token);
array_unshift($text_stack, '');
$strong = $token;
}
} else {
// Here $token_len == 1
if ($italic) {
if (strlen($token_stack[0]) == 1) {
// Closing emphasis marker:
array_shift($token_stack);
$span = Lexer::runGamut('span_gamut', array_shift($text_stack));
$span = Kernel::get('OutputFormatBag')->buildTag('italic', $span);
$text_stack[0] .= parent::hashPart($span);
$italic = '';
} else {
$text_stack[0] .= $token;
}
} else {
array_unshift($token_stack, $token);
array_unshift($text_stack, '');
$italic = $token;
}
}
}
return $text_stack[0];
}
/**
* Process each abbreviation
*
* @param array $matches One set of results form the `transform()` function
* @return string
*/
protected function _callback($matches)
{
$abbr = $matches[0];
$abbr_desciptions = Kernel::getConfig('abbr_desciptions');
if (isset($abbr_desciptions[$abbr])) {
$attributes = array();
$desc = trim($abbr_desciptions[$abbr]);
if (!empty($desc)) {
$attributes['title'] = Lexer::runGamut(GamutLoader::TOOL_ALIAS . ':EncodeAttribute', $desc);
}
$abbr = Kernel::get('OutputFormatBag')->buildTag('abbreviation', $abbr, $attributes);
return parent::hashBlock($abbr);
} else {
return $abbr;
}
}
