public function more($content, $post)
{
$more_text = 'Read the rest →';
$max_paragraphs = 1;
$showmore = false;
$matches = preg_split('/<!--\\s*more\\s*-->/is', $content, 2, PREG_SPLIT_NO_EMPTY);
if (count($matches) > 1) {
$summary = $matches[0];
$remainder = $matches[1];
if (trim($remainder) != '') {
$showmore = true;
}
} else {
$ht = new HtmlTokenizer($content, false);
$set = $ht->parse();
$stack = array();
$para = 0;
$token = $set->current();
$summary = new HTMLTokenSet(false);
$remainder = new HTMLTokenSet(false);
$set->rewind();
for ($token = $set->current(); $set->valid(); $token = $set->next()) {
if ($token['type'] == HTMLTokenizer::NODE_TYPE_ELEMENT_OPEN) {
$stack[$token['name']] = $token['name'];
}
if ($para < $max_paragraphs) {
$summary[] = $token;
}
if ($para >= $max_paragraphs) {
$remainder[] = $token;
$showmore = true;
}
if ($token['type'] == HTMLTokenizer::NODE_TYPE_ELEMENT_CLOSE) {
if (isset($stack[$token['name']])) {
while (end($stack) != $token['name']) {
array_pop($stack);
}
array_pop($stack);
}
if (count($stack) == 0) {
$para++;
}
}
}
}
if ($post->slug == Controller::get_var('slug')) {
$content = $summary . '<div id="more" class="moreanchor">' . 'Continues here →' . '</div>' . $remainder;
} elseif ($showmore == true) {
$content = $summary . '<p class="more"><a href="' . $post->permalink . '#more">' . $more_text . '</a></p>';
} else {
$content = $summary . $remainder;
}
return $content;
}
/**
* Returns a shortened version of whatever is passed in.
* @param string $value A string to shorten
* @param integer $count Maximum words to display [100]
* @param integer $max_paragraphs Maximum paragraphs to display [1]
* @return string The string, shortened
*/
public static function summarize( $text, $count = 100, $max_paragraphs = 1 )
{
$ellipsis = '…';
$showmore = false;
$ht = new HtmlTokenizer($text, false);
$set = $ht->parse();
$stack = array();
$para = 0;
$token = $set->current();
$summary = new HTMLTokenSet();
$set->rewind();
$remaining_words = $count;
// $bail lets the loop end naturally and close all open elements without adding new ones.
$bail = false;
for ( $token = $set->current(); $set->valid(); $token = $set->next() ) {
if ( !$bail && $token['type'] == HTMLTokenizer::NODE_TYPE_ELEMENT_OPEN ) {
$stack[] = $token;
}
if ( !$bail ) {
switch ( $token['type'] ) {
case HTMLTokenizer::NODE_TYPE_TEXT:
$words = preg_split( '/(\\s+)/u', $token['value'], -1, PREG_SPLIT_DELIM_CAPTURE | PREG_SPLIT_NO_EMPTY );
// word count is doubled because spaces between words are captured as their own array elements via PREG_SPLIT_DELIM_CAPTURE
$words = array_slice( $words, 0, $remaining_words * 2 );
$remaining_words -= count( $words ) / 2;
$token['value'] = implode( '', $words );
if ( $remaining_words <= 0 ) {
$token['value'] .= $ellipsis;
$summary[] = $token;
$bail = true;
}
else {
$summary[] = $token;
}
break;
case HTMLTokenizer::NODE_TYPE_ELEMENT_CLOSE;
// don't handle this case here
break;
default:
$summary[] = $token;
break;
}
}
if ( $token['type'] == HTMLTokenizer::NODE_TYPE_ELEMENT_CLOSE ) {
do {
$end = array_pop( $stack );
$end['type'] = HTMLTokenizer::NODE_TYPE_ELEMENT_CLOSE;
$end['attrs'] = null;
$end['value'] = null;
$summary[] = $end;
} while ( ( $bail || $end['name'] != $token['name'] ) && count( $stack ) > 0 );
if ( count( $stack ) == 0 ) {
$para++;
}
if ( $bail || $para >= $max_paragraphs ) {
break;
}
}
}
return (string) $summary;
}
