/**
* This function extracts headings from the html formatted $content. It will pull out
* only the required headings as specified in the options. For all qualifying headings,
* this function populates the $find and $replace arrays (both passed by reference)
* with what to search and replace with.
*
* Returns a HTML formatted string of list items for each qualifying heading. This
* is everything between and NOT including <ul> and </ul>
*
* @access private
* @since 1.0
* @static
*
* @param array $find
* @param array $replace
* @param string $content
*
* @return bool|string
*/
public static function extract_headings(&$find, &$replace, $content = '')
{
global $wp_query;
$post = $wp_query->post;
$matches = array();
$anchor = '';
$items = '';
$headings = get_post_meta($post->ID, '_ez-toc-heading-levels', TRUE);
$exclude = get_post_meta($post->ID, '_ez-toc-exclude', TRUE);
$altText = get_post_meta($post->ID, '_ez-toc-alttext', TRUE);
if (!is_array($headings)) {
$headings = array();
}
if (empty($headings)) {
$headings = ezTOC_Option::get('heading_levels', array());
}
if (empty($exclude)) {
$exclude = ezTOC_Option::get('exclude');
}
// reset the internal collision collection as the_content may have been triggered elsewhere
// eg by themes or other plugins that need to read in content such as metadata fields in
// the head html tag, or to provide descriptions to twitter/facebook
self::$collision_collector = array();
if (is_array($find) && is_array($replace) && $content) {
// get all headings
// the html spec allows for a maximum of 6 heading depths
if (preg_match_all('/(<h([1-6]{1})[^>]*>).*<\\/h\\2>/msuU', $content, $matches, PREG_SET_ORDER)) {
// remove undesired headings (if any) as defined by heading_levels
if (count($headings) != 6) {
$new_matches = array();
for ($i = 0; $i < count($matches); $i++) {
if (in_array($matches[$i][2], $headings)) {
$new_matches[] = $matches[$i];
}
}
$matches = $new_matches;
}
// remove specific headings if provided via the 'exclude' property
if ($exclude) {
$excluded_headings = explode('|', $exclude);
$excluded_count = count($excluded_headings);
if ($excluded_count > 0) {
for ($j = 0; $j < $excluded_count; $j++) {
$excluded_headings[$j] = preg_quote($excluded_headings[$j]);
// escape some regular expression characters
// others: http://www.php.net/manual/en/regexp.reference.meta.php
$excluded_headings[$j] = str_replace(array('\\*'), array('.*'), trim($excluded_headings[$j]));
}
$new_matches = array();
for ($i = 0; $i < count($matches); $i++) {
$found = FALSE;
for ($j = 0; $j < $excluded_count; $j++) {
if (@preg_match('/^' . $excluded_headings[$j] . '$/imU', strip_tags($matches[$i][0]))) {
$found = TRUE;
break;
}
}
if (!$found) {
$new_matches[] = $matches[$i];
}
}
if (count($matches) != count($new_matches)) {
$matches = $new_matches;
}
}
}
// remove empty headings
$new_matches = array();
for ($i = 0; $i < count($matches); $i++) {
if (trim(strip_tags($matches[$i][0])) != FALSE) {
$new_matches[] = $matches[$i];
}
}
if (count($matches) != count($new_matches)) {
$matches = $new_matches;
}
$toc = $matches;
// Replace headers with toc alt text.
if ($altText) {
$alt_headings = array();
$split_headings = preg_split('/\\r\\n|[\\r\\n]/', $altText);
$split_headings_count = count($split_headings);
if ($split_headings) {
for ($k = 0; $k < $split_headings_count; $k++) {
$explode_headings = explode('|', $split_headings[$k]);
if (0 < strlen($explode_headings[0]) && 0 < strlen($explode_headings[1])) {
$alt_headings[$explode_headings[0]] = $explode_headings[1];
}
}
}
if (0 < count($alt_headings)) {
for ($i = 0; $i < count($toc); $i++) {
foreach ($alt_headings as $original_heading => $alt_heading) {
$original_heading = preg_quote($original_heading);
// escape some regular expression characters
// others: http://www.php.net/manual/en/regexp.reference.meta.php
$original_heading = str_replace(array('\\*'), array('.*'), trim($original_heading));
if (@preg_match('/^' . $original_heading . '$/imU', strip_tags($toc[$i][0]))) {
//$matches[ $i ][0] = str_replace( $original_heading, $alt_heading, $matches[ $i ][0] );
$toc[$i][0] = $alt_heading;
}
}
}
}
}
// check minimum number of headings
if (count($matches) >= ezTOC_Option::get('start')) {
for ($i = 0; $i < count($matches); $i++) {
// get anchor and add to find and replace arrays
$anchor = isset($toc[$i][0]) ? self::url_anchor_target($toc[$i][0]) : self::url_anchor_target($matches[$i][0]);
$find[] = $matches[$i][0];
$replace[] = str_replace(array($matches[$i][1], '</h' . $matches[$i][2] . '>'), array($matches[$i][1] . '<span class="ez-toc-section" id="' . $anchor . '">', '</span></h' . $matches[$i][2] . '>'), $matches[$i][0]);
// assemble flat list
if (!ezTOC_Option::get('show_hierarchy')) {
$items .= '<li><a href="' . esc_url('#' . $anchor) . '">';
//$title = apply_filters( 'ez_toc_title', strip_tags( wp_kses_post( $toc[ $i ][0] ) ) );
$title = strip_tags(apply_filters('ez_toc_title', $toc[$i][0]));
//if ( 'decimal' == ezTOC_Option::get( 'counter' ) ) {
//
// $items .= count( $replace ) . ' ';
//}
$items .= $title . '</a></li>';
}
}
// build a hierarchical toc?
// we could have tested for $items but that var can be quite large in some cases
if (ezTOC_Option::get('show_hierarchy')) {
$items = self::build_hierarchy($toc);
}
}
}
}
return $items;
}