tags. // In turn,
tags are correctly discarded later on in this function, as social networks don't support HTML.
// However, this results in separation between paragraphs going from two newlines to one newline.
// Some social media services further drop a single newline, meaning paragraphs become one long block of text, which isn't
// intended.
remove_filter( 'the_content', 'wpautop' );
// Remove some third party Plugin filters that wrongly output content that we don't want in a status.
remove_filter( 'the_content', 'powerpress_content' );
// Apply filters to get true output.
$content = apply_filters( 'the_content', $content );
// Restore wpautop that we just removed.
add_filter( 'the_content', 'wpautop' );
// If the content originates from Gutenberg, remove double newlines and convert breaklines
// into newlines.
$is_gutenberg_request_content = $this->is_gutenberg_post_content( $post );
if ( $is_gutenberg_request_content ) {
// Remove double newlines, which may occur due to using Gutenberg blocks.
// (blocks are separated with HTML comments, stripped using apply_filters( 'the_content' ), which results in double, or even triple, breaklines).
$content = preg_replace( '/(?:(?:\r\n|\r|\n)\s*){2}/s', "\n\n", $content );
// Convert
and
into newlines.
$content = preg_replace( '/
,
, ,
and
to newlines
* - removing double spaces
* - trimming the final result of any leading or trailing spaces
*
* @since 4.6.9
*
* @param string $text Text.
* @param bool $convert_links_to_inline true: Convert e.g. `text` to `text (http://foo.com)`.
* false: Convert e.g. `text` to `text`.
* @param bool $strip_urls Whether to strip URLs from the text.
* @return string Text
*/
private function convert_to_plain_text( $text, $convert_links_to_inline = true, $strip_urls = false ) {
// Strip any shortcodes still remaining.
// If shortcodes need to be processed, they should be processed before calling this function.
$text = strip_shortcodes( $text );
// Wrap content in , and tags with an UTF-8 Content-Type meta tag.
// Forcibly tell DOMDocument that this HTML uses the UTF-8 charset.
// isn't enough, as DOMDocument still interprets the HTML as ISO-8859, which breaks character encoding
// Use of mb_convert_encoding() with HTML-ENTITIES is deprecated in PHP 8.2, so we have to use this method.
// If we don't, special characters render incorrectly.
$text = '' . $text . '';
// Load the HTML into a DOMDocument.
libxml_use_internal_errors( true );
$html = new \DOMDocument();
$html->loadHTML( $text );
// Load DOMDocument into XPath.
$xpath = new \DOMXPath( $html );
// Remove inline style tags and their contents.
foreach ( $xpath->query( '//style' ) as $node ) {
$node->parentNode->removeChild( $node ); // phpcs:ignore WordPress.NamingConventions.ValidVariableName.UsedPropertyNotSnakeCase
}
// Fetch revised HTML.
$text = $html->saveHTML();
// Remove HTML, except breaklines, links and unordered list items.
$retain_tags = ( $strip_urls ? '
and
into newlines.
$text = preg_replace( '/