jonathanbardo · December 29, 2018 20:00
diff --git a/reverse-wpautop.php b/reverse-wpautop.php
 <?php

 /**
 * Do the opposite of wpautop
 *
 * @param string $s
 *
 * @return string
 */
 public static function reverse_wpautop( $s ) {
  //remove any new lines already in there
  $s = str_replace( "\n", '', $s );

  //replace <br /> with \n
  $s = str_replace( [ '<br />', '<br>', '<br/>'], "\n", $s );

  //replace </p> with \n\n
  $s = str_replace( '</p>', "\n\n", $s );

  // Remove all characters from beginning and end of line
  return self::filter_html( $s );
 }

 /**
 * Function to remove all unwanted tags from the html and thus keeping it clean.
 *
 * @param $input
 *
 * @param string $add_to_allowed_tag
 * @return array|mixed|string
 */
 public static function filter_html( $input, $add_to_allowed_tag = '' ) {
  // God WordPress this is not very kind
  $value = wp_unslash( $input );

  // Remove empty paragraph and create
  $value = str_replace( [ '&nbsp;', '<p>&nbsp;</p>' ], [ '', '' ], $value );
  $value = str_replace( [ '<p></p>' ], [ '' ], $value );
  $value = str_replace( [ '<b>', '</b>' ], [ '<strong>', '</strong>' ], $value );
  $value = str_replace( [ '<i>', '</i>' ], [ '<em>', '</em>' ], $value );

  // Remove unallowed tags
  $allowed_tags = '<a><strong><em><ol><ul><li>' . $add_to_allowed_tag;
  $value        = strip_tags( $value, $allowed_tags );

  /**
   * Little hack to remove all attributes in an html string
   */
  $dom = new \DOMDocument;
  libxml_use_internal_errors( true );

  /**
   * This is to have DOMDocument interpret the string as UTF-8
   */
  $dom->loadHTML( '<meta http-equiv="content-type" content="text/html; charset=utf-8">' . $value );
  $xpath = new \DOMXPath( $dom );
  $nodes = $xpath->query( '//@*' );
  foreach ( $nodes as $node ) {
    $parent = $node->parentNode;

    if ( ! is_object( $parent ) ) {
      continue;
    }

    /**
     * We want to keep anchor attributes
     */
    if ( 'a' === $parent->tagName && $parent->hasAttributes() ) {
      $allowed_attrs = [
        'title',
        'target',
        'href',
      ];
      for ( $i = $parent->attributes->length - 1; $i >= 0; --$i ) {
        if ( ! in_array( $parent->attributes->item( 0 )->name, $allowed_attrs ) ) {
          $parent->removeAttributeNode( $parent->attributes->item( $i ) );
        }
      }
    } else {
      $node->parentNode->removeAttribute( $node->nodeName );
    }
  }

  /**
   * Make sure DOMDocument didn't introduce any more unallowed tags like <body>
   */
  $value = strip_tags( $dom->saveHTML(), $allowed_tags );
  libxml_clear_errors();

  return trim( $value );
 }
	<?php

	/**
	* Do the opposite of wpautop
	*
	* @param string $s
	*
	* @return string
	*/
	public static function reverse_wpautop( $s ) {
	//remove any new lines already in there
	$s = str_replace( "\n", '', $s );

	//replace <br /> with \n
	$s = str_replace( [ '<br />', '<br>', '<br/>'], "\n", $s );

	//replace </p> with \n\n
	$s = str_replace( '</p>', "\n\n", $s );

	// Remove all characters from beginning and end of line
	return self::filter_html( $s );
	}

	/**
	* Function to remove all unwanted tags from the html and thus keeping it clean.
	*
	* @param $input
	*
	* @param string $add_to_allowed_tag
	* @return array\|mixed\|string
	*/
	public static function filter_html( $input, $add_to_allowed_tag = '' ) {
	// God WordPress this is not very kind
	$value = wp_unslash( $input );

	// Remove empty paragraph and create
	$value = str_replace( [ ' ', '<p> </p>' ], [ '', '' ], $value );
	$value = str_replace( [ '<p></p>' ], [ '' ], $value );
	$value = str_replace( [ '<b>', '</b>' ], [ '<strong>', '</strong>' ], $value );
	$value = str_replace( [ '<i>', '</i>' ], [ '<em>', '</em>' ], $value );

	// Remove unallowed tags
	$allowed_tags = '<a><strong><em><ol><ul><li>' . $add_to_allowed_tag;
	$value = strip_tags( $value, $allowed_tags );

	/**
	* Little hack to remove all attributes in an html string
	*/
	$dom = new \DOMDocument;
	libxml_use_internal_errors( true );

	/**
	* This is to have DOMDocument interpret the string as UTF-8
	*/
	$dom->loadHTML( '<meta http-equiv="content-type" content="text/html; charset=utf-8">' . $value );
	$xpath = new \DOMXPath( $dom );
	$nodes = $xpath->query( '//@*' );
	foreach ( $nodes as $node ) {
	$parent = $node->parentNode;

	if ( ! is_object( $parent ) ) {
	continue;
	}

	/**
	* We want to keep anchor attributes
	*/
	if ( 'a' === $parent->tagName && $parent->hasAttributes() ) {
	$allowed_attrs = [
	'title',
	'target',
	'href',
	];
	for ( $i = $parent->attributes->length - 1; $i >= 0; --$i ) {
	if ( ! in_array( $parent->attributes->item( 0 )->name, $allowed_attrs ) ) {
	$parent->removeAttributeNode( $parent->attributes->item( $i ) );
	}
	}
	} else {
	$node->parentNode->removeAttribute( $node->nodeName );
	}
	}

	/**
	* Make sure DOMDocument didn't introduce any more unallowed tags like <body>
	*/
	$value = strip_tags( $dom->saveHTML(), $allowed_tags );
	libxml_clear_errors();

	return trim( $value );
	}