public static function MailFormatHelper::htmlToText

public static MailFormatHelper::htmlToText($string, $allowed_tags = NULL)

Transforms an HTML string into plain text, preserving its structure.

The output will be suitable for use as 'format=flowed; delsp=yes' text (RFC 3676) and can be passed directly to MailManagerInterface::mail() for sending.

We deliberately use LF rather than CRLF, see MailManagerInterface::mail().

This function provides suitable alternatives for the following tags: <a> <em> <i> <strong> <b> <br> <p> <blockquote> <ul> <ol> <li> <dl> <dt> <dd> <h1> <h2> <h3> <h4> <h5> <h6> <hr>

Parameters

string $string: The string to be transformed.

array $allowed_tags: (optional) If supplied, a list of tags that will be transformed. If omitted, all supported tags are transformed.

Return value

string The transformed string.

File

core/lib/Drupal/Core/Mail/MailFormatHelper.php, line 104

Class

MailFormatHelper

Defines a class containing utility methods for formatting mail messages.

Namespace

Drupal\Core\Mail

Code

public static function htmlToText($string, $allowed_tags = NULL) {
  // Cache list of supported tags.
  if (empty(static::$supportedTags)) {
    static::$supportedTags = array('a', 'em', 'i', 'strong', 'b', 'br', 'p',
      'blockquote', 'ul', 'ol', 'li', 'dl', 'dt', 'dd', 'h1', 'h2', 'h3',
      'h4', 'h5', 'h6', 'hr');
  }

  // Make sure only supported tags are kept.
  $allowed_tags = isset($allowed_tags) ? array_intersect(static::$supportedTags, $allowed_tags) : static::$supportedTags;

  // Make sure tags, entities and attributes are well-formed and properly
  // nested.
  $string = Html::normalize(Xss::filter($string, $allowed_tags));

  // Apply inline styles.
  $string = preg_replace('!</?(em|i)((?> +)[^>]*)?>!i', '/', $string);
  $string = preg_replace('!</?(strong|b)((?> +)[^>]*)?>!i', '*', $string);

  // Replace inline <a> tags with the text of link and a footnote.
  // 'See <a href="https://www.drupal.org">the Drupal site</a>' becomes
  // 'See the Drupal site [1]' with the URL included as a footnote.
  static::htmlToMailUrls(NULL, TRUE);
  $pattern = '@(<a[^>]+?href="([^"]*)"[^>]*?>(.+?)</a>)@i';
  $string = preg_replace_callback($pattern, 'static::htmlToMailUrls', $string);
  $urls = static::htmlToMailUrls();
  $footnotes = '';
  if (count($urls)) {
    $footnotes .= "\n";
    for ($i = 0, $max = count($urls); $i < $max; $i++) {
      $footnotes .= '[' . ($i + 1) . '] ' . $urls[$i] . "\n";
    }
  }

  // Split tags from text.
  $split = preg_split('/<([^>]+?)>/', $string, -1, PREG_SPLIT_DELIM_CAPTURE);
  // Note: PHP ensures the array consists of alternating delimiters and
  // literals and begins and ends with a literal (inserting $null as
  // required).
  // Odd/even counter (tag or no tag).
  $tag = FALSE;
  // Case conversion function.
  $casing = NULL;
  $output = '';
  // All current indentation string chunks.
  $indent = array();
  // Array of counters for opened lists.
  $lists = array();
  foreach ($split as $value) {
    // Holds a string ready to be formatted and output.
    $chunk = NULL;

    // Process HTML tags (but don't output any literally).
    if ($tag) {
      list($tagname) = explode(' ', strtolower($value), 2);
      switch ($tagname) {
        // List counters.
        case 'ul':
          array_unshift($lists, '*');
          break;

        case 'ol':
          array_unshift($lists, 1);
          break;

        case '/ul':
        case '/ol':
          array_shift($lists);
          // Ensure blank new-line.
          $chunk = '';
          break;

          // Quotation/list markers, non-fancy headers.
        case 'blockquote':
          // Format=flowed indentation cannot be mixed with lists.
          $indent[] = count($lists) ? ' "' : '>';
          break;

        case 'li':
          $indent[] = isset($lists[0]) && is_numeric($lists[0]) ? ' ' . $lists[0]++ . ') ' : ' * ';
          break;

        case 'dd':
          $indent[] = '    ';
          break;

        case 'h3':
          $indent[] = '.... ';
          break;

        case 'h4':
          $indent[] = '.. ';
          break;

        case '/blockquote':
          if (count($lists)) {
            // Append closing quote for inline quotes (immediately).
            $output = rtrim($output, "> \n") . "\"\n";
            // Ensure blank new-line.
            $chunk = '';
          }

          // Fall-through.
        case '/li':
        case '/dd':
          array_pop($indent);
          break;

        case '/h3':
        case '/h4':
          array_pop($indent);
        case '/h5':
        case '/h6':
          // Ensure blank new-line.
          $chunk = '';
          break;

          // Fancy headers.
        case 'h1':
          $indent[] = '======== ';
          $casing = '\Drupal\Component\Utility\Unicode::strtoupper';
          break;

        case 'h2':
          $indent[] = '-------- ';
          $casing = '\Drupal\Component\Utility\Unicode::strtoupper';
          break;

        case '/h1':
        case '/h2':
          $casing = NULL;
          // Pad the line with dashes.
          $output = static::htmlToTextPad($output, ($tagname == '/h1') ? '=' : '-', ' ');
          array_pop($indent);
          // Ensure blank new-line.
          $chunk = '';
          break;

          // Horizontal rulers.
        case 'hr':
          // Insert immediately.
          $output .= static::wrapMail('', implode('', $indent)) . "\n";
          $output = static::htmlToTextPad($output, '-');
          break;

          // Paragraphs and definition lists.
        case '/p':
        case '/dl':
          // Ensure blank new-line.
          $chunk = '';
          break;
      }
    }
    // Process blocks of text.
    else {
      // Convert inline HTML text to plain text; not removing line-breaks or
      // white-space, since that breaks newlines when sanitizing plain-text.
      $value = trim(Html::decodeEntities($value));
      if (Unicode::strlen($value)) {
        $chunk = $value;
      }
    }

    // See if there is something waiting to be output.
    if (isset($chunk)) {
      // Apply any necessary case conversion.
      if (isset($casing)) {
        $chunk = call_user_func($casing, $chunk);
      }
      $line_endings = Settings::get('mail_line_endings', PHP_EOL);
      // Format it and apply the current indentation.
      $output .= static::wrapMail($chunk, implode('', $indent)) . $line_endings;
      // Remove non-quotation markers from indentation.
      $indent = array_map('\Drupal\Core\Mail\MailFormatHelper::htmlToTextClean', $indent);
    }

    $tag = !$tag;
  }

  return $output . $footnotes;
}