filter.module

<?php


/**
 * @file
 * Framework for handling the filtering of content.
 */
use Drupal\Core\Url;
use Drupal\Component\Utility\Html;
use Drupal\Component\Utility\Unicode;
use Drupal\Core\Cache\Cache;
use Drupal\Core\Routing\RouteMatchInterface;
use Drupal\Core\Session\AccountInterface;
use Drupal\Core\Template\Attribute;
use Drupal\filter\FilterFormatInterface;

/**
 * Implements hook_help().
 */
function filter_help($route_name, RouteMatchInterface $route_match) {
    switch ($route_name) {
        case 'help.page.filter':
            $output = '';
            $output .= '<h3>' . t('About') . '</h3>';
            $output .= '<p>' . t('The Filter module allows administrators to configure text formats. Text formats change how HTML tags and other text will be <em>processed and displayed</em> in the site. They are used to transform text, and also help to defend your website against potentially damaging input from malicious users. Visual text editors can be associated with text formats by using the <a href=":editor_help">Text Editor module</a>. For more information, see the <a href=":filter_do">online documentation for the Filter module</a>.', [
                ':filter_do' => 'https://www.drupal.org/documentation/modules/filter/',
                ':editor_help' => \Drupal::moduleHandler()->moduleExists('editor') ? Url::fromRoute('help.page', [
                    'name' => 'editor',
                ])->toString() : '#',
            ]) . '</p>';
            $output .= '<h3>' . t('Uses') . '</h3>';
            $output .= '<dl>';
            $output .= '<dt>' . t('Managing text formats') . '</dt>';
            $output .= '<dd>' . t('You can create and edit text formats on the <a href=":formats">Text formats page</a> (if the Text Editor module is enabled, this page is named Text formats and editors). One text format is included by default: Plain text (which removes all HTML tags). Additional text formats may be created during installation. You can create a text format by clicking "<a href=":add_format">Add text format</a>".', [
                ':formats' => Url::fromRoute('filter.admin_overview')->toString(),
                ':add_format' => Url::fromRoute('filter.format_add')->toString(),
            ]) . '</dd>';
            $output .= '<dt>' . t('Assigning roles to text formats') . '</dt>';
            $output .= '<dd>' . t('You can define which users will be able to use each text format by selecting roles. To ensure security, anonymous and untrusted users should only have access to text formats that restrict them to either plain text or a safe set of HTML tags. This is because HTML tags can allow embedding malicious links or scripts in text. More trusted registered users may be granted permission to use less restrictive text formats in order to create rich text. <strong>Improper text format configuration is a security risk.</strong>') . '</dd>';
            $output .= '<dt>' . t('Selecting filters') . '</dt>';
            $output .= '<dd>' . t('Each text format uses filters that add, remove, or transform elements within user-entered text. For example, one filter removes unapproved HTML tags, while another transforms URLs into clickable links. Filters are applied in a specific order. They do not change the <em>stored</em> content: they define how it is processed and displayed.') . '</dd>';
            $output .= '<dd>' . t('Each filter can have additional configuration options. For example, for the "Limit allowed HTML tags" filter you need to define the list of HTML tags that the filter leaves in the text.') . '</dd>';
            $output .= '<dt>' . t('Using text fields with text formats') . '</dt>';
            $output .= '<dd>' . t('Text fields that allow text formats are those with "formatted" in the description. These are <em>Text (formatted, long, with summary)</em>, <em>Text (formatted)</em>, and <em>Text (formatted, long)</em>. You cannot change the type of field once a field has been created.') . '</dd>';
            $output .= '<dt>' . t('Choosing a text format') . '</dt>';
            $output .= '<dd>' . t('When creating or editing data in a field that has text formats enabled, users can select the format under the field from the Text format select list.') . '</dd>';
            $output .= '</dl>';
            return $output;
        case 'filter.admin_overview':
            $output = '<p>' . t('Text formats define how text is filtered for output and how HTML tags and other text is displayed, replaced, or removed. <strong>Improper text format configuration is a security risk.</strong> Learn more on the <a href=":filter_help">Filter module help page</a>.', [
                ':filter_help' => Url::fromRoute('help.page', [
                    'name' => 'filter',
                ])->toString(),
            ]) . '</p>';
            $output .= '<p>' . t('Text formats are presented on content editing pages in the order defined on this page. The first format available to a user will be selected by default.') . '</p>';
            return $output;
        case 'entity.filter_format.edit_form':
            $output = '<p>' . t('A text format contains filters that change the display of user input; for example, stripping out malicious HTML or making URLs clickable. Filters are executed from top to bottom and the order is important, since one filter may prevent another filter from doing its job. For example, when URLs are converted into links before disallowed HTML tags are removed, all links may be removed. When this happens, the order of filters may need to be rearranged.') . '</p>';
            return $output;
    }
}

/**
 * Implements hook_theme().
 */
function filter_theme() {
    return [
        'filter_tips' => [
            'variables' => [
                'tips' => NULL,
                'long' => FALSE,
            ],
        ],
        'text_format_wrapper' => [
            'variables' => [
                'children' => NULL,
                'description' => NULL,
                'attributes' => [],
            ],
        ],
        'filter_guidelines' => [
            'variables' => [
                'format' => NULL,
            ],
        ],
        'filter_caption' => [
            'variables' => [
                'node' => NULL,
                'tag' => NULL,
                'caption' => NULL,
                'classes' => NULL,
            ],
        ],
    ];
}

/**
 * Retrieves a list of enabled text formats, ordered by weight.
 *
 * @param \Drupal\Core\Session\AccountInterface|null $account
 *   (optional) If provided, only those formats that are allowed for this user
 *   account will be returned. All enabled formats will be returned otherwise.
 *   Defaults to NULL.
 *
 * @return \Drupal\filter\FilterFormatInterface[]
 *   An array of text format objects, keyed by the format ID and ordered by
 *   weight.
 *
 * @see filter_formats_reset()
 */
function filter_formats(AccountInterface $account = NULL) {
    $formats =& drupal_static(__FUNCTION__, []);
    // All available formats are cached for performance.
    if (!isset($formats['all'])) {
        $language_interface = \Drupal::languageManager()->getCurrentLanguage();
        if ($cache = \Drupal::cache()->get("filter_formats:{$language_interface->getId()}")) {
            $formats['all'] = $cache->data;
        }
        else {
            $formats['all'] = \Drupal::entityTypeManager()->getStorage('filter_format')
                ->loadByProperties([
                'status' => TRUE,
            ]);
            uasort($formats['all'], 'Drupal\\Core\\Config\\Entity\\ConfigEntityBase::sort');
            \Drupal::cache()->set("filter_formats:{$language_interface->getId()}", $formats['all'], Cache::PERMANENT, \Drupal::entityTypeManager()->getDefinition('filter_format')
                ->getListCacheTags());
        }
    }
    // If no user was specified, return all formats.
    if (!isset($account)) {
        return $formats['all'];
    }
    // Build a list of user-specific formats.
    $account_id = $account->id();
    if (!isset($formats['user'][$account_id])) {
        $formats['user'][$account_id] = [];
        foreach ($formats['all'] as $format) {
            if ($format->access('use', $account)) {
                $formats['user'][$account_id][$format->id()] = $format;
            }
        }
    }
    return $formats['user'][$account_id];
}

/**
 * Resets the text format caches.
 *
 * @see filter_formats()
 */
function filter_formats_reset() {
    drupal_static_reset('filter_formats');
}

/**
 * Retrieves a list of roles that are allowed to use a given text format.
 *
 * @param \Drupal\filter\FilterFormatInterface $format
 *   An object representing the text format.
 *
 * @return array
 *   An array of role names, keyed by role ID.
 */
function filter_get_roles_by_format(FilterFormatInterface $format) {
    // Handle the fallback format upfront (all roles have access to this format).
    if ($format->isFallbackFormat()) {
        return user_role_names();
    }
    // Do not list any roles if the permission does not exist.
    $permission = $format->getPermissionName();
    return !empty($permission) ? user_role_names(FALSE, $permission) : [];
}

/**
 * Retrieves a list of text formats that are allowed for a given role.
 *
 * @param string $rid
 *   The user role ID to retrieve text formats for.
 *
 * @return \Drupal\filter\FilterFormatInterface[]
 *   An array of text format objects that are allowed for the role, keyed by
 *   the text format ID and ordered by weight.
 */
function filter_get_formats_by_role($rid) {
    $formats = [];
    foreach (filter_formats() as $format) {
        $roles = filter_get_roles_by_format($format);
        if (isset($roles[$rid])) {
            $formats[$format->id()] = $format;
        }
    }
    return $formats;
}

/**
 * Returns the ID of the default text format for a particular user.
 *
 * The default text format is the first available format that the user is
 * allowed to access, when the formats are ordered by weight. It should
 * generally be used as a default choice when presenting the user with a list
 * of possible text formats (for example, in a node creation form).
 *
 * Conversely, when existing content that does not have an assigned text format
 * needs to be filtered for display, the default text format is the wrong
 * choice, because it is not guaranteed to be consistent from user to user, and
 * some trusted users may have an unsafe text format set by default, which
 * should not be used on text of unknown origin. Instead, the fallback format
 * returned by filter_fallback_format() should be used, since that is intended
 * to be a safe, consistent format that is always available to all users.
 *
 * @param \Drupal\Core\Session\AccountInterface|null $account
 *   (optional) The user account to check. Defaults to the currently logged-in
 *   user. Defaults to NULL.
 *
 * @return string
 *   The ID of the user's default text format.
 *
 * @see filter_fallback_format()
 */
function filter_default_format(AccountInterface $account = NULL) {
    if (!isset($account)) {
        $account = \Drupal::currentUser();
    }
    // Get a list of formats for this user, ordered by weight. The first one
    // available is the user's default format.
    $formats = filter_formats($account);
    $format = reset($formats);
    return $format->id();
}

/**
 * Returns the ID of the fallback text format that all users have access to.
 *
 * The fallback text format is a regular text format in every respect, except
 * it does not participate in the filter permission system and cannot be
 * disabled. It needs to exist because any user who has permission to create
 * formatted content must always have at least one text format they can use.
 *
 * Because the fallback format is available to all users, it should always be
 * configured securely. For example, when the Filter module is installed, this
 * format is initialized to output plain text. Installation profiles and site
 * administrators have the freedom to configure it further.
 *
 * Note that the fallback format is completely distinct from the default format,
 * which differs per user and is simply the first format which that user has
 * access to. The default and fallback formats are only guaranteed to be the
 * same for users who do not have access to any other format; otherwise, the
 * fallback format's weight determines its placement with respect to the user's
 * other formats.
 *
 * Any modules implementing a format deletion functionality must not delete this
 * format.
 *
 * @return string|null
 *   The ID of the fallback text format.
 *
 * @see hook_filter_format_disable()
 * @see filter_default_format()
 */
function filter_fallback_format() {
    // This variable is automatically set in the database for all installations
    // of Drupal. In the event that it gets disabled or deleted somehow, there
    // is no safe default to return, since we do not want to risk making an
    // existing (and potentially unsafe) text format on the site automatically
    // available to all users. Returning NULL at least guarantees that this
    // cannot happen.
    return \Drupal::config('filter.settings')->get('fallback_format');
}

/**
 * Runs all the enabled filters on a piece of text.
 *
 * Note: Because filters can inject JavaScript or execute PHP code, security is
 * vital here. When a user supplies a text format, you should validate it using
 * $format->access() before accepting/using it. This is normally done in the
 * validation stage of the Form API. You should for example never make a
 * preview of content in a disallowed format.
 *
 * Note: this function should only be used when filtering text for use elsewhere
 * than on a rendered HTML page. If this is part of an HTML page, then a
 * renderable array with a #type 'processed_text' element should be used instead
 * of this, because that will allow cacheability metadata to be set and bubbled
 * up and attachments to be associated (assets, placeholders, etc.). In other
 * words: if you are presenting the filtered text in an HTML page, the only way
 * this will be presented correctly, is by using the 'processed_text' element.
 *
 * @param string $text
 *   The text to be filtered.
 * @param string|null $format_id
 *   (optional) The machine name of the filter format to be used to filter the
 *   text. Defaults to the fallback format. See filter_fallback_format().
 * @param string $langcode
 *   (optional) The language code of the text to be filtered, e.g. 'en' for
 *   English. This allows filters to be language-aware so language-specific
 *   text replacement can be implemented. Defaults to an empty string.
 * @param array $filter_types_to_skip
 *   (optional) An array of filter types to skip, or an empty array (default)
 *   to skip no filter types. All of the format's filters will be applied,
 *   except for filters of the types that are marked to be skipped.
 *   FilterInterface::TYPE_HTML_RESTRICTOR is the only type that cannot be
 *   skipped.
 *
 * @return \Drupal\Component\Render\MarkupInterface
 *   The filtered text.
 *
 * @see \Drupal\filter\Plugin\FilterInterface::process()
 *
 * @ingroup sanitization
 */
function check_markup($text, $format_id = NULL, $langcode = '', $filter_types_to_skip = []) {
    $build = [
        '#type' => 'processed_text',
        '#text' => $text,
        '#format' => $format_id,
        '#filter_types_to_skip' => $filter_types_to_skip,
        '#langcode' => $langcode,
    ];
    return \Drupal::service('renderer')->renderPlain($build);
}

/**
 * Retrieves the filter tips.
 *
 * @param string $format_id
 *   The ID of the text format for which to retrieve tips, or -1 to return tips
 *   for all formats accessible to the current user.
 * @param bool $long
 *   (optional) Boolean indicating whether the long form of tips should be
 *   returned. Defaults to FALSE.
 *
 * @return array
 *   An associative array of filtering tips, keyed by filter name. Each
 *   filtering tip is an associative array with elements:
 *   - tip: Tip text.
 *   - id: Filter ID.
 */
function _filter_tips($format_id, $long = FALSE) {
    $formats = filter_formats(\Drupal::currentUser());
    $tips = [];
    // If only listing one format, extract it from the $formats array.
    if ($format_id != -1) {
        $formats = [
            $formats[$format_id],
        ];
    }
    foreach ($formats as $format) {
        foreach ($format->filters() as $name => $filter) {
            if ($filter->status) {
                $tip = $filter->tips($long);
                if (isset($tip)) {
                    $tips[$format->label()][$name] = [
                        'tip' => [
                            '#markup' => $tip,
                        ],
                        'id' => $name,
                    ];
                }
            }
        }
    }
    return $tips;
}

/**
 * Prepares variables for text format guideline templates.
 *
 * Default template: filter-guidelines.html.twig.
 *
 * @param array $variables
 *   An associative array containing:
 *   - format: An object representing a text format.
 */
function template_preprocess_filter_guidelines(&$variables) {
    $format = $variables['format'];
    $variables['tips'] = [
        '#theme' => 'filter_tips',
        '#tips' => _filter_tips($format->id(), FALSE),
    ];
    // Add format id for filter.es6.js.
    $variables['attributes']['data-drupal-format-id'] = $format->id();
}

/**
 * Prepares variables for text format wrapper templates.
 *
 * Default template: text-format-wrapper.html.twig.
 *
 * @param array $variables
 *   An associative array containing:
 *   - attributes: An associative array containing properties of the element.
 */
function template_preprocess_text_format_wrapper(&$variables) {
    $variables['aria_description'] = FALSE;
    // Add element class and id for screen readers.
    if (isset($variables['attributes']['aria-describedby'])) {
        $variables['aria_description'] = TRUE;
        $variables['attributes']['id'] = $variables['attributes']['aria-describedby'];
        // Remove aria-describedby attribute as it shouldn't be visible here.
        unset($variables['attributes']['aria-describedby']);
    }
}

/**
 * Prepares variables for filter tips templates.
 *
 * Default template: filter-tips.html.twig.
 *
 * @param array $variables
 *   An associative array containing:
 *   - tips: An array containing descriptions and a CSS ID in the form of
 *     'module-name/filter-id' (only used when $long is TRUE) for each
 *     filter in one or more text formats. Example:
 *     @code
 *       array(
 *         'Full HTML' => array(
 *           0 => array(
 *             'tip' => 'Web page addresses and email addresses turn into links automatically.',
 *             'id' => 'filter/2',
 *           ),
 *         ),
 *       );
 *     @endcode
 *   - long: (optional) Whether the passed-in filter tips contain extended
 *     explanations, i.e. intended to be output on the path 'filter/tips'
 *     (TRUE), or are in a short format, i.e. suitable to be displayed below a
 *     form element. Defaults to FALSE.
 */
function template_preprocess_filter_tips(&$variables) {
    $tips = $variables['tips'];
    foreach ($variables['tips'] as $name => $tiplist) {
        foreach ($tiplist as $tip_key => $tip) {
            $tiplist[$tip_key]['attributes'] = new Attribute();
        }
        $variables['tips'][$name] = [
            'attributes' => new Attribute(),
            'name' => $name,
            'list' => $tiplist,
        ];
    }
    $variables['multiple'] = count($tips) > 1;
}

/**
 * @defgroup standard_filters Standard filters
 * @{
 * Filters implemented by the Filter module.
 */

/**
 * Converts text into hyperlinks automatically.
 *
 * This filter identifies and makes clickable three types of "links".
 * - URLs like http://example.com.
 * - Email addresses like name@example.com.
 * - Web addresses without the "http://" protocol defined, like
 *   www.example.com.
 * Each type must be processed separately, as there is no one regular
 * expression that could possibly match all of the cases in one pass.
 */
function _filter_url($text, $filter) {
    // Store the current text in case any of the preg_* functions fail.
    $saved_text = $text;
    // Tags to skip and not recurse into.
    $ignore_tags = 'a|script|style|code|pre';
    // Pass length to regexp callback.
    _filter_url_trim(NULL, $filter->settings['filter_url_length']);
    // Create an array which contains the regexps for each type of link.
    // The key to the regexp is the name of a function that is used as
    // callback function to process matches of the regexp. The callback function
    // is to return the replacement for the match. The array is used and
    // matching/replacement done below inside some loops.
    $tasks = [];
    // Prepare protocols pattern for absolute URLs.
    // \Drupal\Component\Utility\UrlHelper::stripDangerousProtocols() will replace
    // any bad protocols with HTTP, so we need to support the identical list.
    // While '//' is technically optional for MAILTO only, we cannot cleanly
    // differ between protocols here without hard-coding MAILTO, so '//' is
    // optional for all protocols.
    // @see \Drupal\Component\Utility\UrlHelper::stripDangerousProtocols()
    $protocols = \Drupal::getContainer()->getParameter('filter_protocols');
    $protocols = implode(':(?://)?|', $protocols) . ':(?://)?';
    $valid_url_path_characters = "[\\p{L}\\p{M}\\p{N}!\\*\\';:=\\+,\\.\$\\/%#\\[\\]\\-_~@&]";
    // Allow URL paths to contain balanced parens
    // 1. Used in Wikipedia URLs like /Primer_(film)
    // 2. Used in IIS sessions like /S(dfd346)/
    $valid_url_balanced_parens = '\\(' . $valid_url_path_characters . '+\\)';
    // Valid end-of-path characters (so /foo. does not gobble the period).
    // 1. Allow =&# for empty URL parameters and other URL-join artifacts
    $valid_url_ending_characters = '[\\p{L}\\p{M}\\p{N}:_+~#=/]|(?:' . $valid_url_balanced_parens . ')';
    $valid_url_query_chars = '[a-zA-Z0-9!?\\*\'@\\(\\);:&=\\+\\$\\/%#\\[\\]\\-_\\.,~|]';
    $valid_url_query_ending_chars = '[a-zA-Z0-9_&=#\\/]';
    // full path
    // and allow @ in a url, but only in the middle. Catch things like http://example.com/@user/
    $valid_url_path = '(?:(?:' . $valid_url_path_characters . '*(?:' . $valid_url_balanced_parens . $valid_url_path_characters . '*)*' . $valid_url_ending_characters . ')|(?:@' . $valid_url_path_characters . '+\\/))';
    // Prepare domain name pattern.
    // The ICANN seems to be on track towards accepting more diverse top level
    // domains (TLDs), so this pattern has been "future-proofed" to allow for
    // TLDs of length 2-64.
    $domain = '(?:[\\p{L}\\p{M}\\p{N}._+-]+\\.)?[\\p{L}\\p{M}]{2,64}\\b';
    // Mail domains differ from the generic domain pattern, specifically:
    // A . character must be present in the string that follows the @ character.
    $email_domain = '(?:[\\p{L}\\p{M}\\p{N}._+-]+\\.)+[\\p{L}\\p{M}]{2,64}\\b';
    $ip = '(?:[0-9]{1,3}\\.){3}[0-9]{1,3}';
    $auth = '[\\p{L}\\p{M}\\p{N}:%_+*~#?&=.,/;-]+@';
    $trail = '(' . $valid_url_path . '*)?(\\?' . $valid_url_query_chars . '*' . $valid_url_query_ending_chars . ')?';
    // Match absolute URLs.
    $url_pattern = "(?:{$auth})?(?:{$domain}|{$ip})/?(?:{$trail})?";
    $pattern = "`((?:{$protocols})(?:{$url_pattern}))`u";
    $tasks['_filter_url_parse_full_links'] = $pattern;
    // Match email addresses.
    $url_pattern = "[\\p{L}\\p{M}\\p{N}._+-]{1,254}@(?:{$email_domain})";
    $pattern = "`({$url_pattern})`u";
    $tasks['_filter_url_parse_email_links'] = $pattern;
    // Match www domains.
    $url_pattern = "www\\.(?:{$domain})/?(?:{$trail})?";
    $pattern = "`({$url_pattern})`u";
    $tasks['_filter_url_parse_partial_links'] = $pattern;
    // Each type of URL needs to be processed separately. The text is joined and
    // re-split after each task, since all injected HTML tags must be correctly
    // protected before the next task.
    foreach ($tasks as $task => $pattern) {
        // HTML comments need to be handled separately, as they may contain HTML
        // markup, especially a '>'. Therefore, remove all comment contents and add
        // them back later.
        _filter_url_escape_comments('', TRUE);
        $text = is_null($text) ? '' : preg_replace_callback('`<!--(.*?)-->`s', '_filter_url_escape_comments', $text);
        // Split at all tags; ensures that no tags or attributes are processed.
        $chunks = is_null($text) ? [
            '',
        ] : preg_split('/(<.+?>)/is', $text, -1, PREG_SPLIT_DELIM_CAPTURE);
        // Do not attempt to convert links into URLs if preg_split() fails.
        if ($chunks !== FALSE) {
            // PHP ensures that the array consists of alternating delimiters and
            // literals, and begins and ends with a literal (inserting NULL as
            // required). Therefore, the first chunk is always text:
            $chunk_type = 'text';
            // If a tag of $ignore_tags is found, it is stored in $open_tag and only
            // removed when the closing tag is found. Until the closing tag is found,
            // no replacements are made.
            $open_tag = '';
            for ($i = 0; $i < count($chunks); $i++) {
                if ($chunk_type == 'text') {
                    // Only process this text if there are no unclosed $ignore_tags.
                    if ($open_tag == '') {
                        // If there is a match, inject a link into this chunk via the callback
                        // function contained in $task.
                        $chunks[$i] = preg_replace_callback($pattern, $task, $chunks[$i]);
                    }
                    // Text chunk is done, so next chunk must be a tag.
                    $chunk_type = 'tag';
                }
                else {
                    // Only process this tag if there are no unclosed $ignore_tags.
                    if ($open_tag == '') {
                        // Check whether this tag is contained in $ignore_tags.
                        if (preg_match("`<({$ignore_tags})(?:\\s|>)`i", $chunks[$i], $matches)) {
                            $open_tag = $matches[1];
                        }
                    }
                    else {
                        if (preg_match("`<\\/{$open_tag}>`i", $chunks[$i], $matches)) {
                            $open_tag = '';
                        }
                    }
                    // Tag chunk is done, so next chunk must be text.
                    $chunk_type = 'text';
                }
            }
            $text = implode($chunks);
        }
        // Revert to the original comment contents
        _filter_url_escape_comments('', FALSE);
        $text = $text ? preg_replace_callback('`<!--(.*?)-->`', '_filter_url_escape_comments', $text) : $text;
    }
    // If there is no text at this point revert to the previous text.
    return strlen((string) $text) > 0 ? $text : $saved_text;
}

/**
 * Makes links out of absolute URLs.
 *
 * Callback for preg_replace_callback() within _filter_url().
 */
function _filter_url_parse_full_links($match) {
    // The $i:th parenthesis in the regexp contains the URL.
    $i = 1;
    $match[$i] = Html::decodeEntities($match[$i]);
    $caption = Html::escape(_filter_url_trim($match[$i]));
    $match[$i] = Html::escape($match[$i]);
    return '<a href="' . $match[$i] . '">' . $caption . '</a>';
}

/**
 * Makes links out of email addresses.
 *
 * Callback for preg_replace_callback() within _filter_url().
 */
function _filter_url_parse_email_links($match) {
    // The $i:th parenthesis in the regexp contains the URL.
    $i = 0;
    $match[$i] = Html::decodeEntities($match[$i]);
    $caption = Html::escape(_filter_url_trim($match[$i]));
    $match[$i] = Html::escape($match[$i]);
    return '<a href="mailto:' . $match[$i] . '">' . $caption . '</a>';
}

/**
 * Makes links out of domain names starting with "www.".
 *
 * Callback for preg_replace_callback() within _filter_url().
 */
function _filter_url_parse_partial_links($match) {
    // The $i:th parenthesis in the regexp contains the URL.
    $i = 1;
    $match[$i] = Html::decodeEntities($match[$i]);
    $caption = Html::escape(_filter_url_trim($match[$i]));
    $match[$i] = Html::escape($match[$i]);
    return '<a href="http://' . $match[$i] . '">' . $caption . '</a>';
}

/**
 * Escapes the contents of HTML comments.
 *
 * Callback for preg_replace_callback() within _filter_url().
 *
 * @param array $match
 *   An array containing matches to replace from preg_replace_callback(),
 *   whereas $match[1] is expected to contain the content to be filtered.
 * @param bool|null $escape
 *   (optional) A Boolean indicating whether to escape (TRUE) or unescape
 *   comments (FALSE). Defaults to NULL, indicating neither. If TRUE, statically
 *   cached $comments are reset.
 */
function _filter_url_escape_comments($match, $escape = NULL) {
    static $mode, $comments = [];
    if (isset($escape)) {
        $mode = $escape;
        if ($escape) {
            $comments = [];
        }
        return;
    }
    // Replace all HTML comments with a '<!-- [hash] -->' placeholder.
    if ($mode) {
        $content = $match[1];
        $hash = hash('sha256', $content);
        $comments[$hash] = $content;
        return "<!-- {$hash} -->";
    }
    else {
        $hash = $match[1];
        $hash = trim($hash);
        $content = $comments[$hash];
        return "<!--{$content}-->";
    }
}

/**
 * Shortens a long URL to a given length ending with an ellipsis.
 */
function _filter_url_trim($text, $length = NULL) {
    static $_length;
    if ($length !== NULL) {
        $_length = $length;
    }
    if (isset($text) && isset($_length)) {
        $text = Unicode::truncate($text, $_length, FALSE, TRUE);
    }
    return $text;
}

/**
 * Converts line breaks into <p> and <br> in an intelligent fashion.
 *
 * Based on: http://photomatt.net/scripts/autop
 */
function _filter_autop($text) {
    // All block level tags
    $block = '(?:table|thead|tfoot|caption|col|colgroup|tbody|tr|td|th|div|dl|dd|dt|ul|ol|li|pre|select|option|form|map|area|blockquote|address|math|input|p|h[1-6]|fieldset|legend|hr|article|aside|details|figcaption|figure|footer|header|hgroup|menu|nav|section|summary)';
    // Split at opening and closing PRE, SCRIPT, STYLE, OBJECT, IFRAME tags
    // and comments. We don't apply any processing to the contents of these tags
    // to avoid messing up code. We look for matched pairs and allow basic
    // nesting. For example:
    // "processed <pre> ignored <script> ignored </script> ignored </pre> processed"
    $chunks = preg_split('@(<!--.*?-->|</?(?:pre|script|style|object|iframe|drupal-media|svg|!--)[^>]*>)@i', $text, -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).
    $ignore = FALSE;
    $ignoretag = '';
    $output = '';
    foreach ($chunks as $i => $chunk) {
        if ($i % 2) {
            $comment = substr($chunk, 0, 4) == '<!--';
            if ($comment) {
                // Nothing to do, this is a comment.
                $output .= $chunk;
                continue;
            }
            // Opening or closing tag?
            $open = $chunk[1] != '/';
            [
                $tag,
            ] = preg_split('/[ >]/', substr($chunk, 2 - $open), 2);
            if (!$ignore) {
                if ($open) {
                    $ignore = TRUE;
                    $ignoretag = $tag;
                }
            }
            elseif (!$open && $ignoretag == $tag) {
                $ignore = FALSE;
                $ignoretag = '';
            }
        }
        elseif (!$ignore) {
            // just to make things a little easier, pad the end
            $chunk = preg_replace('|\\n*$|', '', $chunk) . "\n\n";
            $chunk = preg_replace('|<br />\\s*<br />|', "\n\n", $chunk);
            // Space things out a little
            $chunk = preg_replace('!(<' . $block . '[^>]*>)!', "\n\$1", $chunk);
            // Space things out a little
            $chunk = preg_replace('!(</' . $block . '>)!', "\$1\n\n", $chunk);
            // take care of duplicates
            $chunk = preg_replace("/\n\n+/", "\n\n", $chunk);
            $chunk = preg_replace('/^\\n|\\n\\s*\\n$/', '', $chunk);
            // make paragraphs, including one at the end
            $chunk = '<p>' . preg_replace('/\\n\\s*\\n\\n?(.)/', "</p>\n<p>\$1", $chunk) . "</p>\n";
            // problem with nested lists
            $chunk = preg_replace("|<p>(<li.+?)</p>|", "\$1", $chunk);
            $chunk = preg_replace('|<p><blockquote([^>]*)>|i', "<blockquote\$1><p>", $chunk);
            $chunk = str_replace('</blockquote></p>', '</p></blockquote>', $chunk);
            // under certain strange conditions it could create a P of entirely whitespace
            $chunk = preg_replace('|<p>\\s*</p>\\n?|', '', $chunk);
            $chunk = preg_replace('!<p>\\s*(</?' . $block . '[^>]*>)!', "\$1", $chunk);
            $chunk = preg_replace('!(</?' . $block . '[^>]*>)\\s*</p>!', "\$1", $chunk);
            // make line breaks
            $chunk = preg_replace('|(?<!<br />)\\s*\\n|', "<br />\n", $chunk);
            $chunk = preg_replace('!(</?' . $block . '[^>]*>)\\s*<br />!', "\$1", $chunk);
            $chunk = preg_replace('!<br />(\\s*</?(?:p|li|div|dl|dd|dt|th|pre|td|ul|ol)>)!', '$1', $chunk);
            $chunk = preg_replace('/&([^#])(?![A-Za-z0-9]{1,8};)/', '&amp;$1', $chunk);
        }
        $output .= $chunk;
    }
    return $output;
}

/**
 * Escapes all HTML tags, so they will be visible instead of being effective.
 */
function _filter_html_escape($text) {
    return trim(Html::escape($text));
}

/**
 * Process callback for local image filter.
 */
function _filter_html_image_secure_process($text) {
    // Find the path (e.g. '/') to Drupal root.
    $base_path = base_path();
    $base_path_length = mb_strlen($base_path);
    // Find the directory on the server where index.php resides.
    $local_dir = \Drupal::root() . '/';
    $html_dom = Html::load($text);
    $images = $html_dom->getElementsByTagName('img');
    
    /** @var \Drupal\Core\File\FileUrlGeneratorInterface $file_url_generator */
    $file_url_generator = \Drupal::service('file_url_generator');
    foreach ($images as $image) {
        $src = $image->getAttribute('src');
        // Transform absolute image URLs to relative image URLs: prevent problems on
        // multisite set-ups and prevent mixed content errors.
        $image->setAttribute('src', $file_url_generator->transformRelative($src));
        // Verify that $src starts with $base_path.
        // This also ensures that external images cannot be referenced.
        $src = $image->getAttribute('src');
        if (mb_substr($src, 0, $base_path_length) === $base_path) {
            // Remove the $base_path to get the path relative to the Drupal root.
            // Ensure the path refers to an actual image by prefixing the image source
            // with the Drupal root and running getimagesize() on it.
            $local_image_path = $local_dir . mb_substr($src, $base_path_length);
            $local_image_path = rawurldecode($local_image_path);
            if (@getimagesize($local_image_path)) {
                // The image has the right path. Erroneous images are dealt with below.
                continue;
            }
        }
        // Allow modules and themes to replace an invalid image with an error
        // indicator. See filter_filter_secure_image_alter().
        \Drupal::moduleHandler()->alter('filter_secure_image', $image);
    }
    $text = Html::serialize($html_dom);
    return $text;
}

/**
 * Implements hook_filter_secure_image_alter().
 *
 * Formats an image DOM element that has an invalid source.
 *
 * @see _filter_html_image_secure_process()
 */
function filter_filter_secure_image_alter(&$image) {
    // Turn an invalid image into an error indicator.
    $image->setAttribute('src', base_path() . 'core/misc/icons/e32700/error.svg');
    $image->setAttribute('alt', t('Image removed.'));
    $image->setAttribute('title', t('This image has been removed. For security reasons, only images from the local domain are allowed.'));
    $image->setAttribute('height', '16');
    $image->setAttribute('width', '16');
    // Add a CSS class to aid in styling.
    $class = $image->getAttribute('class') ? trim($image->getAttribute('class')) . ' ' : '';
    $class .= 'filter-image-invalid';
    $image->setAttribute('class', $class);
}

/**
 * @} End of "defgroup standard_filters".
 */