[moodle.git] / lib / weblib.php

<?php
// This file is part of Moodle - http://moodle.org/
//
// Moodle is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// Moodle is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with Moodle.  If not, see <http://www.gnu.org/licenses/>.

/**
 * Library of functions for web output
 *
 * Library of all general-purpose Moodle PHP functions and constants
 * that produce HTML output
 *
 * Other main libraries:
 * - datalib.php - functions that access the database.
 * - moodlelib.php - general-purpose Moodle functions.
 *
 * @package    core
 * @subpackage lib
 * @copyright  1999 onwards Martin Dougiamas {@link http://moodle.com}
 * @license    http://www.gnu.org/copyleft/gpl.html GNU GPL v3 or later
 */

defined('MOODLE_INTERNAL') || die();

// Constants.

// Define text formatting types ... eventually we can add Wiki, BBcode etc.

/**
 * Does all sorts of transformations and filtering.
 */
define('FORMAT_MOODLE',   '0');

/**
 * Plain HTML (with some tags stripped).
 */
define('FORMAT_HTML',     '1');

/**
 * Plain text (even tags are printed in full).
 */
define('FORMAT_PLAIN',    '2');

/**
 * Wiki-formatted text.
 * Deprecated: left here just to note that '3' is not used (at the moment)
 * and to catch any latent wiki-like text (which generates an error)
 * @deprecated since 2005!
 */
define('FORMAT_WIKI',     '3');

/**
 * Markdown-formatted text http://daringfireball.net/projects/markdown/
 */
define('FORMAT_MARKDOWN', '4');

/**
 * A moodle_url comparison using this flag will return true if the base URLs match, params are ignored.
 */
define('URL_MATCH_BASE', 0);

/**
 * A moodle_url comparison using this flag will return true if the base URLs match and the params of url1 are part of url2.
 */
define('URL_MATCH_PARAMS', 1);

/**
 * A moodle_url comparison using this flag will return true if the two URLs are identical, except for the order of the params.
 */
define('URL_MATCH_EXACT', 2);

// Functions.

/**
 * Add quotes to HTML characters.
 *
 * Returns $var with HTML characters (like "<", ">", etc.) properly quoted.
 * Related function {@link p()} simply prints the output of this function.
 *
 * @param string $var the string potentially containing HTML characters
 * @return string
 */
function s($var) {

    if ($var === false) {
        return '0';
    }

    return preg_replace('/&amp;#(\d+|x[0-9a-f]+);/i', '&#$1;',
            htmlspecialchars($var, ENT_QUOTES | ENT_HTML401 | ENT_SUBSTITUTE));
}

/**
 * Add quotes to HTML characters.
 *
 * Prints $var with HTML characters (like "<", ">", etc.) properly quoted.
 * This function simply calls & displays {@link s()}.
 * @see s()
 *
 * @param string $var the string potentially containing HTML characters
 * @return string
 */
function p($var) {
    echo s($var);
}

/**
 * Does proper javascript quoting.
 *
 * Do not use addslashes anymore, because it does not work when magic_quotes_sybase is enabled.
 *
 * @param mixed $var String, Array, or Object to add slashes to
 * @return mixed quoted result
 */
function addslashes_js($var) {
    if (is_string($var)) {
        $var = str_replace('\\', '\\\\', $var);
        $var = str_replace(array('\'', '"', "\n", "\r", "\0"), array('\\\'', '\\"', '\\n', '\\r', '\\0'), $var);
        $var = str_replace('</', '<\/', $var);   // XHTML compliance.
    } else if (is_array($var)) {
        $var = array_map('addslashes_js', $var);
    } else if (is_object($var)) {
        $a = get_object_vars($var);
        foreach ($a as $key => $value) {
            $a[$key] = addslashes_js($value);
        }
        $var = (object)$a;
    }
    return $var;
}

/**
 * Remove query string from url.
 *
 * Takes in a URL and returns it without the querystring portion.
 *
 * @param string $url the url which may have a query string attached.
 * @return string The remaining URL.
 */
function strip_querystring($url) {

    if ($commapos = strpos($url, '?')) {
        return substr($url, 0, $commapos);
    } else {
        return $url;
    }
}

/**
 * Returns the name of the current script, WITH the querystring portion.
 *
 * This function is necessary because PHP_SELF and REQUEST_URI and SCRIPT_NAME
 * return different things depending on a lot of things like your OS, Web
 * server, and the way PHP is compiled (ie. as a CGI, module, ISAPI, etc.)
 * <b>NOTE:</b> This function returns false if the global variables needed are not set.
 *
 * @return mixed String or false if the global variables needed are not set.
 */
function me() {
    global $ME;
    return $ME;
}

/**
 * Guesses the full URL of the current script.
 *
 * This function is using $PAGE->url, but may fall back to $FULLME which
 * is constructed from  PHP_SELF and REQUEST_URI or SCRIPT_NAME
 *
 * @return mixed full page URL string or false if unknown
 */
function qualified_me() {
    global $FULLME, $PAGE, $CFG;

    if (isset($PAGE) and $PAGE->has_set_url()) {
        // This is the only recommended way to find out current page.
        return $PAGE->url->out(false);

    } else {
        if ($FULLME === null) {
            // CLI script most probably.
            return false;
        }
        if (!empty($CFG->sslproxy)) {
            // Return only https links when using SSL proxy.
            return preg_replace('/^http:/', 'https:', $FULLME, 1);
        } else {
            return $FULLME;
        }
    }
}

/**
 * Determines whether or not the Moodle site is being served over HTTPS.
 *
 * This is done simply by checking the value of $CFG->wwwroot, which seems
 * to be the only reliable method.
 *
 * @return boolean True if site is served over HTTPS, false otherwise.
 */
function is_https() {
    global $CFG;

    return (strpos($CFG->wwwroot, 'https://') === 0);
}

/**
 * Returns the cleaned local URL of the HTTP_REFERER less the URL query string parameters if required.
 *
 * @param bool $stripquery if true, also removes the query part of the url.
 * @return string The resulting referer or empty string.
 */
function get_local_referer($stripquery = true) {
    if (isset($_SERVER['HTTP_REFERER'])) {
        $referer = clean_param($_SERVER['HTTP_REFERER'], PARAM_LOCALURL);
        if ($stripquery) {
            return strip_querystring($referer);
        } else {
            return $referer;
        }
    } else {
        return '';
    }
}

/**
 * Class for creating and manipulating urls.
 *
 * It can be used in moodle pages where config.php has been included without any further includes.
 *
 * It is useful for manipulating urls with long lists of params.
 * One situation where it will be useful is a page which links to itself to perform various actions
 * and / or to process form data. A moodle_url object :
 * can be created for a page to refer to itself with all the proper get params being passed from page call to
 * page call and methods can be used to output a url including all the params, optionally adding and overriding
 * params and can also be used to
 *     - output the url without any get params
 *     - and output the params as hidden fields to be output within a form
 *
 * @copyright 2007 jamiesensei
 * @link http://docs.moodle.org/dev/lib/weblib.php_moodle_url See short write up here
 * @license http://www.gnu.org/copyleft/gpl.html GNU GPL v3 or later
 * @package core
 */
class moodle_url {

    /**
     * Scheme, ex.: http, https
     * @var string
     */
    protected $scheme = '';

    /**
     * Hostname.
     * @var string
     */
    protected $host = '';

    /**
     * Port number, empty means default 80 or 443 in case of http.
     * @var int
     */
    protected $port = '';

    /**
     * Username for http auth.
     * @var string
     */
    protected $user = '';

    /**
     * Password for http auth.
     * @var string
     */
    protected $pass = '';

    /**
     * Script path.
     * @var string
     */
    protected $path = '';

    /**
     * Optional slash argument value.
     * @var string
     */
    protected $slashargument = '';

    /**
     * Anchor, may be also empty, null means none.
     * @var string
     */
    protected $anchor = null;

    /**
     * Url parameters as associative array.
     * @var array
     */
    protected $params = array();

    /**
     * Create new instance of moodle_url.
     *
     * @param moodle_url|string $url - moodle_url means make a copy of another
     *      moodle_url and change parameters, string means full url or shortened
     *      form (ex.: '/course/view.php'). It is strongly encouraged to not include
     *      query string because it may result in double encoded values. Use the
     *      $params instead. For admin URLs, just use /admin/script.php, this
     *      class takes care of the $CFG->admin issue.
     * @param array $params these params override current params or add new
     * @param string $anchor The anchor to use as part of the URL if there is one.
     * @throws moodle_exception
     */
    public function __construct($url, array $params = null, $anchor = null) {
        global $CFG;

        if ($url instanceof moodle_url) {
            $this->scheme = $url->scheme;
            $this->host = $url->host;
            $this->port = $url->port;
            $this->user = $url->user;
            $this->pass = $url->pass;
            $this->path = $url->path;
            $this->slashargument = $url->slashargument;
            $this->params = $url->params;
            $this->anchor = $url->anchor;

        } else {
            // Detect if anchor used.
            $apos = strpos($url, '#');
            if ($apos !== false) {
                $anchor = substr($url, $apos);
                $anchor = ltrim($anchor, '#');
                $this->set_anchor($anchor);
                $url = substr($url, 0, $apos);
            }

            // Normalise shortened form of our url ex.: '/course/view.php'.
            if (strpos($url, '/') === 0) {
                $url = $CFG->wwwroot.$url;
            }

            if ($CFG->admin !== 'admin') {
                if (strpos($url, "$CFG->wwwroot/admin/") === 0) {
                    $url = str_replace("$CFG->wwwroot/admin/", "$CFG->wwwroot/$CFG->admin/", $url);
                }
            }

            // Parse the $url.
            $parts = parse_url($url);
            if ($parts === false) {
                throw new moodle_exception('invalidurl');
            }
            if (isset($parts['query'])) {
                // Note: the values may not be correctly decoded, url parameters should be always passed as array.
                parse_str(str_replace('&amp;', '&', $parts['query']), $this->params);
            }
            unset($parts['query']);
            foreach ($parts as $key => $value) {
                $this->$key = $value;
            }

            // Detect slashargument value from path - we do not support directory names ending with .php.
            $pos = strpos($this->path, '.php/');
            if ($pos !== false) {
                $this->slashargument = substr($this->path, $pos + 4);
                $this->path = substr($this->path, 0, $pos + 4);
            }
        }

        $this->params($params);
        if ($anchor !== null) {
            $this->anchor = (string)$anchor;
        }
    }

    /**
     * Add an array of params to the params for this url.
     *
     * The added params override existing ones if they have the same name.
     *
     * @param array $params Defaults to null. If null then returns all params.
     * @return array Array of Params for url.
     * @throws coding_exception
     */
    public function params(array $params = null) {
        $params = (array)$params;

        foreach ($params as $key => $value) {
            if (is_int($key)) {
                throw new coding_exception('Url parameters can not have numeric keys!');
            }
            if (!is_string($value)) {
                if (is_array($value)) {
                    throw new coding_exception('Url parameters values can not be arrays!');
                }
                if (is_object($value) and !method_exists($value, '__toString')) {
                    throw new coding_exception('Url parameters values can not be objects, unless __toString() is defined!');
                }
            }
            $this->params[$key] = (string)$value;
        }
        return $this->params;
    }

    /**
     * Remove all params if no arguments passed.
     * Remove selected params if arguments are passed.
     *
     * Can be called as either remove_params('param1', 'param2')
     * or remove_params(array('param1', 'param2')).
     *
     * @param string[]|string $params,... either an array of param names, or 1..n string params to remove as args.
     * @return array url parameters
     */
    public function remove_params($params = null) {
        if (!is_array($params)) {
            $params = func_get_args();
        }
        foreach ($params as $param) {
            unset($this->params[$param]);
        }
        return $this->params;
    }

    /**
     * Remove all url parameters.
     *
     * @todo remove the unused param.
     * @param array $params Unused param
     * @return void
     */
    public function remove_all_params($params = null) {
        $this->params = array();
        $this->slashargument = '';
    }

    /**
     * Add a param to the params for this url.
     *
     * The added param overrides existing one if they have the same name.
     *
     * @param string $paramname name
     * @param string $newvalue Param value. If new value specified current value is overriden or parameter is added
     * @return mixed string parameter value, null if parameter does not exist
     */
    public function param($paramname, $newvalue = '') {
        if (func_num_args() > 1) {
            // Set new value.
            $this->params(array($paramname => $newvalue));
        }
        if (isset($this->params[$paramname])) {
            return $this->params[$paramname];
        } else {
            return null;
        }
    }

    /**
     * Merges parameters and validates them
     *
     * @param array $overrideparams
     * @return array merged parameters
     * @throws coding_exception
     */
    protected function merge_overrideparams(array $overrideparams = null) {
        $overrideparams = (array)$overrideparams;
        $params = $this->params;
        foreach ($overrideparams as $key => $value) {
            if (is_int($key)) {
                throw new coding_exception('Overridden parameters can not have numeric keys!');
            }
            if (is_array($value)) {
                throw new coding_exception('Overridden parameters values can not be arrays!');
            }
            if (is_object($value) and !method_exists($value, '__toString')) {
                throw new coding_exception('Overridden parameters values can not be objects, unless __toString() is defined!');
            }
            $params[$key] = (string)$value;
        }
        return $params;
    }

    /**
     * Get the params as as a query string.
     *
     * This method should not be used outside of this method.
     *
     * @param bool $escaped Use &amp; as params separator instead of plain &
     * @param array $overrideparams params to add to the output params, these
     *      override existing ones with the same name.
     * @return string query string that can be added to a url.
     */
    public function get_query_string($escaped = true, array $overrideparams = null) {
        $arr = array();
        if ($overrideparams !== null) {
            $params = $this->merge_overrideparams($overrideparams);
        } else {
            $params = $this->params;
        }
        foreach ($params as $key => $val) {
            if (is_array($val)) {
                foreach ($val as $index => $value) {
                    $arr[] = rawurlencode($key.'['.$index.']')."=".rawurlencode($value);
                }
            } else {
                if (isset($val) && $val !== '') {
                    $arr[] = rawurlencode($key)."=".rawurlencode($val);
                } else {
                    $arr[] = rawurlencode($key);
                }
            }
        }
        if ($escaped) {
            return implode('&amp;', $arr);
        } else {
            return implode('&', $arr);
        }
    }

    /**
     * Shortcut for printing of encoded URL.
     *
     * @return string
     */
    public function __toString() {
        return $this->out(true);
    }

    /**
     * Output url.
     *
     * If you use the returned URL in HTML code, you want the escaped ampersands. If you use
     * the returned URL in HTTP headers, you want $escaped=false.
     *
     * @param bool $escaped Use &amp; as params separator instead of plain &
     * @param array $overrideparams params to add to the output url, these override existing ones with the same name.
     * @return string Resulting URL
     */
    public function out($escaped = true, array $overrideparams = null) {

        global $CFG;

        if (!is_bool($escaped)) {
            debugging('Escape parameter must be of type boolean, '.gettype($escaped).' given instead.');
        }

        $url = $this;

        // Allow url's to be rewritten by a plugin.
        if (isset($CFG->urlrewriteclass) && !isset($CFG->upgraderunning)) {
            $class = $CFG->urlrewriteclass;
            $pluginurl = $class::url_rewrite($url);
            if ($pluginurl instanceof moodle_url) {
                $url = $pluginurl;
            }
        }

        return $url->raw_out($escaped, $overrideparams);

    }

    /**
     * Output url without any rewrites
     *
     * This is identical in signature and use to out() but doesn't call the rewrite handler.
     *
     * @param bool $escaped Use &amp; as params separator instead of plain &
     * @param array $overrideparams params to add to the output url, these override existing ones with the same name.
     * @return string Resulting URL
     */
    public function raw_out($escaped = true, array $overrideparams = null) {
        if (!is_bool($escaped)) {
            debugging('Escape parameter must be of type boolean, '.gettype($escaped).' given instead.');
        }

        $uri = $this->out_omit_querystring().$this->slashargument;

        $querystring = $this->get_query_string($escaped, $overrideparams);
        if ($querystring !== '') {
            $uri .= '?' . $querystring;
        }
        if (!is_null($this->anchor)) {
            $uri .= '#'.$this->anchor;
        }

        return $uri;
    }

    /**
     * Returns url without parameters, everything before '?'.
     *
     * @param bool $includeanchor if {@link self::anchor} is defined, should it be returned?
     * @return string
     */
    public function out_omit_querystring($includeanchor = false) {

        $uri = $this->scheme ? $this->scheme.':'.((strtolower($this->scheme) == 'mailto') ? '':'//'): '';
        $uri .= $this->user ? $this->user.($this->pass? ':'.$this->pass:'').'@':'';
        $uri .= $this->host ? $this->host : '';
        $uri .= $this->port ? ':'.$this->port : '';
        $uri .= $this->path ? $this->path : '';
        if ($includeanchor and !is_null($this->anchor)) {
            $uri .= '#' . $this->anchor;
        }

        return $uri;
    }

    /**
     * Compares this moodle_url with another.
     *
     * See documentation of constants for an explanation of the comparison flags.
     *
     * @param moodle_url $url The moodle_url object to compare
     * @param int $matchtype The type of comparison (URL_MATCH_BASE, URL_MATCH_PARAMS, URL_MATCH_EXACT)
     * @return bool
     */
    public function compare(moodle_url $url, $matchtype = URL_MATCH_EXACT) {

        $baseself = $this->out_omit_querystring();
        $baseother = $url->out_omit_querystring();

        // Append index.php if there is no specific file.
        if (substr($baseself, -1) == '/') {
            $baseself .= 'index.php';
        }
        if (substr($baseother, -1) == '/') {
            $baseother .= 'index.php';
        }

        // Compare the two base URLs.
        if ($baseself != $baseother) {
            return false;
        }

        if ($matchtype == URL_MATCH_BASE) {
            return true;
        }

        $urlparams = $url->params();
        foreach ($this->params() as $param => $value) {
            if ($param == 'sesskey') {
                continue;
            }
            if (!array_key_exists($param, $urlparams) || $urlparams[$param] != $value) {
                return false;
            }
        }

        if ($matchtype == URL_MATCH_PARAMS) {
            return true;
        }

        foreach ($urlparams as $param => $value) {
            if ($param == 'sesskey') {
                continue;
            }
            if (!array_key_exists($param, $this->params()) || $this->param($param) != $value) {
                return false;
            }
        }

        if ($url->anchor !== $this->anchor) {
            return false;
        }

        return true;
    }

    /**
     * Sets the anchor for the URI (the bit after the hash)
     *
     * @param string $anchor null means remove previous
     */
    public function set_anchor($anchor) {
        if (is_null($anchor)) {
            // Remove.
            $this->anchor = null;
        } else if ($anchor === '') {
            // Special case, used as empty link.
            $this->anchor = '';
        } else if (preg_match('|[a-zA-Z\_\:][a-zA-Z0-9\_\-\.\:]*|', $anchor)) {
            // Match the anchor against the NMTOKEN spec.
            $this->anchor = $anchor;
        } else {
            // Bad luck, no valid anchor found.
            $this->anchor = null;
        }
    }

    /**
     * Sets the scheme for the URI (the bit before ://)
     *
     * @param string $scheme
     */
    public function set_scheme($scheme) {
        // See http://www.ietf.org/rfc/rfc3986.txt part 3.1.
        if (preg_match('/^[a-zA-Z][a-zA-Z0-9+.-]*$/', $scheme)) {
            $this->scheme = $scheme;
        } else {
            throw new coding_exception('Bad URL scheme.');
        }
    }

    /**
     * Sets the url slashargument value.
     *
     * @param string $path usually file path
     * @param string $parameter name of page parameter if slasharguments not supported
     * @param bool $supported usually null, then it depends on $CFG->slasharguments, use true or false for other servers
     * @return void
     */
    public function set_slashargument($path, $parameter = 'file', $supported = null) {
        global $CFG;
        if (is_null($supported)) {
            $supported = !empty($CFG->slasharguments);
        }

        if ($supported) {
            $parts = explode('/', $path);
            $parts = array_map('rawurlencode', $parts);
            $path  = implode('/', $parts);
            $this->slashargument = $path;
            unset($this->params[$parameter]);

        } else {
            $this->slashargument = '';
            $this->params[$parameter] = $path;
        }
    }

    // Static factory methods.

    /**
     * General moodle file url.
     *
     * @param string $urlbase the script serving the file
     * @param string $path
     * @param bool $forcedownload
     * @return moodle_url
     */
    public static function make_file_url($urlbase, $path, $forcedownload = false) {
        $params = array();
        if ($forcedownload) {
            $params['forcedownload'] = 1;
        }
        $url = new moodle_url($urlbase, $params);
        $url->set_slashargument($path);
        return $url;
    }

    /**
     * Factory method for creation of url pointing to plugin file.
     *
     * Please note this method can be used only from the plugins to
     * create urls of own files, it must not be used outside of plugins!
     *
     * @param int $contextid
     * @param string $component
     * @param string $area
     * @param int $itemid
     * @param string $pathname
     * @param string $filename
     * @param bool $forcedownload
     * @param mixed $includetoken Whether to use a user token when displaying this group image.
     *                True indicates to generate a token for current user, and integer value indicates to generate a token for the
     *                user whose id is the value indicated.
     *                If the group picture is included in an e-mail or some other location where the audience is a specific
     *                user who will not be logged in when viewing, then we use a token to authenticate the user.
     * @return moodle_url
     */
    public static function make_pluginfile_url($contextid, $component, $area, $itemid, $pathname, $filename,
                                               $forcedownload = false, $includetoken = false) {
        global $CFG, $USER;

        $path = [];

        if ($includetoken) {
            $urlbase = "$CFG->wwwroot/tokenpluginfile.php";
            $userid = $includetoken === true ? $USER->id : $includetoken;
            $token = get_user_key('core_files', $userid);
            if ($CFG->slasharguments) {
                $path[] = $token;
            }
        } else {
            $urlbase = "$CFG->wwwroot/pluginfile.php";
        }
        $path[] = $contextid;
        $path[] = $component;
        $path[] = $area;

        if ($itemid !== null) {
            $path[] = $itemid;
        }

        $path = "/" . implode('/', $path) . "{$pathname}{$filename}";

        $url = self::make_file_url($urlbase, $path, $forcedownload, $includetoken);
        if ($includetoken && empty($CFG->slasharguments)) {
            $url->param('token', $token);
        }
        return $url;
    }

    /**
     * Factory method for creation of url pointing to plugin file.
     * This method is the same that make_pluginfile_url but pointing to the webservice pluginfile.php script.
     * It should be used only in external functions.
     *
     * @since  2.8
     * @param int $contextid
     * @param string $component
     * @param string $area
     * @param int $itemid
     * @param string $pathname
     * @param string $filename
     * @param bool $forcedownload
     * @return moodle_url
     */
    public static function make_webservice_pluginfile_url($contextid, $component, $area, $itemid, $pathname, $filename,
                                               $forcedownload = false) {
        global $CFG;
        $urlbase = "$CFG->wwwroot/webservice/pluginfile.php";
        if ($itemid === null) {
            return self::make_file_url($urlbase, "/$contextid/$component/$area".$pathname.$filename, $forcedownload);
        } else {
            return self::make_file_url($urlbase, "/$contextid/$component/$area/$itemid".$pathname.$filename, $forcedownload);
        }
    }

    /**
     * Factory method for creation of url pointing to draft file of current user.
     *
     * @param int $draftid draft item id
     * @param string $pathname
     * @param string $filename
     * @param bool $forcedownload
     * @return moodle_url
     */
    public static function make_draftfile_url($draftid, $pathname, $filename, $forcedownload = false) {
        global $CFG, $USER;
        $urlbase = "$CFG->wwwroot/draftfile.php";
        $context = context_user::instance($USER->id);

        return self::make_file_url($urlbase, "/$context->id/user/draft/$draftid".$pathname.$filename, $forcedownload);
    }

    /**
     * Factory method for creating of links to legacy course files.
     *
     * @param int $courseid
     * @param string $filepath
     * @param bool $forcedownload
     * @return moodle_url
     */
    public static function make_legacyfile_url($courseid, $filepath, $forcedownload = false) {
        global $CFG;

        $urlbase = "$CFG->wwwroot/file.php";
        return self::make_file_url($urlbase, '/'.$courseid.'/'.$filepath, $forcedownload);
    }

    /**
     * Returns URL a relative path from $CFG->wwwroot
     *
     * Can be used for passing around urls with the wwwroot stripped
     *
     * @param boolean $escaped Use &amp; as params separator instead of plain &
     * @param array $overrideparams params to add to the output url, these override existing ones with the same name.
     * @return string Resulting URL
     * @throws coding_exception if called on a non-local url
     */
    public function out_as_local_url($escaped = true, array $overrideparams = null) {
        global $CFG;

        $url = $this->out($escaped, $overrideparams);

        // Url should be equal to wwwroot. If not then throw exception.
        if (($url === $CFG->wwwroot) || (strpos($url, $CFG->wwwroot.'/') === 0)) {
            $localurl = substr($url, strlen($CFG->wwwroot));
            return !empty($localurl) ? $localurl : '';
        } else {
            throw new coding_exception('out_as_local_url called on a non-local URL');
        }
    }

    /**
     * Returns the 'path' portion of a URL. For example, if the URL is
     * http://www.example.org:447/my/file/is/here.txt?really=1 then this will
     * return '/my/file/is/here.txt'.
     *
     * By default the path includes slash-arguments (for example,
     * '/myfile.php/extra/arguments') so it is what you would expect from a
     * URL path. If you don't want this behaviour, you can opt to exclude the
     * slash arguments. (Be careful: if the $CFG variable slasharguments is
     * disabled, these URLs will have a different format and you may need to
     * look at the 'file' parameter too.)
     *
     * @param bool $includeslashargument If true, includes slash arguments
     * @return string Path of URL
     */
    public function get_path($includeslashargument = true) {
        return $this->path . ($includeslashargument ? $this->slashargument : '');
    }

    /**
     * Returns a given parameter value from the URL.
     *
     * @param string $name Name of parameter
     * @return string Value of parameter or null if not set
     */
    public function get_param($name) {
        if (array_key_exists($name, $this->params)) {
            return $this->params[$name];
        } else {
            return null;
        }
    }

    /**
     * Returns the 'scheme' portion of a URL. For example, if the URL is
     * http://www.example.org:447/my/file/is/here.txt?really=1 then this will
     * return 'http' (without the colon).
     *
     * @return string Scheme of the URL.
     */
    public function get_scheme() {
        return $this->scheme;
    }

    /**
     * Returns the 'host' portion of a URL. For example, if the URL is
     * http://www.example.org:447/my/file/is/here.txt?really=1 then this will
     * return 'www.example.org'.
     *
     * @return string Host of the URL.
     */
    public function get_host() {
        return $this->host;
    }

    /**
     * Returns the 'port' portion of a URL. For example, if the URL is
     * http://www.example.org:447/my/file/is/here.txt?really=1 then this will
     * return '447'.
     *
     * @return string Port of the URL.
     */
    public function get_port() {
        return $this->port;
    }
}

/**
 * Determine if there is data waiting to be processed from a form
 *
 * Used on most forms in Moodle to check for data
 * Returns the data as an object, if it's found.
 * This object can be used in foreach loops without
 * casting because it's cast to (array) automatically
 *
 * Checks that submitted POST data exists and returns it as object.
 *
 * @return mixed false or object
 */
function data_submitted() {

    if (empty($_POST)) {
        return false;
    } else {
        return (object)fix_utf8($_POST);
    }
}

/**
 * Given some normal text this function will break up any
 * long words to a given size by inserting the given character
 *
 * It's multibyte savvy and doesn't change anything inside html tags.
 *
 * @param string $string the string to be modified
 * @param int $maxsize maximum length of the string to be returned
 * @param string $cutchar the string used to represent word breaks
 * @return string
 */
function break_up_long_words($string, $maxsize=20, $cutchar=' ') {

    // First of all, save all the tags inside the text to skip them.
    $tags = array();
    filter_save_tags($string, $tags);

    // Process the string adding the cut when necessary.
    $output = '';
    $length = core_text::strlen($string);
    $wordlength = 0;

    for ($i=0; $i<$length; $i++) {
        $char = core_text::substr($string, $i, 1);
        if ($char == ' ' or $char == "\t" or $char == "\n" or $char == "\r" or $char == "<" or $char == ">") {
            $wordlength = 0;
        } else {
            $wordlength++;
            if ($wordlength > $maxsize) {
                $output .= $cutchar;
                $wordlength = 0;
            }
        }
        $output .= $char;
    }

    // Finally load the tags back again.
    if (!empty($tags)) {
        $output = str_replace(array_keys($tags), $tags, $output);
    }

    return $output;
}

/**
 * Try and close the current window using JavaScript, either immediately, or after a delay.
 *
 * Echo's out the resulting XHTML & javascript
 *
 * @param integer $delay a delay in seconds before closing the window. Default 0.
 * @param boolean $reloadopener if true, we will see if this window was a pop-up, and try
 *      to reload the parent window before this one closes.
 */
function close_window($delay = 0, $reloadopener = false) {
    global $PAGE, $OUTPUT;

    if (!$PAGE->headerprinted) {
        $PAGE->set_title(get_string('closewindow'));
        echo $OUTPUT->header();
    } else {
        $OUTPUT->container_end_all(false);
    }

    if ($reloadopener) {
        // Trigger the reload immediately, even if the reload is after a delay.
        $PAGE->requires->js_function_call('window.opener.location.reload', array(true));
    }
    $OUTPUT->notification(get_string('windowclosing'), 'notifysuccess');

    $PAGE->requires->js_function_call('close_window', array(new stdClass()), false, $delay);

    echo $OUTPUT->footer();
    exit;
}

/**
 * Returns a string containing a link to the user documentation for the current page.
 *
 * Also contains an icon by default. Shown to teachers and admin only.
 *
 * @param string $text The text to be displayed for the link
 * @return string The link to user documentation for this current page
 */
function page_doc_link($text='') {
    global $OUTPUT, $PAGE;
    $path = page_get_doc_link_path($PAGE);
    if (!$path) {
        return '';
    }
    return $OUTPUT->doc_link($path, $text);
}

/**
 * Returns the path to use when constructing a link to the docs.
 *
 * @since Moodle 2.5.1 2.6
 * @param moodle_page $page
 * @return string
 */
function page_get_doc_link_path(moodle_page $page) {
    global $CFG;

    if (empty($CFG->docroot) || during_initial_install()) {
        return '';
    }
    if (!has_capability('moodle/site:doclinks', $page->context)) {
        return '';
    }

    $path = $page->docspath;
    if (!$path) {
        return '';
    }
    return $path;
}


/**
 * Validates an email to make sure it makes sense.
 *
 * @param string $address The email address to validate.
 * @return boolean
 */
function validate_email($address) {
    global $CFG;
    require_once($CFG->libdir.'/phpmailer/moodle_phpmailer.php');

    return moodle_phpmailer::validateAddress($address);
}

/**
 * Extracts file argument either from file parameter or PATH_INFO
 *
 * Note: $scriptname parameter is not needed anymore
 *
 * @return string file path (only safe characters)
 */
function get_file_argument() {
    global $SCRIPT;

    $relativepath = false;
    $hasforcedslashargs = false;

    if (isset($_SERVER['REQUEST_URI']) && !empty($_SERVER['REQUEST_URI'])) {
        // Checks whether $_SERVER['REQUEST_URI'] contains '/pluginfile.php/'
        // instead of '/pluginfile.php?', when serving a file from e.g. mod_imscp or mod_scorm.
        if ((strpos($_SERVER['REQUEST_URI'], '/pluginfile.php/') !== false)
                && isset($_SERVER['PATH_INFO']) && !empty($_SERVER['PATH_INFO'])) {
            // Exclude edge cases like '/pluginfile.php/?file='.
            $args = explode('/', ltrim($_SERVER['PATH_INFO'], '/'));
            $hasforcedslashargs = (count($args) > 2); // Always at least: context, component and filearea.
        }
    }
    if (!$hasforcedslashargs) {
        $relativepath = optional_param('file', false, PARAM_PATH);
    }

    if ($relativepath !== false and $relativepath !== '') {
        return $relativepath;
    }
    $relativepath = false;

    // Then try extract file from the slasharguments.
    if (stripos($_SERVER['SERVER_SOFTWARE'], 'iis') !== false) {
        // NOTE: IIS tends to convert all file paths to single byte DOS encoding,
        //       we can not use other methods because they break unicode chars,
        //       the only ways are to use URL rewriting
        //       OR
        //       to properly set the 'FastCGIUtf8ServerVariables' registry key.
        if (isset($_SERVER['PATH_INFO']) and $_SERVER['PATH_INFO'] !== '') {
            // Check that PATH_INFO works == must not contain the script name.
            if (strpos($_SERVER['PATH_INFO'], $SCRIPT) === false) {
                $relativepath = clean_param(urldecode($_SERVER['PATH_INFO']), PARAM_PATH);
            }
        }
    } else {
        // All other apache-like servers depend on PATH_INFO.
        if (isset($_SERVER['PATH_INFO'])) {
            if (isset($_SERVER['SCRIPT_NAME']) and strpos($_SERVER['PATH_INFO'], $_SERVER['SCRIPT_NAME']) === 0) {
                $relativepath = substr($_SERVER['PATH_INFO'], strlen($_SERVER['SCRIPT_NAME']));
            } else {
                $relativepath = $_SERVER['PATH_INFO'];
            }
            $relativepath = clean_param($relativepath, PARAM_PATH);
        }
    }

    return $relativepath;
}

/**
 * Just returns an array of text formats suitable for a popup menu
 *
 * @return array
 */
function format_text_menu() {
    return array (FORMAT_MOODLE => get_string('formattext'),
                  FORMAT_HTML => get_string('formathtml'),
                  FORMAT_PLAIN => get_string('formatplain'),
                  FORMAT_MARKDOWN => get_string('formatmarkdown'));
}

/**
 * Given text in a variety of format codings, this function returns the text as safe HTML.
 *
 * This function should mainly be used for long strings like posts,
 * answers, glossary items etc. For short strings {@link format_string()}.
 *
 * <pre>
 * Options:
 *      trusted     :   If true the string won't be cleaned. Default false required noclean=true.
 *      noclean     :   If true the string won't be cleaned, unless $CFG->forceclean is set. Default false required trusted=true.
 *      nocache     :   If true the strign will not be cached and will be formatted every call. Default false.
 *      filter      :   If true the string will be run through applicable filters as well. Default true.
 *      para        :   If true then the returned string will be wrapped in div tags. Default true.
 *      newlines    :   If true then lines newline breaks will be converted to HTML newline breaks. Default true.
 *      context     :   The context that will be used for filtering.
 *      overflowdiv :   If set to true the formatted text will be encased in a div
 *                      with the class no-overflow before being returned. Default false.
 *      allowid     :   If true then id attributes will not be removed, even when
 *                      using htmlpurifier. Default false.
 *      blanktarget :   If true all <a> tags will have target="_blank" added unless target is explicitly specified.
 * </pre>
 *
 * @staticvar array $croncache
 * @param string $text The text to be formatted. This is raw text originally from user input.
 * @param int $format Identifier of the text format to be used
 *            [FORMAT_MOODLE, FORMAT_HTML, FORMAT_PLAIN, FORMAT_MARKDOWN]
 * @param object/array $options text formatting options
 * @param int $courseiddonotuse deprecated course id, use context option instead
 * @return string
 */
function format_text($text, $format = FORMAT_MOODLE, $options = null, $courseiddonotuse = null) {
    global $CFG, $DB, $PAGE;

    if ($text === '' || is_null($text)) {
        // No need to do any filters and cleaning.
        return '';
    }

    // Detach object, we can not modify it.
    $options = (array)$options;

    if (!isset($options['trusted'])) {
        $options['trusted'] = false;
    }
    if (!isset($options['noclean'])) {
        if ($options['trusted'] and trusttext_active()) {
            // No cleaning if text trusted and noclean not specified.
            $options['noclean'] = true;
        } else {
            $options['noclean'] = false;
        }
    }
    if (!empty($CFG->forceclean)) {
        // Whatever the caller claims, the admin wants all content cleaned anyway.
        $options['noclean'] = false;
    }
    if (!isset($options['nocache'])) {
        $options['nocache'] = false;
    }
    if (!isset($options['filter'])) {
        $options['filter'] = true;
    }
    if (!isset($options['para'])) {
        $options['para'] = true;
    }
    if (!isset($options['newlines'])) {
        $options['newlines'] = true;
    }
    if (!isset($options['overflowdiv'])) {
        $options['overflowdiv'] = false;
    }
    $options['blanktarget'] = !empty($options['blanktarget']);

    // Calculate best context.
    if (empty($CFG->version) or $CFG->version < 2013051400 or during_initial_install()) {
        // Do not filter anything during installation or before upgrade completes.
        $context = null;

    } else if (isset($options['context'])) { // First by explicit passed context option.
        if (is_object($options['context'])) {
            $context = $options['context'];
        } else {
            $context = context::instance_by_id($options['context']);
        }
    } else if ($courseiddonotuse) {
        // Legacy courseid.
        $context = context_course::instance($courseiddonotuse);
    } else {
        // Fallback to $PAGE->context this may be problematic in CLI and other non-standard pages :-(.
        $context = $PAGE->context;
    }

    if (!$context) {
        // Either install/upgrade or something has gone really wrong because context does not exist (yet?).
        $options['nocache'] = true;
        $options['filter']  = false;
    }

    if ($options['filter']) {
        $filtermanager = filter_manager::instance();
        $filtermanager->setup_page_for_filters($PAGE, $context); // Setup global stuff filters may have.
        $filteroptions = array(
            'originalformat' => $format,
            'noclean' => $options['noclean'],
        );
    } else {
        $filtermanager = new null_filter_manager();
        $filteroptions = array();
    }

    switch ($format) {
        case FORMAT_HTML:
            if (!$options['noclean']) {
                $text = clean_text($text, FORMAT_HTML, $options);
            }
            $text = $filtermanager->filter_text($text, $context, $filteroptions);
            break;

        case FORMAT_PLAIN:
            $text = s($text); // Cleans dangerous JS.
            $text = rebuildnolinktag($text);
            $text = str_replace('  ', '&nbsp; ', $text);
            $text = nl2br($text);
            break;

        case FORMAT_WIKI:
            // This format is deprecated.
            $text = '<p>NOTICE: Wiki-like formatting has been removed from Moodle.  You should not be seeing
                     this message as all texts should have been converted to Markdown format instead.
                     Please post a bug report to http://moodle.org/bugs with information about where you
                     saw this message.</p>'.s($text);
            break;

        case FORMAT_MARKDOWN:
            $text = markdown_to_html($text);
            if (!$options['noclean']) {
                $text = clean_text($text, FORMAT_HTML, $options);
            }
            $text = $filtermanager->filter_text($text, $context, $filteroptions);
            break;

        default:  // FORMAT_MOODLE or anything else.
            $text = text_to_html($text, null, $options['para'], $options['newlines']);
            if (!$options['noclean']) {
                $text = clean_text($text, FORMAT_HTML, $options);
            }
            $text = $filtermanager->filter_text($text, $context, $filteroptions);
            break;
    }
    if ($options['filter']) {
        // At this point there should not be any draftfile links any more,
        // this happens when developers forget to post process the text.
        // The only potential problem is that somebody might try to format
        // the text before storing into database which would be itself big bug..
        $text = str_replace("\"$CFG->wwwroot/draftfile.php", "\"$CFG->wwwroot/brokenfile.php#", $text);

        if ($CFG->debugdeveloper) {
            if (strpos($text, '@@PLUGINFILE@@/') !== false) {
                debugging('Before calling format_text(), the content must be processed with file_rewrite_pluginfile_urls()',
                    DEBUG_DEVELOPER);
            }
        }
    }

    if (!empty($options['overflowdiv'])) {
        $text = html_writer::tag('div', $text, array('class' => 'no-overflow'));
    }

    if ($options['blanktarget']) {
        $domdoc = new DOMDocument();
        libxml_use_internal_errors(true);
        $domdoc->loadHTML('<?xml version="1.0" encoding="UTF-8" ?>' . $text);
        libxml_clear_errors();
        foreach ($domdoc->getElementsByTagName('a') as $link) {
            if ($link->hasAttribute('target') && strpos($link->getAttribute('target'), '_blank') === false) {
                continue;
            }
            $link->setAttribute('target', '_blank');
            if (strpos($link->getAttribute('rel'), 'noreferrer') === false) {
                $link->setAttribute('rel', trim($link->getAttribute('rel') . ' noreferrer'));
            }
        }

        // This regex is nasty and I don't like it. The correct way to solve this is by loading the HTML like so:
        // $domdoc->loadHTML($text, LIBXML_HTML_NOIMPLIED | LIBXML_HTML_NODEFDTD); however it seems like the libxml
        // version that travis uses doesn't work properly and ends up leaving <html><body>, so I'm forced to use
        // this regex to remove those tags.
        $text = trim(preg_replace('~<(?:!DOCTYPE|/?(?:html|body))[^>]*>\s*~i', '', $domdoc->saveHTML($domdoc->documentElement)));
    }

    return $text;
}

/**
 * Resets some data related to filters, called during upgrade or when general filter settings change.
 *
 * @param bool $phpunitreset true means called from our PHPUnit integration test reset
 * @return void
 */
function reset_text_filters_cache($phpunitreset = false) {
    global $CFG, $DB;

    if ($phpunitreset) {
        // HTMLPurifier does not change, DB is already reset to defaults,
        // nothing to do here, the dataroot was cleared too.
        return;
    }

    // The purge_all_caches() deals with cachedir and localcachedir purging,
    // the individual filter caches are invalidated as necessary elsewhere.

    // Update $CFG->filterall cache flag.
    if (empty($CFG->stringfilters)) {
        set_config('filterall', 0);
        return;
    }
    $installedfilters = core_component::get_plugin_list('filter');
    $filters = explode(',', $CFG->stringfilters);
    foreach ($filters as $filter) {
        if (isset($installedfilters[$filter])) {
            set_config('filterall', 1);
            return;
        }
    }
    set_config('filterall', 0);
}

/**
 * Given a simple string, this function returns the string
 * processed by enabled string filters if $CFG->filterall is enabled
 *
 * This function should be used to print short strings (non html) that
 * need filter processing e.g. activity titles, post subjects,
 * glossary concepts.
 *
 * @staticvar bool $strcache
 * @param string $string The string to be filtered. Should be plain text, expect
 * possibly for multilang tags.
 * @param boolean $striplinks To strip any link in the result text. Moodle 1.8 default changed from false to true! MDL-8713
 * @param array $options options array/object or courseid
 * @return string
 */
function format_string($string, $striplinks = true, $options = null) {
    global $CFG, $PAGE;

    // We'll use a in-memory cache here to speed up repeated strings.
    static $strcache = false;

    if (empty($CFG->version) or $CFG->version < 2013051400 or during_initial_install()) {
        // Do not filter anything during installation or before upgrade completes.
        return $string = strip_tags($string);
    }

    if ($strcache === false or count($strcache) > 2000) {
        // This number might need some tuning to limit memory usage in cron.
        $strcache = array();
    }

    if (is_numeric($options)) {
        // Legacy courseid usage.
        $options  = array('context' => context_course::instance($options));
    } else {
        // Detach object, we can not modify it.
        $options = (array)$options;
    }

    if (empty($options['context'])) {
        // Fallback to $PAGE->context this may be problematic in CLI and other non-standard pages :-(.
        $options['context'] = $PAGE->context;
    } else if (is_numeric($options['context'])) {
        $options['context'] = context::instance_by_id($options['context']);
    }
    if (!isset($options['filter'])) {
        $options['filter'] = true;
    }

    $options['escape'] = !isset($options['escape']) || $options['escape'];

    if (!$options['context']) {
        // We did not find any context? weird.
        return $string = strip_tags($string);
    }

    // Calculate md5.
    $cachekeys = array($string, $striplinks, $options['context']->id,
        $options['escape'], current_language(), $options['filter']);
    $md5 = md5(implode('<+>', $cachekeys));

    // Fetch from cache if possible.
    if (isset($strcache[$md5])) {
        return $strcache[$md5];
    }

    // First replace all ampersands not followed by html entity code
    // Regular expression moved to its own method for easier unit testing.
    $string = $options['escape'] ? replace_ampersands_not_followed_by_entity($string) : $string;

    if (!empty($CFG->filterall) && $options['filter']) {
        $filtermanager = filter_manager::instance();
        $filtermanager->setup_page_for_filters($PAGE, $options['context']); // Setup global stuff filters may have.
        $string = $filtermanager->filter_string($string, $options['context']);
    }

    // If the site requires it, strip ALL tags from this string.
    if (!empty($CFG->formatstringstriptags)) {
        if ($options['escape']) {
            $string = str_replace(array('<', '>'), array('&lt;', '&gt;'), strip_tags($string));
        } else {
            $string = strip_tags($string);
        }
    } else {
        // Otherwise strip just links if that is required (default).
        if ($striplinks) {
            // Strip links in string.
            $string = strip_links($string);
        }
        $string = clean_text($string);
    }

    // Store to cache.
    $strcache[$md5] = $string;

    return $string;
}

/**
 * Given a string, performs a negative lookahead looking for any ampersand character
 * that is not followed by a proper HTML entity. If any is found, it is replaced
 * by &amp;. The string is then returned.
 *
 * @param string $string
 * @return string
 */
function replace_ampersands_not_followed_by_entity($string) {
    return preg_replace("/\&(?![a-zA-Z0-9#]{1,8};)/", "&amp;", $string);
}

/**
 * Given a string, replaces all <a>.*</a> by .* and returns the string.
 *
 * @param string $string
 * @return string
 */
function strip_links($string) {
    return preg_replace('/(<a\s[^>]+?>)(.+?)(<\/a>)/is', '$2', $string);
}

/**
 * This expression turns links into something nice in a text format. (Russell Jungwirth)
 *
 * @param string $string
 * @return string
 */
function wikify_links($string) {
    return preg_replace('~(<a [^<]*href=["|\']?([^ "\']*)["|\']?[^>]*>([^<]*)</a>)~i', '$3 [ $2 ]', $string);
}

/**
 * Given text in a variety of format codings, this function returns the text as plain text suitable for plain email.
 *
 * @param string $text The text to be formatted. This is raw text originally from user input.
 * @param int $format Identifier of the text format to be used
 *            [FORMAT_MOODLE, FORMAT_HTML, FORMAT_PLAIN, FORMAT_WIKI, FORMAT_MARKDOWN]
 * @return string
 */
function format_text_email($text, $format) {

    switch ($format) {

        case FORMAT_PLAIN:
            return $text;
            break;

        case FORMAT_WIKI:
            // There should not be any of these any more!
            $text = wikify_links($text);
            return core_text::entities_to_utf8(strip_tags($text), true);
            break;

        case FORMAT_HTML:
            return html_to_text($text);
            break;

        case FORMAT_MOODLE:
        case FORMAT_MARKDOWN:
        default:
            $text = wikify_links($text);
            return core_text::entities_to_utf8(strip_tags($text), true);
            break;
    }
}

/**
 * Formats activity intro text
 *
 * @param string $module name of module
 * @param object $activity instance of activity
 * @param int $cmid course module id
 * @param bool $filter filter resulting html text
 * @return string
 */
function format_module_intro($module, $activity, $cmid, $filter=true) {
    global $CFG;
    require_once("$CFG->libdir/filelib.php");
    $context = context_module::instance($cmid);
    $options = array('noclean' => true, 'para' => false, 'filter' => $filter, 'context' => $context, 'overflowdiv' => true);
    $intro = file_rewrite_pluginfile_urls($activity->intro, 'pluginfile.php', $context->id, 'mod_'.$module, 'intro', null);
    return trim(format_text($intro, $activity->introformat, $options, null));
}

/**
 * Removes the usage of Moodle files from a text.
 *
 * In some rare cases we need to re-use a text that already has embedded links
 * to some files hosted within Moodle. But the new area in which we will push
 * this content does not support files... therefore we need to remove those files.
 *
 * @param string $source The text
 * @return string The stripped text
 */
function strip_pluginfile_content($source) {
    $baseurl = '@@PLUGINFILE@@';
    // Looking for something like < .* "@@pluginfile@@.*" .* >
    $pattern = '$<[^<>]+["\']' . $baseurl . '[^"\']*["\'][^<>]*>$';
    $stripped = preg_replace($pattern, '', $source);
    // Use purify html to rebalence potentially mismatched tags and generally cleanup.
    return purify_html($stripped);
}

/**
 * Legacy function, used for cleaning of old forum and glossary text only.
 *
 * @param string $text text that may contain legacy TRUSTTEXT marker
 * @return string text without legacy TRUSTTEXT marker
 */
function trusttext_strip($text) {
    if (!is_string($text)) {
        // This avoids the potential for an endless loop below.
        throw new coding_exception('trusttext_strip parameter must be a string');
    }
    while (true) { // Removing nested TRUSTTEXT.
        $orig = $text;
        $text = str_replace('#####TRUSTTEXT#####', '', $text);
        if (strcmp($orig, $text) === 0) {
            return $text;
        }
    }
}

/**
 * Must be called before editing of all texts with trust flag. Removes all XSS nasties from texts stored in database if needed.
 *
 * @param stdClass $object data object with xxx, xxxformat and xxxtrust fields
 * @param string $field name of text field
 * @param context $context active context
 * @return stdClass updated $object
 */
function trusttext_pre_edit($object, $field, $context) {
    $trustfield  = $field.'trust';
    $formatfield = $field.'format';

    if (!$object->$trustfield or !trusttext_trusted($context)) {
        $object->$field = clean_text($object->$field, $object->$formatfield);
    }

    return $object;
}

/**
 * Is current user trusted to enter no dangerous XSS in this context?
 *
 * Please note the user must be in fact trusted everywhere on this server!!
 *
 * @param context $context
 * @return bool true if user trusted
 */
function trusttext_trusted($context) {
    return (trusttext_active() and has_capability('moodle/site:trustcontent', $context));
}

/**
 * Is trusttext feature active?
 *
 * @return bool
 */
function trusttext_active() {
    global $CFG;

    return !empty($CFG->enabletrusttext);
}

/**
 * Cleans raw text removing nasties.
 *
 * Given raw text (eg typed in by a user) this function cleans it up and removes any nasty tags that could mess up
 * Moodle pages through XSS attacks.
 *
 * The result must be used as a HTML text fragment, this function can not cleanup random
 * parts of html tags such as url or src attributes.
 *
 * NOTE: the format parameter was deprecated because we can safely clean only HTML.
 *
 * @param string $text The text to be cleaned
 * @param int|string $format deprecated parameter, should always contain FORMAT_HTML or FORMAT_MOODLE
 * @param array $options Array of options; currently only option supported is 'allowid' (if true,
 *   does not remove id attributes when cleaning)
 * @return string The cleaned up text
 */
function clean_text($text, $format = FORMAT_HTML, $options = array()) {
    $text = (string)$text;

    if ($format != FORMAT_HTML and $format != FORMAT_HTML) {
        // TODO: we need to standardise cleanup of text when loading it into editor first.
        // debugging('clean_text() is designed to work only with html');.
    }

    if ($format == FORMAT_PLAIN) {
        return $text;
    }

    if (is_purify_html_necessary($text)) {
        $text = purify_html($text, $options);
    }

    // Originally we tried to neutralise some script events here, it was a wrong approach because
    // it was trivial to work around that (for example using style based XSS exploits).
    // We must not give false sense of security here - all developers MUST understand how to use
    // rawurlencode(), htmlentities(), htmlspecialchars(), p(), s(), moodle_url, html_writer and friends!!!

    return $text;
}

/**
 * Is it necessary to use HTMLPurifier?
 *
 * @private
 * @param string $text
 * @return bool false means html is safe and valid, true means use HTMLPurifier
 */
function is_purify_html_necessary($text) {
    if ($text === '') {
        return false;
    }

    if ($text === (string)((int)$text)) {
        return false;
    }

    if (strpos($text, '&') !== false or preg_match('|<[^pesb/]|', $text)) {
        // We need to normalise entities or other tags except p, em, strong and br present.
        return true;
    }

    $altered = htmlspecialchars($text, ENT_NOQUOTES, 'UTF-8', true);
    if ($altered === $text) {
        // No < > or other special chars means this must be safe.
        return false;
    }

    // Let's try to convert back some safe html tags.
    $altered = preg_replace('|&lt;p&gt;(.*?)&lt;/p&gt;|m', '<p>$1</p>', $altered);
    if ($altered === $text) {
        return false;
    }
    $altered = preg_replace('|&lt;em&gt;([^<>]+?)&lt;/em&gt;|m', '<em>$1</em>', $altered);
    if ($altered === $text) {
        return false;
    }
    $altered = preg_replace('|&lt;strong&gt;([^<>]+?)&lt;/strong&gt;|m', '<strong>$1</strong>', $altered);
    if ($altered === $text) {
        return false;
    }
    $altered = str_replace('&lt;br /&gt;', '<br />', $altered);
    if ($altered === $text) {
        return false;
    }

    return true;
}

/**
 * KSES replacement cleaning function - uses HTML Purifier.
 *
 * @param string $text The (X)HTML string to purify
 * @param array $options Array of options; currently only option supported is 'allowid' (if set,
 *   does not remove id attributes when cleaning)
 * @return string
 */
function purify_html($text, $options = array()) {
    global $CFG;

    $text = (string)$text;

    static $purifiers = array();
    static $caches = array();

    // Purifier code can change only during major version upgrade.
    $version = empty($CFG->version) ? 0 : $CFG->version;
    $cachedir = "$CFG->localcachedir/htmlpurifier/$version";
    if (!file_exists($cachedir)) {
        // Purging of caches may remove the cache dir at any time,
        // luckily file_exists() results should be cached for all existing directories.
        $purifiers = array();
        $caches = array();
        gc_collect_cycles();

        make_localcache_directory('htmlpurifier', false);
        check_dir_exists($cachedir);
    }

    $allowid = empty($options['allowid']) ? 0 : 1;
    $allowobjectembed = empty($CFG->allowobjectembed) ? 0 : 1;

    $type = 'type_'.$allowid.'_'.$allowobjectembed;

    if (!array_key_exists($type, $caches)) {
        $caches[$type] = cache::make('core', 'htmlpurifier', array('type' => $type));
    }
    $cache = $caches[$type];

    // Add revision number and all options to the text key so that it is compatible with local cluster node caches.
    $key = "|$version|$allowobjectembed|$allowid|$text";
    $filteredtext = $cache->get($key);

    if ($filteredtext === true) {
        // The filtering did not change the text last time, no need to filter anything again.
        return $text;
    } else if ($filteredtext !== false) {
        return $filteredtext;
    }

    if (empty($purifiers[$type])) {
        require_once $CFG->libdir.'/htmlpurifier/HTMLPurifier.safe-includes.php';
        require_once $CFG->libdir.'/htmlpurifier/locallib.php';
        $config = HTMLPurifier_Config::createDefault();

        $config->set('HTML.DefinitionID', 'moodlehtml');
        $config->set('HTML.DefinitionRev', 6);
        $config->set('Cache.SerializerPath', $cachedir);
        $config->set('Cache.SerializerPermissions', $CFG->directorypermissions);
        $config->set('Core.NormalizeNewlines', false);
        $config->set('Core.ConvertDocumentToFragment', true);
        $config->set('Core.Encoding', 'UTF-8');
        $config->set('HTML.Doctype', 'XHTML 1.0 Transitional');
        $config->set('URI.AllowedSchemes', array(
            'http' => true,
            'https' => true,
            'ftp' => true,
            'irc' => true,
            'nntp' => true,
            'news' => true,
            'rtsp' => true,
            'rtmp' => true,
            'teamspeak' => true,
            'gopher' => true,
            'mms' => true,
            'mailto' => true
        ));
        $config->set('Attr.AllowedFrameTargets', array('_blank'));

        if ($allowobjectembed) {
            $config->set('HTML.SafeObject', true);
            $config->set('Output.FlashCompat', true);
            $config->set('HTML.SafeEmbed', true);
        }

        if ($allowid) {
            $config->set('Attr.EnableID', true);
        }

        if ($def = $config->maybeGetRawHTMLDefinition()) {
            $def->addElement('nolink', 'Block', 'Flow', array());                       // Skip our filters inside.
            $def->addElement('tex', 'Inline', 'Inline', array());                       // Tex syntax, equivalent to $$xx$$.
            $def->addElement('algebra', 'Inline', 'Inline', array());                   // Algebra syntax, equivalent to @@xx@@.
            $def->addElement('lang', 'Block', 'Flow', array(), array('lang'=>'CDATA')); // Original multilang style - only our hacked lang attribute.
            $def->addAttribute('span', 'xxxlang', 'CDATA');                             // Current very problematic multilang.

            // Media elements.
            // https://html.spec.whatwg.org/#the-video-element
            $def->addElement('video', 'Block', 'Optional: #PCDATA | Flow | source | track', 'Common', [
                'src' => 'URI',
                'crossorigin' => 'Enum#anonymous,use-credentials',
                'poster' => 'URI',
                'preload' => 'Enum#auto,metadata,none',
                'autoplay' => 'Bool',
                'playsinline' => 'Bool',
                'loop' => 'Bool',
                'muted' => 'Bool',
                'controls' => 'Bool',
                'width' => 'Length',
                'height' => 'Length',
            ]);
            // https://html.spec.whatwg.org/#the-audio-element
            $def->addElement('audio', 'Block', 'Optional: #PCDATA | Flow | source | track', 'Common', [
                'src' => 'URI',
                'crossorigin' => 'Enum#anonymous,use-credentials',
                'preload' => 'Enum#auto,metadata,none',
                'autoplay' => 'Bool',
                'loop' => 'Bool',
                'muted' => 'Bool',
                'controls' => 'Bool'
            ]);
            // https://html.spec.whatwg.org/#the-source-element
            $def->addElement('source', false, 'Empty', null, [
                'src' => 'URI',
                'type' => 'Text'
            ]);
            // https://html.spec.whatwg.org/#the-track-element
            $def->addElement('track', false, 'Empty', null, [
                'src' => 'URI',
                'kind' => 'Enum#subtitles,captions,descriptions,chapters,metadata',
                'srclang' => 'Text',
                'label' => 'Text',
                'default' => 'Bool',
            ]);

            // Use the built-in Ruby module to add annotation support.
            $def->manager->addModule(new HTMLPurifier_HTMLModule_Ruby());
        }

        $purifier = new HTMLPurifier($config);
        $purifiers[$type] = $purifier;
    } else {
        $purifier = $purifiers[$type];
    }

    $multilang = (strpos($text, 'class="multilang"') !== false);

    $filteredtext = $text;
    if ($multilang) {
        $filteredtextregex = '/<span(\s+lang="([a-zA-Z0-9_-]+)"|\s+class="multilang"){2}\s*>/';
        $filteredtext = preg_replace($filteredtextregex, '<span xxxlang="${2}">', $filteredtext);
    }
    $filteredtext = (string)$purifier->purify($filteredtext);
    if ($multilang) {
        $filteredtext = preg_replace('/<span xxxlang="([a-zA-Z0-9_-]+)">/', '<span lang="${1}" class="multilang">', $filteredtext);
    }

    if ($text === $filteredtext) {
        // No need to store the filtered text, next time we will just return unfiltered text
        // because it was not changed by purifying.
        $cache->set($key, true);
    } else {
        $cache->set($key, $filteredtext);
    }

    return $filteredtext;
}

/**
 * Given plain text, makes it into HTML as nicely as possible.
 *
 * May contain HTML tags already.
 *
 * Do not abuse this function. It is intended as lower level formatting feature used
 * by {@link format_text()} to convert FORMAT_MOODLE to HTML. You are supposed
 * to call format_text() in most of cases.
 *
 * @param string $text The string to convert.
 * @param boolean $smileyignored Was used to determine if smiley characters should convert to smiley images, ignored now
 * @param boolean $para If true then the returned string will be wrapped in div tags
 * @param boolean $newlines If true then lines newline breaks will be converted to HTML newline breaks.
 * @return string
 */
function text_to_html($text, $smileyignored = null, $para = true, $newlines = true) {
    // Remove any whitespace that may be between HTML tags.
    $text = preg_replace("~>([[:space:]]+)<~i", "><", $text);

    // Remove any returns that precede or follow HTML tags.
    $text = preg_replace("~([\n\r])<~i", " <", $text);
    $text = preg_replace("~>([\n\r])~i", "> ", $text);

    // Make returns into HTML newlines.
    if ($newlines) {
        $text = nl2br($text);
    }

    // Wrap the whole thing in a div if required.
    if ($para) {
        // In 1.9 this was changed from a p => div.
        return '<div class="text_to_html">'.$text.'</div>';
    } else {
        return $text;
    }
}

/**
 * Given Markdown formatted text, make it into XHTML using external function
 *
 * @param string $text The markdown formatted text to be converted.
 * @return string Converted text
 */
function markdown_to_html($text) {
    global $CFG;

    if ($text === '' or $text === null) {
        return $text;
    }

    require_once($CFG->libdir .'/markdown/MarkdownInterface.php');
    require_once($CFG->libdir .'/markdown/Markdown.php');
    require_once($CFG->libdir .'/markdown/MarkdownExtra.php');

    return \Michelf\MarkdownExtra::defaultTransform($text);
}

/**
 * Given HTML text, make it into plain text using external function
 *
 * @param string $html The text to be converted.
 * @param integer $width Width to wrap the text at. (optional, default 75 which
 *      is a good value for email. 0 means do not limit line length.)
 * @param boolean $dolinks By default, any links in the HTML are collected, and
 *      printed as a list at the end of the HTML. If you don't want that, set this
 *      argument to false.
 * @return string plain text equivalent of the HTML.
 */
function html_to_text($html, $width = 75, $dolinks = true) {
    global $CFG;

    require_once($CFG->libdir .'/html2text/lib.php');

    $options = array(
        'width'     => $width,
        'do_links'  => 'table',
    );

    if (empty($dolinks)) {
        $options['do_links'] = 'none';
    }
    $h2t = new core_html2text($html, $options);
    $result = $h2t->getText();

    return $result;
}

/**
 * Converts texts or strings to plain text.
 *
 * - When used to convert user input introduced in an editor the text format needs to be passed in $contentformat like we usually
 *   do in format_text.
 * - When this function is used for strings that are usually passed through format_string before displaying them
 *   we need to set $contentformat to false. This will execute html_to_text as these strings can contain multilang tags if
 *   multilang filter is applied to headings.
 *
 * @param string $content The text as entered by the user
 * @param int|false $contentformat False for strings or the text format: FORMAT_MOODLE/FORMAT_HTML/FORMAT_PLAIN/FORMAT_MARKDOWN
 * @return string Plain text.
 */
function content_to_text($content, $contentformat) {

    switch ($contentformat) {
        case FORMAT_PLAIN:
            // Nothing here.
            break;
        case FORMAT_MARKDOWN:
            $content = markdown_to_html($content);
            $content = html_to_text($content, 75, false);
            break;
        default:
            // FORMAT_HTML, FORMAT_MOODLE and $contentformat = false, the later one are strings usually formatted through
            // format_string, we need to convert them from html because they can contain HTML (multilang filter).
            $content = html_to_text($content, 75, false);
    }

    return trim($content, "\r\n ");
}

/**
 * Factory method for extracting draft file links from arbitrary text using regular expressions. Only text
 * is required; other file fields may be passed to filter.
 *
 * @param string $text Some html content.
 * @param bool $forcehttps force https urls.
 * @param int $contextid This parameter and the next three identify the file area to save to.
 * @param string $component The component name.
 * @param string $filearea The filearea.
 * @param int $itemid The item id for the filearea.
 * @param string $filename The specific filename of the file.
 * @return array
 */
function extract_draft_file_urls_from_text($text, $forcehttps = false, $contextid = null, $component = null,
                                           $filearea = null, $itemid = null, $filename = null) {
    global $CFG;

    $wwwroot = $CFG->wwwroot;
    if ($forcehttps) {
        $wwwroot = str_replace('http://', 'https://', $wwwroot);
    }
    $urlstring = '/' . preg_quote($wwwroot, '/');

    $urlbase = preg_quote('draftfile.php');
    $urlstring .= "\/(?<urlbase>{$urlbase})";

    if (is_null($contextid)) {
        $contextid = '[0-9]+';
    }
    $urlstring .= "\/(?<contextid>{$contextid})";

    if (is_null($component)) {
        $component = '[a-z_]+';
    }
    $urlstring .= "\/(?<component>{$component})";

    if (is_null($filearea)) {
        $filearea = '[a-z_]+';
    }
    $urlstring .= "\/(?<filearea>{$filearea})";

    if (is_null($itemid)) {
        $itemid = '[0-9]+';
    }
    $urlstring .= "\/(?<itemid>{$itemid})";

    // Filename matching magic based on file_rewrite_urls_to_pluginfile().
    if (is_null($filename)) {
        $filename = '[^\'\",&<>|`\s:\\\\]+';
    }
    $urlstring .= "\/(?<filename>{$filename})/";

    // Regular expression which matches URLs and returns their components.
    preg_match_all($urlstring, $text, $urls, PREG_SET_ORDER);
    return $urls;
}

/**
 * This function will highlight search words in a given string
 *
 * It cares about HTML and will not ruin links.  It's best to use
 * this function after performing any conversions to HTML.
 *
 * @param string $needle The search string. Syntax like "word1 +word2 -word3" is dealt with correctly.
 * @param string $haystack The string (HTML) within which to highlight the search terms.
 * @param boolean $matchcase whether to do case-sensitive. Default case-insensitive.
 * @param string $prefix the string to put before each search term found.
 * @param string $suffix the string to put after each search term found.
 * @return string The highlighted HTML.
 */
function highlight($needle, $haystack, $matchcase = false,
        $prefix = '<span class="highlight">', $suffix = '</span>') {

    // Quick bail-out in trivial cases.
    if (empty($needle) or empty($haystack)) {
        return $haystack;
    }

    // Break up the search term into words, discard any -words and build a regexp.
    $words = preg_split('/ +/', trim($needle));
    foreach ($words as $index => $word) {
        if (strpos($word, '-') === 0) {
            unset($words[$index]);
        } else if (strpos($word, '+') === 0) {
            $words[$index] = '\b' . preg_quote(ltrim($word, '+'), '/') . '\b'; // Match only as a complete word.
        } else {
            $words[$index] = preg_quote($word, '/');
        }
    }
    $regexp = '/(' . implode('|', $words) . ')/u'; // Char u is to do UTF-8 matching.
    if (!$matchcase) {
        $regexp .= 'i';
    }

    // Another chance to bail-out if $search was only -words.
    if (empty($words)) {
        return $haystack;
    }

    // Split the string into HTML tags and real content.
    $chunks = preg_split('/((?:<[^>]*>)+)/', $haystack, -1, PREG_SPLIT_DELIM_CAPTURE);

    // We have an array of alternating blocks of text, then HTML tags, then text, ...
    // Loop through replacing search terms in the text, and leaving the HTML unchanged.
    $ishtmlchunk = false;
    $result = '';
    foreach ($chunks as $chunk) {
        if ($ishtmlchunk) {
            $result .= $chunk;
        } else {
            $result .= preg_replace($regexp, $prefix . '$1' . $suffix, $chunk);
        }
        $ishtmlchunk = !$ishtmlchunk;
    }

    return $result;
}

/**
 * This function will highlight instances of $needle in $haystack
 *
 * It's faster that the above function {@link highlight()} and doesn't care about
 * HTML or anything.
 *
 * @param string $needle The string to search for
 * @param string $haystack The string to search for $needle in
 * @return string The highlighted HTML
 */
function highlightfast($needle, $haystack) {

    if (empty($needle) or empty($haystack)) {
        return $haystack;
    }

    $parts = explode(core_text::strtolower($needle), core_text::strtolower($haystack));

    if (count($parts) === 1) {
        return $haystack;
    }

    $pos = 0;

    foreach ($parts as $key => $part) {
        $parts[$key] = substr($haystack, $pos, strlen($part));
        $pos += strlen($part);

        $parts[$key] .= '<span class="highlight">'.substr($haystack, $pos, strlen($needle)).'</span>';
        $pos += strlen($needle);
    }

    return str_replace('<span class="highlight"></span>', '', join('', $parts));
}

/**
 * Return a string containing 'lang', xml:lang and optionally 'dir' HTML attributes.
 *
 * Internationalisation, for print_header and backup/restorelib.
 *
 * @param bool $dir Default false
 * @return string Attributes
 */
function get_html_lang($dir = false) {
    $direction = '';
    if ($dir) {
        if (right_to_left()) {
            $direction = ' dir="rtl"';
        } else {
            $direction = ' dir="ltr"';
        }
    }
    // Accessibility: added the 'lang' attribute to $direction, used in theme <html> tag.
    $language = str_replace('_', '-', current_language());
    @header('Content-Language: '.$language);
    return ($direction.' lang="'.$language.'" xml:lang="'.$language.'"');
}


// STANDARD WEB PAGE PARTS.

/**
 * Send the HTTP headers that Moodle requires.
 *
 * There is a backwards compatibility hack for legacy code
 * that needs to add custom IE compatibility directive.
 *
 * Example:
 * <code>
 * if (!isset($CFG->additionalhtmlhead)) {
 *     $CFG->additionalhtmlhead = '';
 * }
 * $CFG->additionalhtmlhead .= '<meta http-equiv="X-UA-Compatible" content="IE=8" />';
 * header('X-UA-Compatible: IE=8');
 * echo $OUTPUT->header();
 * </code>
 *
 * Please note the $CFG->additionalhtmlhead alone might not work,
 * you should send the IE compatibility header() too.
 *
 * @param string $contenttype
 * @param bool $cacheable Can this page be cached on back?
 * @return void, sends HTTP headers
 */
function send_headers($contenttype, $cacheable = true) {
    global $CFG;

    @header('Content-Type: ' . $contenttype);
    @header('Content-Script-Type: text/javascript');
    @header('Content-Style-Type: text/css');

    if (empty($CFG->additionalhtmlhead) or stripos($CFG->additionalhtmlhead, 'X-UA-Compatible') === false) {
        @header('X-UA-Compatible: IE=edge');
    }

    if ($cacheable) {
        // Allow caching on "back" (but not on normal clicks).
        @header('Cache-Control: private, pre-check=0, post-check=0, max-age=0, no-transform');
        @header('Pragma: no-cache');
        @header('Expires: ');
    } else {
        // Do everything we can to always prevent clients and proxies caching.
        @header('Cache-Control: no-store, no-cache, must-revalidate');
        @header('Cache-Control: post-check=0, pre-check=0, no-transform', false);
        @header('Pragma: no-cache');
        @header('Expires: Mon, 20 Aug 1969 09:23:00 GMT');
        @header('Last-Modified: ' . gmdate('D, d M Y H:i:s') . ' GMT');
    }
    @header('Accept-Ranges: none');

    // The Moodle app must be allowed to embed content always.
    if (empty($CFG->allowframembedding) && !core_useragent::is_moodle_app()) {
        @header('X-Frame-Options: sameorigin');
    }
}

/**
 * Return the right arrow with text ('next'), and optionally embedded in a link.
 *
 * @param string $text HTML/plain text label (set to blank only for breadcrumb separator cases).
 * @param string $url An optional link to use in a surrounding HTML anchor.
 * @param bool $accesshide True if text should be hidden (for screen readers only).
 * @param string $addclass Additional class names for the link, or the arrow character.
 * @return string HTML string.
 */
function link_arrow_right($text, $url='', $accesshide=false, $addclass='', $addparams = []) {
    global $OUTPUT; // TODO: move to output renderer.
    $arrowclass = 'arrow ';
    if (!$url) {
        $arrowclass .= $addclass;
    }
    $arrow = '<span class="'.$arrowclass.'">'.$OUTPUT->rarrow().'</span>';
    $htmltext = '';
    if ($text) {
        $htmltext = '<span class="arrow_text">'.$text.'</span>&nbsp;';
        if ($accesshide) {
            $htmltext = get_accesshide($htmltext);
        }
    }
    if ($url) {
        $class = 'arrow_link';
        if ($addclass) {
            $class .= ' '.$addclass;
        }

        $linkparams = [
            'class' => $class,
            'href' => $url,
            'title' => preg_replace('/<.*?>/', '', $text),
        ];

        $linkparams += $addparams;

        return html_writer::link($url, $htmltext . $arrow, $linkparams);
    }
    return $htmltext.$arrow;
}

/**
 * Return the left arrow with text ('previous'), and optionally embedded in a link.
 *
 * @param string $text HTML/plain text label (set to blank only for breadcrumb separator cases).
 * @param string $url An optional link to use in a surrounding HTML anchor.
 * @param bool $accesshide True if text should be hidden (for screen readers only).
 * @param string $addclass Additional class names for the link, or the arrow character.
 * @return string HTML string.
 */
function link_arrow_left($text, $url='', $accesshide=false, $addclass='', $addparams = []) {
    global $OUTPUT; // TODO: move to utput renderer.
    $arrowclass = 'arrow ';
    if (! $url) {
        $arrowclass .= $addclass;
    }
    $arrow = '<span class="'.$arrowclass.'">'.$OUTPUT->larrow().'</span>';
    $htmltext = '';
    if ($text) {
        $htmltext = '&nbsp;<span class="arrow_text">'.$text.'</span>';
        if ($accesshide) {
            $htmltext = get_accesshide($htmltext);
        }
    }
    if ($url) {
        $class = 'arrow_link';
        if ($addclass) {
            $class .= ' '.$addclass;
        }

        $linkparams = [
            'class' => $class,
            'href' => $url,
            'title' => preg_replace('/<.*?>/', '', $text),
        ];

        $linkparams += $addparams;

        return html_writer::link($url, $arrow . $htmltext, $linkparams);
    }
    return $arrow.$htmltext;
}

/**
 * Return a HTML element with the class "accesshide", for accessibility.
 *
 * Please use cautiously - where possible, text should be visible!
 *
 * @param string $text Plain text.
 * @param string $elem Lowercase element name, default "span".
 * @param string $class Additional classes for the element.
 * @param string $attrs Additional attributes string in the form, "name='value' name2='value2'"
 * @return string HTML string.
 */
function get_accesshide($text, $elem='span', $class='', $attrs='') {
    return "<$elem class=\"accesshide $class\" $attrs>$text</$elem>";
}

/**
 * Return the breadcrumb trail navigation separator.
 *
 * @return string HTML string.
 */
function get_separator() {
    // Accessibility: the 'hidden' slash is preferred for screen readers.
    return ' '.link_arrow_right($text='/', $url='', $accesshide=true, 'sep').' ';
}

/**
 * Print (or return) a collapsible region, that has a caption that can be clicked to expand or collapse the region.
 *
 * If JavaScript is off, then the region will always be expanded.
 *
 * @param string $contents the contents of the box.
 * @param string $classes class names added to the div that is output.
 * @param string $id id added to the div that is output. Must not be blank.
 * @param string $caption text displayed at the top. Clicking on this will cause the region to expand or contract.
 * @param string $userpref the name of the user preference that stores the user's preferred default state.
 *      (May be blank if you do not wish the state to be persisted.
 * @param boolean $default Initial collapsed state to use if the user_preference it not set.
 * @param boolean $return if true, return the HTML as a string, rather than printing it.
 * @return string|void If $return is false, returns nothing, otherwise returns a string of HTML.
 */
function print_collapsible_region($contents, $classes, $id, $caption, $userpref = '', $default = false, $return = false) {
    $output  = print_collapsible_region_start($classes, $id, $caption, $userpref, $default, true);
    $output .= $contents;
    $output .= print_collapsible_region_end(true);

    if ($return) {
        return $output;
    } else {
        echo $output;
    }
}

/**
 * Print (or return) the start of a collapsible region
 *
 * The collapsibleregion has a caption that can be clicked to expand or collapse the region. If JavaScript is off, then the region
 * will always be expanded.
 *
 * @param string $classes class names added to the div that is output.
 * @param string $id id added to the div that is output. Must not be blank.
 * @param string $caption text displayed at the top. Clicking on this will cause the region to expand or contract.
 * @param string $userpref the name of the user preference that stores the user's preferred default state.
 *      (May be blank if you do not wish the state to be persisted.
 * @param boolean $default Initial collapsed state to use if the user_preference it not set.
 * @param boolean $return if true, return the HTML as a string, rather than printing it.
 * @return string|void if $return is false, returns nothing, otherwise returns a string of HTML.
 */
function print_collapsible_region_start($classes, $id, $caption, $userpref = '', $default = false, $return = false) {
    global $PAGE;

    // Work out the initial state.
    if (!empty($userpref) and is_string($userpref)) {
        user_preference_allow_ajax_update($userpref, PARAM_BOOL);
        $collapsed = get_user_preferences($userpref, $default);
    } else {
        $collapsed = $default;
        $userpref = false;
    }

    if ($collapsed) {
        $classes .= ' collapsed';
    }

    $output = '';
    $output .= '<div id="' . $id . '" class="collapsibleregion ' . $classes . '">';
    $output .= '<div id="' . $id . '_sizer">';
    $output .= '<div id="' . $id . '_caption" class="collapsibleregioncaption">';
    $output .= $caption . ' ';
    $output .= '</div><div id="' . $id . '_inner" class="collapsibleregioninner">';
    $PAGE->requires->js_init_call('M.util.init_collapsible_region', array($id, $userpref, get_string('clicktohideshow')));

    if ($return) {
        return $output;
    } else {
        echo $output;
    }
}

/**
 * Close a region started with print_collapsible_region_start.
 *
 * @param boolean $return if true, return the HTML as a string, rather than printing it.
 * @return string|void if $return is false, returns nothing, otherwise returns a string of HTML.
 */
function print_collapsible_region_end($return = false) {
    $output = '</div></div></div>';

    if ($return) {
        return $output;
    } else {
        echo $output;
    }
}

/**
 * Print a specified group's avatar.
 *
 * @param array|stdClass $group A single {@link group} object OR array of groups.
 * @param int $courseid The course ID.
 * @param boolean $large Default small picture, or large.
 * @param boolean $return If false print picture, otherwise return the output as string
 * @param boolean $link Enclose image in a link to view specified course?
 * @param boolean $includetoken Whether to use a user token when displaying this group image.
 *                True indicates to generate a token for current user, and integer value indicates to generate a token for the
 *                user whose id is the value indicated.
 *                If the group picture is included in an e-mail or some other location where the audience is a specific
 *                user who will not be logged in when viewing, then we use a token to authenticate the user.
 * @return string|void Depending on the setting of $return
 */
function print_group_picture($group, $courseid, $large = false, $return = false, $link = true, $includetoken = false) {
    global $CFG;

    if (is_array($group)) {
        $output = '';
        foreach ($group as $g) {
            $output .= print_group_picture($g, $courseid, $large, true, $link, $includetoken);
        }
        if ($return) {
            return $output;
        } else {
            echo $output;
            return;
        }
    }

    $pictureurl = get_group_picture_url($group, $courseid, $large, $includetoken);

    // If there is no picture, do nothing.
    if (!isset($pictureurl)) {
        return;
    }

    $context = context_course::instance($courseid);

    $groupname = s($group->name);
    $pictureimage = html_writer::img($pictureurl, $groupname, ['title' => $groupname]);

    $output = '';
    if ($link or has_capability('moodle/site:accessallgroups', $context)) {
        $linkurl = new moodle_url('/user/index.php', ['id' => $courseid, 'group' => $group->id]);
        $output .= html_writer::link($linkurl, $pictureimage);
    } else {
        $output .= $pictureimage;
    }

    if ($return) {
        return $output;
    } else {
        echo $output;
    }
}

/**
 * Return the url to the group picture.
 *
 * @param  stdClass $group A group object.
 * @param  int $courseid The course ID for the group.
 * @param  bool $large A large or small group picture? Default is small.
 * @param  boolean $includetoken Whether to use a user token when displaying this group image.
 *                 True indicates to generate a token for current user, and integer value indicates to generate a token for the
 *                 user whose id is the value indicated.
 *                 If the group picture is included in an e-mail or some other location where the audience is a specific
 *                 user who will not be logged in when viewing, then we use a token to authenticate the user.
 * @return moodle_url Returns the url for the group picture.
 */
function get_group_picture_url($group, $courseid, $large = false, $includetoken = false) {
    global $CFG;

    $context = context_course::instance($courseid);

    // If there is no picture, do nothing.
    if (!$group->picture) {
        return;
    }

    // If picture is hidden, only show to those with course:managegroups.
    if ($group->hidepicture and !has_capability('moodle/course:managegroups', $context)) {
        return;
    }

    if ($large) {
        $file = 'f1';
    } else {
        $file = 'f2';
    }

    $grouppictureurl = moodle_url::make_pluginfile_url(
            $context->id, 'group', 'icon', $group->id, '/', $file, false, $includetoken);
    $grouppictureurl->param('rev', $group->picture);
    return $grouppictureurl;
}


/**
 * Display a recent activity note
 *
 * @staticvar string $strftimerecent
 * @param int $time A timestamp int.
 * @param stdClass $user A user object from the database.
 * @param string $text Text for display for the note
 * @param string $link The link to wrap around the text
 * @param bool $return If set to true the HTML is returned rather than echo'd
 * @param string $viewfullnames
 * @return string If $retrun was true returns HTML for a recent activity notice.
 */
function print_recent_activity_note($time, $user, $text, $link, $return=false, $viewfullnames=null) {
    static $strftimerecent = null;
    $output = '';

    if (is_null($viewfullnames)) {
        $context = context_system::instance();
        $viewfullnames = has_capability('moodle/site:viewfullnames', $context);
    }

    if (is_null($strftimerecent)) {
        $strftimerecent = get_string('strftimerecent');
    }

    $output .= '<div class="head">';
    $output .= '<div class="date">'.userdate($time, $strftimerecent).'</div>';
    $output .= '<div class="name">'.fullname($user, $viewfullnames).'</div>';
    $output .= '</div>';
    $output .= '<div class="info"><a href="'.$link.'">'.format_string($text, true).'</a></div>';

    if ($return) {
        return $output;
    } else {
        echo $output;
    }
}

/**
 * Returns a popup menu with course activity modules
 *
 * Given a course this function returns a small popup menu with all the course activity modules in it, as a navigation menu
 * outputs a simple list structure in XHTML.
 * The data is taken from the serialised array stored in the course record.
 *
 * @param course $course A {@link $COURSE} object.
 * @param array $sections
 * @param course_modinfo $modinfo
 * @param string $strsection
 * @param string $strjumpto
 * @param int $width
 * @param string $cmid
 * @return string The HTML block
 */
function navmenulist($course, $sections, $modinfo, $strsection, $strjumpto, $width=50, $cmid=0) {

    global $CFG, $OUTPUT;

    $section = -1;
    $menu = array();
    $doneheading = false;

    $courseformatoptions = course_get_format($course)->get_format_options();
    $coursecontext = context_course::instance($course->id);

    $menu[] = '<ul class="navmenulist"><li class="jumpto section"><span>'.$strjumpto.'</span><ul>';
    foreach ($modinfo->cms as $mod) {
        if (!$mod->has_view()) {
            // Don't show modules which you can't link to!
            continue;
        }

        // For course formats using 'numsections' do not show extra sections.
        if (isset($courseformatoptions['numsections']) && $mod->sectionnum > $courseformatoptions['numsections']) {
            break;
        }

        if (!$mod->uservisible) { // Do not icnlude empty sections at all.
            continue;
        }

        if ($mod->sectionnum >= 0 and $section != $mod->sectionnum) {
            $thissection = $sections[$mod->sectionnum];

            if ($thissection->visible or
                    (isset($courseformatoptions['hiddensections']) and !$courseformatoptions['hiddensections']) or
                    has_capability('moodle/course:viewhiddensections', $coursecontext)) {
                $thissection->summary = strip_tags(format_string($thissection->summary, true));
                if (!$doneheading) {
                    $menu[] = '</ul></li>';
                }
                if ($course->format == 'weeks' or empty($thissection->summary)) {
                    $item = $strsection ." ". $mod->sectionnum;
                } else {
                    if (core_text::strlen($thissection->summary) < ($width-3)) {
                        $item = $thissection->summary;
                    } else {
                        $item = core_text::substr($thissection->summary, 0, $width).'...';
                    }
                }
                $menu[] = '<li class="section"><span>'.$item.'</span>';
                $menu[] = '<ul>';
                $doneheading = true;

                $section = $mod->sectionnum;
            } else {
                // No activities from this hidden section shown.
                continue;
            }
        }

        $url = $mod->modname .'/view.php?id='. $mod->id;
        $mod->name = strip_tags(format_string($mod->name ,true));
        if (core_text::strlen($mod->name) > ($width+5)) {
            $mod->name = core_text::substr($mod->name, 0, $width).'...';
        }
        if (!$mod->visible) {
            $mod->name = '('.$mod->name.')';
        }
        $class = 'activity '.$mod->modname;
        $class .= ($cmid == $mod->id) ? ' selected' : '';
        $menu[] = '<li class="'.$class.'">'.
                  $OUTPUT->image_icon('icon', '', $mod->modname).
                  '<a href="'.$CFG->wwwroot.'/mod/'.$url.'">'.$mod->name.'</a></li>';
    }

    if ($doneheading) {
        $menu[] = '</ul></li>';
    }
    $menu[] = '</ul></li></ul>';

    return implode("\n", $menu);
}

/**
 * Prints a grade menu (as part of an existing form) with help showing all possible numerical grades and scales.
 *
 * @todo Finish documenting this function
 * @todo Deprecate: this is only used in a few contrib modules
 *
 * @param int $courseid The course ID
 * @param string $name
 * @param string $current
 * @param boolean $includenograde Include those with no grades
 * @param boolean $return If set to true returns rather than echo's
 * @return string|bool Depending on value of $return
 */
function print_grade_menu($courseid, $name, $current, $includenograde=true, $return=false) {
    global $OUTPUT;

    $output = '';
    $strscale = get_string('scale');
    $strscales = get_string('scales');

    $scales = get_scales_menu($courseid);
    foreach ($scales as $i => $scalename) {
        $grades[-$i] = $strscale .': '. $scalename;
    }
    if ($includenograde) {
        $grades[0] = get_string('nograde');
    }
    for ($i=100; $i>=1; $i--) {
        $grades[$i] = $i;
    }
    $output .= html_writer::select($grades, $name, $current, false);

    $linkobject = '<span class="helplink">' . $OUTPUT->pix_icon('help', $strscales) . '</span>';
    $link = new moodle_url('/course/scales.php', array('id' => $courseid, 'list' => 1));
    $action = new popup_action('click', $link, 'ratingscales', array('height' => 400, 'width' => 500));
    $output .= $OUTPUT->action_link($link, $linkobject, $action, array('title' => $strscales));

    if ($return) {
        return $output;
    } else {
        echo $output;
    }
}

/**
 * Print an error to STDOUT and exit with a non-zero code. For commandline scripts.
 *
 * Default errorcode is 1.
 *
 * Very useful for perl-like error-handling:
 * do_somethting() or mdie("Something went wrong");
 *
 * @param string  $msg       Error message
 * @param integer $errorcode Error code to emit
 */
function mdie($msg='', $errorcode=1) {
    trigger_error($msg);
    exit($errorcode);
}

/**
 * Print a message and exit.
 *
 * @param string $message The message to print in the notice
 * @param moodle_url|string $link The link to use for the continue button
 * @param object $course A course object. Unused.
 * @return void This function simply exits
 */
function notice ($message, $link='', $course=null) {
    global $PAGE, $OUTPUT;

    $message = clean_text($message);   // In case nasties are in here.

    if (CLI_SCRIPT) {
        echo("!!$message!!\n");
        exit(1); // No success.
    }

    if (!$PAGE->headerprinted) {
        // Header not yet printed.
        $PAGE->set_title(get_string('notice'));
        echo $OUTPUT->header();
    } else {
        echo $OUTPUT->container_end_all(false);
    }

    echo $OUTPUT->box($message, 'generalbox', 'notice');
    echo $OUTPUT->continue_button($link);

    echo $OUTPUT->footer();
    exit(1); // General error code.
}

/**
 * Redirects the user to another page, after printing a notice.
 *
 * This function calls the OUTPUT redirect method, echo's the output and then dies to ensure nothing else happens.
 *
 * <strong>Good practice:</strong> You should call this method before starting page
 * output by using any of the OUTPUT methods.
 *
 * @param moodle_url|string $url A moodle_url to redirect to. Strings are not to be trusted!
 * @param string $message The message to display to the user
 * @param int $delay The delay before redirecting
 * @param string $messagetype The type of notification to show the message in. See constants on \core\output\notification.
 * @throws moodle_exception
 */
function redirect($url, $message='', $delay=null, $messagetype = \core\output\notification::NOTIFY_INFO) {
    global $OUTPUT, $PAGE, $CFG;

    if (CLI_SCRIPT or AJAX_SCRIPT) {
        // This is wrong - developers should not use redirect in these scripts but it should not be very likely.
        throw new moodle_exception('redirecterrordetected', 'error');
    }

    if ($delay === null) {
        $delay = -1;
    }

    // Prevent debug errors - make sure context is properly initialised.
    if ($PAGE) {
        $PAGE->set_context(null);
        $PAGE->set_pagelayout('redirect');  // No header and footer needed.
        $PAGE->set_title(get_string('pageshouldredirect', 'moodle'));
    }

    if ($url instanceof moodle_url) {
        $url = $url->out(false);
    }

    $debugdisableredirect = false;
    do {
        if (defined('DEBUGGING_PRINTED')) {
            // Some debugging already printed, no need to look more.
            $debugdisableredirect = true;
            break;
        }

        if (core_useragent::is_msword()) {
            // Clicking a URL from MS Word sends a request to the server without cookies. If that
            // causes a redirect Word will open a browser pointing the new URL. If not, the URL that
            // was clicked is opened. Because the request from Word is without cookies, it almost
            // always results in a redirect to the login page, even if the user is logged in in their
            // browser. This is not what we want, so prevent the redirect for requests from Word.
            $debugdisableredirect = true;
            break;
        }

        if (empty($CFG->debugdisplay) or empty($CFG->debug)) {
            // No errors should be displayed.
            break;
        }

        if (!function_exists('error_get_last') or !$lasterror = error_get_last()) {
            break;
        }

        if (!($lasterror['type'] & $CFG->debug)) {
            // Last error not interesting.
            break;
        }

        // Watch out here, @hidden() errors are returned from error_get_last() too.
        if (headers_sent()) {
            // We already started printing something - that means errors likely printed.
            $debugdisableredirect = true;
            break;
        }

        if (ob_get_level() and ob_get_contents()) {
            // There is something waiting to be printed, hopefully it is the errors,
            // but it might be some error hidden by @ too - such as the timezone mess from setup.php.
            $debugdisableredirect = true;
            break;
        }
    } while (false);

    // Technically, HTTP/1.1 requires Location: header to contain the absolute path.
    // (In practice browsers accept relative paths - but still, might as well do it properly.)
    // This code turns relative into absolute.
    if (!preg_match('|^[a-z]+:|i', $url)) {
        // Get host name http://www.wherever.com.
        $hostpart = preg_replace('|^(.*?[^:/])/.*$|', '$1', $CFG->wwwroot);
        if (preg_match('|^/|', $url)) {
            // URLs beginning with / are relative to web server root so we just add them in.
            $url = $hostpart.$url;
        } else {
            // URLs not beginning with / are relative to path of current script, so add that on.
            $url = $hostpart.preg_replace('|\?.*$|', '', me()).'/../'.$url;
        }
        // Replace all ..s.
        while (true) {
            $newurl = preg_replace('|/(?!\.\.)[^/]*/\.\./|', '/', $url);
            if ($newurl == $url) {
                break;
            }
            $url = $newurl;
        }
    }

    // Sanitise url - we can not rely on moodle_url or our URL cleaning
    // because they do not support all valid external URLs.
    $url = preg_replace('/[\x00-\x1F\x7F]/', '', $url);
    $url = str_replace('"', '%22', $url);
    $encodedurl = preg_replace("/\&(?![a-zA-Z0-9#]{1,8};)/", "&amp;", $url);
    $encodedurl = preg_replace('/^.*href="([^"]*)".*$/', "\\1", clean_text('<a href="'.$encodedurl.'" />', FORMAT_HTML));
    $url = str_replace('&amp;', '&', $encodedurl);

    if (!empty($message)) {
        if (!$debugdisableredirect && !headers_sent()) {
            // A message has been provided, and the headers have not yet been sent.
            // Display the message as a notification on the subsequent page.
            \core\notification::add($message, $messagetype);
            $message = null;
            $delay = 0;
        } else {
            if ($delay === -1 || !is_numeric($delay)) {
                $delay = 3;
            }
            $message = clean_text($message);
        }
    } else {
        $message = get_string('pageshouldredirect');
        $delay = 0;
    }

    // Make sure the session is closed properly, this prevents problems in IIS
    // and also some potential PHP shutdown issues.
    \core\session\manager::write_close();

    if ($delay == 0 && !$debugdisableredirect && !headers_sent()) {
        // 302 might not work for POST requests, 303 is ignored by obsolete clients.
        @header($_SERVER['SERVER_PROTOCOL'] . ' 303 See Other');
        @header('Location: '.$url);
        echo bootstrap_renderer::plain_redirect_message($encodedurl);
        exit;
    }

    // Include a redirect message, even with a HTTP redirect, because that is recommended practice.
    if ($PAGE) {
        $CFG->docroot = false; // To prevent the link to moodle docs from being displayed on redirect page.
        echo $OUTPUT->redirect_message($encodedurl, $message, $delay, $debugdisableredirect, $messagetype);
        exit;
    } else {
        echo bootstrap_renderer::early_redirect_message($encodedurl, $message, $delay);
        exit;
    }
}

/**
 * Given an email address, this function will return an obfuscated version of it.
 *
 * @param string $email The email address to obfuscate
 * @return string The obfuscated email address
 */
function obfuscate_email($email) {
    $i = 0;
    $length = strlen($email);
    $obfuscated = '';
    while ($i < $length) {
        if (rand(0, 2) && $email{$i}!='@') { // MDL-20619 some browsers have problems unobfuscating @.
            $obfuscated.='%'.dechex(ord($email{$i}));
        } else {
            $obfuscated.=$email{$i};
        }
        $i++;
    }
    return $obfuscated;
}

/**
 * This function takes some text and replaces about half of the characters
 * with HTML entity equivalents.   Return string is obviously longer.
 *
 * @param string $plaintext The text to be obfuscated
 * @return string The obfuscated text
 */
function obfuscate_text($plaintext) {
    $i=0;
    $length = core_text::strlen($plaintext);
    $obfuscated='';
    $prevobfuscated = false;
    while ($i < $length) {
        $char = core_text::substr($plaintext, $i, 1);
        $ord = core_text::utf8ord($char);
        $numerical = ($ord >= ord('0')) && ($ord <= ord('9'));
        if ($prevobfuscated and $numerical ) {
            $obfuscated.='&#'.$ord.';';
        } else if (rand(0, 2)) {
            $obfuscated.='&#'.$ord.';';
            $prevobfuscated = true;
        } else {
            $obfuscated.=$char;
            $prevobfuscated = false;
        }
        $i++;
    }
    return $obfuscated;
}

/**
 * This function uses the {@link obfuscate_email()} and {@link obfuscate_text()}
 * to generate a fully obfuscated email link, ready to use.
 *
 * @param string $email The email address to display
 * @param string $label The text to displayed as hyperlink to $email
 * @param boolean $dimmed If true then use css class 'dimmed' for hyperlink
 * @param string $subject The subject of the email in the mailto link
 * @param string $body The content of the email in the mailto link
 * @return string The obfuscated mailto link
 */
function obfuscate_mailto($email, $label='', $dimmed=false, $subject = '', $body = '') {

    if (empty($label)) {
        $label = $email;
    }

    $label = obfuscate_text($label);
    $email = obfuscate_email($email);
    $mailto = obfuscate_text('mailto');
    $url = new moodle_url("mailto:$email");
    $attrs = array();

    if (!empty($subject)) {
        $url->param('subject', format_string($subject));
    }
    if (!empty($body)) {
        $url->param('body', format_string($body));
    }

    // Use the obfuscated mailto.
    $url = preg_replace('/^mailto/', $mailto, $url->out());

    if ($dimmed) {
        $attrs['title'] = get_string('emaildisable');
        $attrs['class'] = 'dimmed';
    }

    return html_writer::link($url, $label, $attrs);
}

/**
 * This function is used to rebuild the <nolink> tag because some formats (PLAIN and WIKI)
 * will transform it to html entities
 *
 * @param string $text Text to search for nolink tag in
 * @return string
 */
function rebuildnolinktag($text) {

    $text = preg_replace('/&lt;(\/*nolink)&gt;/i', '<$1>', $text);

    return $text;
}

/**
 * Prints a maintenance message from $CFG->maintenance_message or default if empty.
 */
function print_maintenance_message() {
    global $CFG, $SITE, $PAGE, $OUTPUT;

    header($_SERVER['SERVER_PROTOCOL'] . ' 503 Moodle under maintenance');
    header('Status: 503 Moodle under maintenance');
    header('Retry-After: 300');

    $PAGE->set_pagetype('maintenance-message');
    $PAGE->set_pagelayout('maintenance');
    $PAGE->set_title(strip_tags($SITE->fullname));
    $PAGE->set_heading($SITE->fullname);
    echo $OUTPUT->header();
    echo $OUTPUT->heading(get_string('sitemaintenance', 'admin'));
    if (isset($CFG->maintenance_message) and !html_is_blank($CFG->maintenance_message)) {
        echo $OUTPUT->box_start('maintenance_message generalbox boxwidthwide boxaligncenter');
        echo $CFG->maintenance_message;
        echo $OUTPUT->box_end();
    }
    echo $OUTPUT->footer();
    die;
}

/**
 * Returns a string containing a nested list, suitable for formatting into tabs with CSS.
 *
 * It is not recommended to use this function in Moodle 2.5 but it is left for backward
 * compartibility.
 *
 * Example how to print a single line tabs:
 * $rows = array(
 *    new tabobject(...),
 *    new tabobject(...)
 * );
 * echo $OUTPUT->tabtree($rows, $selectedid);
 *
 * Multiple row tabs may not look good on some devices but if you want to use them
 * you can specify ->subtree for the active tabobject.
 *
 * @param array $tabrows An array of rows where each row is an array of tab objects
 * @param string $selected  The id of the selected tab (whatever row it's on)
 * @param array  $inactive  An array of ids of inactive tabs that are not selectable.
 * @param array  $activated An array of ids of other tabs that are currently activated
 * @param bool $return If true output is returned rather then echo'd
 * @return string HTML output if $return was set to true.
 */