message.php

См. документацию.

<?php
 
namespace Bitrix\Mail;
 
use Bitrix\Mail\Integration\UI\EntitySelector\AddressBookProvider;
use Bitrix\Main;
use Bitrix\Main\Context;
use Bitrix\Main\Localization\Loc;
use Bitrix\UI\EntitySelector\Dialog;
use Bitrix\UI\EntitySelector\ItemCollection;
use CTimeZone;
 
Main\Localization\Loc::loadMessages(__FILE__);
 
class Message
{
 
    //const QUOTE_START_MARKER = '-- Bitrix24 Mail begin ---';
    //const QUOTE_END_MARKER   = '-- Bitrix24 Mail end ---';
 
    const QUOTE_START_MARKER_HTML = '<div id="srvb24mqsm" style="font-family: \'srvb24mqsm\', serif;">&nbsp;</div>';
    const QUOTE_END_MARKER_HTML   = '<div id="qemb24msrv" style="font-family: \'qemb24msrv\', serif;">&nbsp;</div>';
 
    const QUOTE_HTML_REGEX = '/<div\s[^>]+srvb24mqsm[^>]+>.*?<\/div>(.*)<div\s[^>]+qemb24msrv[^>]+>.*?<\/div>/is';
 
    const QUOTE_PLACEHOLDER = '__QUOTE_PLACEHOLDER__';
 
    protected $type;
    protected $headers, $subject, $from, $to;
    protected $html, $text, $attachments;
    protected $secret;
 
    public function __construct(array &$message, $type)
    {
        $this->type = $type;
 
        $properties = array(
            'headers', 'subject', 'from', 'to',
            'text', 'html', 'attachments',
            'secret'
        );
 
        foreach ($properties as $property)
        {
            if (isset($message[$property]))
                $this->$property = $message[$property];
        }
    }
 
    public static function stripQuotes($text): string
    {
        return preg_replace('/^("(.*)"|\'(.*)\')$/', '$2$3', $text);
    }
 
    private static function convertContactListToString($list): string
    {
        $string = '';
 
        foreach ($list as $contact)
        {
            $name = static::stripQuotes($contact['name']);
            $email = $contact['email'];
 
            if ($name === $email)
            {
                $name = '';
            }
 
            $string .= Loc::getMessage('MAIL_QUOTE_MESSAGE_HEADER_CONTACT', ['#NAME#' => $name,'#EMAIL#' => $email]);
 
            if (next($list))
            {
                $string .= ', ';
            }
        }
 
        return $string;
    }
 
    final public static function getSelectedRecipientsForDialog($recipients = [], $updateData = false): ItemCollection
    {
        $options = [
            'entities' => [
                [
                    'id' => AddressBookProvider::PROVIDER_ENTITY_ID,
                ],
            ],
        ];
 
        $dialog = new Dialog($options);
        $provider = $dialog->getEntity(AddressBookProvider::PROVIDER_ENTITY_ID)->getProvider();
        $provider->setPreinstalledItems($recipients, $updateData);
        $items = $provider->getItems([]);
        $itemCollection = new ItemCollection();
 
        foreach ($items as $item)
        {
            $itemCollection->add($item);
        }
 
        return $itemCollection;
    }
 
    final public static function wrapTheMessageWithAQuote($body, $subject, $timeString, $from = [], $to = [], $cc = [], bool $sanitized = false): string
    {
        $fieldDateInTimeStamp = makeTimestamp($timeString);
        $titleDateFormat = Context::getCurrent()->getCulture()->getFullDateFormat()."&#013;H:i:s";
        $formattedDate = FormatDate($titleDateFormat, $fieldDateInTimeStamp, (time() + CTimeZone::getOffset()));
 
        $wrap = '';
 
        $fromList = static::convertContactListToString($from);
        $toList = static::convertContactListToString($to);
        $ccList = static::convertContactListToString($cc);
        if (!$sanitized)
        {
            $body = Helper\Message::sanitizeHtml($body);
        }
 
        if (empty($ccList))
        {
            $wrap .= Loc::getMessage('MAIL_QUOTE_MESSAGE_HEADER_WITHOUT_CC', [
                '#DATE#' => $formattedDate,
                '#SUBJECT#' => $subject,
                '#BODY#' => $body,
                '#FROM_LIST#' => $fromList,
                '#TO_LIST#' => $toList,
                '[blockquote]' => '<blockquote style="margin: 0 0 0 5px; padding: 5px 5px 5px 8px; border-left: 4px solid #e2e3e5; ">',
                '[/blockquote]' => '</blockquote>'
            ]);
        }
        else
        {
            $wrap .= Loc::getMessage('MAIL_QUOTE_MESSAGE_HEADER', [
                '#DATE#' => $formattedDate,
                '#SUBJECT#' => $subject,
                '#BODY#' => $body,
                '#FROM_LIST#' => $fromList,
                '#TO_LIST#' => $toList,
                '#CC_LIST#' => $ccList,
                '[blockquote]' => '<blockquote style="margin: 0 0 0 5px; padding: 5px 5px 5px 8px; border-left: 4px solid #e2e3e5; ">',
                '[/blockquote]' => '</blockquote>'
            ]);
        }
 
        return $wrap;
    }

    final public static function getQuoteStartMarker($html = false)
    {
        return $html ? static::QUOTE_START_MARKER_HTML : static::QUOTE_START_MARKER;
    }

    final public static function getQuoteEndMarker($html = false)
    {
        return $html ? static::QUOTE_END_MARKER_HTML : static::QUOTE_END_MARKER;
    }

    public function attachmentsCount()
    {
        return is_array($this->attachments) ? count($this->attachments) : 0;
    }

    protected function parse()
    {
        if (isset($this->html))
        {
            $html = $this->html;
 
            $html = str_replace(array("\r", "\n"), '', $html);
            $html = preg_replace('/<br\s*\/?>/is', "\n", $html);
 
            $html = str_ireplace('</div>', "</div>\n", $html);
            $html = str_ireplace('</p>', "</p>\n", $html);
            $html = preg_replace('/<\/h([1-6])>/i', "</h\\1>\n", $html);
            $html = str_ireplace('</table>', "</table>\n", $html);
            $html = str_ireplace('</tr>', "</tr>\n", $html);
            $html = str_ireplace('</pre>', "</pre>\n", $html);
 
            $html = preg_replace('/(\n\s*)?<div/i', "\n<div", $html);
            $html = preg_replace('/(\n\s*)?<p(?=\s|>)/i', "\n<p", $html);
            $html = preg_replace('/(\n\s*)?<h([1-6])/i', "\n<h\\2", $html);
            $html = preg_replace('/(\n\s*)?<table/i', "\n<table", $html);
            $html = preg_replace('/(\n\s*)?<tr/i', "\n<tr", $html);
            $html = preg_replace('/(\n\s*)?<pre/i', "\n<pre", $html);
 
            $html = preg_replace('/(\n\s*)?<hr[^>]*>(\s*\n)?/i', "\n<hr>\n", $html);
 
            if ($this->type == 'reply' and $parts = $this->splitHtml($html))
            {
                list($before, $quote, $after) = $parts;
                $html = sprintf('%s%s%s', $before, static::QUOTE_PLACEHOLDER, $after);
            }
 
            if ($this->attachmentsCount())
            {
                foreach ($this->attachments as $item)
                {
                    $html = preg_replace(
                        sprintf('/<img[^>]+src\s*=\s*(\'|\")?\s*(cid:%s)\s*\1[^>]*>/is', preg_quote($item['contentId'], '/')),
                        sprintf('[ATTACHMENT=%s]', $item['uniqueId']),
                        $html
                    );
                }
            }
 
            // TODO: Sanitizer
            $html = preg_replace('/<style[^>]*>.*?<\/style>/is', '', $html);
            $html = preg_replace('/<script[^>]*>.*?<\/script>/is', '', $html);
            $html = preg_replace('/<title[^>]*>.*?<\/title>/is', '', $html);
            $html = preg_replace('/<caption[^>]*>.*?<\/caption>/is', '', $html);
 
            // TODO: Sanitizer
            $html = preg_replace('/<a\s[^>]*href\s*=\s*([^\'\"\s>]+)\s*[^>]*>/is', '<a href="\1">', $html);
 
            // TODO: TextParser
            $html = preg_replace('/<strong[^>]*>(.*?)<\/strong>/is', '<b>\1</b>', $html);
            $html = preg_replace('/<em[^>]*>(.*?)<\/em>/is', '<i>\1</i>', $html);
            $html = preg_replace('/<blockquote[^>]*>(.*?)<\/blockquote>/is', '<quote>\1</quote>', $html);
            $html = preg_replace('/<hr[^>]*>/is', '________________________________________', $html);
            $html = preg_replace('/<del[^>]*>(.*?)<\/del>/is', '<s>\1</s>', $html);
            $html = preg_replace('/<ins[^>]*>(.*?)<\/ins>/is', '<u>\1</u>', $html);
            $html = preg_replace('/<h([1-6])[^>]*>(.*?)<\/h\1>/is', '<b>\2</b>', $html);
            $html = preg_replace('/<dl[^>]*>(.*?)<\/dl>/is', '<ul>\1</ul>', $html);
            $html = preg_replace('/<dt[^>]*>/is', '<li>', $html);
            $html = preg_replace('/<dd[^>]*>/is', ' - ', $html);
            $html = preg_replace('/<(sub|sup)[^>]*>(.*?)<\/\1>/is', '(\2)', $html);
 
            // TODO: TextParser
            $html = preg_replace('/<th[^>]*>(.*?)<\/th>/is', '<td>\1</td>', $html);
 
            $sanitizer = new \CBXSanitizer();
            //$sanitizer->setLevel(\CBXSanitizer::SECURE_LEVEL_MIDDLE);
            $sanitizer->addTags(array(
                'a'     => array('href'),
                'b'     => array(),
                'u'     => array(),
                's'     => array(),
                'i'     => array(),
                'img'   => array('src'),
                'font'  => array('color', 'size', 'face'),
                'ul'    => array(),
                'ol'    => array(),
                'li'    => array(),
                'table' => array(),
                'tr'    => array(),
                'td'    => array(),
                'th'    => array(),
                'quote' => array(),
                'br'    => array(),
                //'big'   => array(),
                //'small' => array(),
            ));
            $sanitizer->applyDoubleEncode(false);
            $html = $sanitizer->sanitizeHtml($html);
 
            $parser = new \CTextParser();
            $text = $parser->convertHtmlToBB($html);
 
            $text = html_entity_decode($text, ENT_QUOTES | ENT_HTML401, LANG_CHARSET);
 
            // TODO: TextParser
            $text = preg_replace('/<\/?([abuis]|img|font|ul|ol|li|table|tr|td|th|quote|br)(?=\s|>)[^>]*>/i', '', $text);
 
            $text = preg_replace('/[\t\x20]+/', "\x20", $text);
        }
        else
        {
            $text = $this->text;
 
            $text = str_replace("\r\n", "\n", $text);
            $text = str_replace("\r", "\n", $text);
 
            if ($this->type == 'reply' and $parts = $this->splitText($text))
            {
                list($before, $quote, $after) = $parts;
                $text = sprintf('%s%s%s', $before, static::QUOTE_PLACEHOLDER, $after);
            }
 
            if ($this->attachmentsCount())
            {
                foreach ($this->attachments as $item)
                {
                    $text = str_replace(
                        sprintf('[cid:%s]', $item['contentId']),
                        sprintf('[ATTACHMENT=%s]', $item['uniqueId']),
                        $text
                    );
                }
            }
        }
 
        if ($this->type == 'reply' && mb_strpos($text, static::QUOTE_PLACEHOLDER))
        {
            $text = $this->removeReplyHead($text);
            $text = preg_replace(sprintf('/\s*%s\s*/', preg_quote(static::QUOTE_PLACEHOLDER, '/')), "\n\n", $text);
        }
 
        if ($this->type == 'forward')
            $text = $this->removeForwardHead($text);
 
        // TODO: TextParser
        $text = preg_replace('/\[tr\]\s*\[\/tr\]/is', '', $text);
        $text = preg_replace('/\[table\]\s*\[\/table\]/is', '', $text);
 
        $text = trim($text);
        $text = preg_replace('/(\s*\n){2,}/', "\n\n", $text);
 
        if (empty($text) && $this->attachmentsCount() == 1)
            $text = sprintf('[ATTACHMENT=%s]', $item['uniqueId']);
 
        if (!empty($this->secret))
            $text = str_replace($this->secret, 'xxxxxxxx', $text);
 
        return $text;
    }
 
    public static function parseMessage(array &$message)
    {
        $message = new static($message, null);
 
        return $message->parse();
    }

    public static function parseReply(array &$message)
    {
        $reply = new static($message, 'reply');
 
        return $reply->parse();
    }

    public static function parseForward(array &$message)
    {
        $forward = new static($message, 'forward');
 
        return $forward->parse();
    }

    protected function splitHtml(&$html)
    {
        $parts = preg_split('/(<blockquote.+?<\/blockquote>)/is', $html, null, PREG_SPLIT_DELIM_CAPTURE);
 
        if (count($parts) > 3)
        {
            $parts = array_merge(
                array(join(array_slice($parts, 0, -2))),
                array_slice($parts, -2)
            );
        }
        else
        {
            if (count($parts) == 3)
                $parts = preg_split('/(<blockquote.+<\/blockquote>)/is', $html, null, PREG_SPLIT_DELIM_CAPTURE);
        }
 
        if (count($parts) < 3)
            $parts = preg_split(static::QUOTE_HTML_REGEX, $html, null, PREG_SPLIT_DELIM_CAPTURE);
 
        if (count($parts) == 3)
            return $parts;
 
        return false;
    }

    protected function splitText(&$text)
    {
        $parts = preg_split('/((?:^>.*$\n?){2,})/m', $text, null, PREG_SPLIT_DELIM_CAPTURE);
 
        if (count($parts) < 3)
            $parts = preg_split('/((?:^\|.*$\n?){2,})/m', $text, null, PREG_SPLIT_DELIM_CAPTURE);
 
        if (count($parts) < 3)
        {
            $outlookRegex = '/(
                (?:^_{20,}\n(?:[\t\x20]*\n)?)?
                (?:^(?:from|to|subject|sent|date):\x20[^\n]+$\n?){2,8}.*
            )/ismx';
            $parts = preg_split($outlookRegex, $text, null, PREG_SPLIT_DELIM_CAPTURE);
        }
 
        if (count($parts) == 3)
            return $parts;
 
        return false;
    }

    protected function scoreFullHead(&$head)
    {
        $score = 0;
 
        if (preg_match_all('/^([^\:\n]{1,20}):[\t\x20]+(.+)$/mu', $head, $matches, PREG_SET_ORDER))
        {
            $subject = array(
                'value'  => $this->subject,
                'strlen' => mb_strlen($this->subject),
                'sgnlen' => mb_strlen(trim($this->subject))
            );
 
            $isHeader = function($key, $value) use (&$subject)
            {
                if (mb_strlen(trim($value)) >= 10 && $subject['sgnlen'] >= 10)
                {
                    $dist = $subject['strlen'] - mb_strlen($value);
 
                    if (abs($dist) < 10)
                    {
                        if ($dist >= 0 && mb_strpos($subject['value'], $value) !== false)
                        {
                            return true;
                        }
 
                        if (max($subject['strlen'], mb_strlen($value)) < 256 && levenshtein($subject['value'], $value) < 10)
                        {
                            return true;
                        }
                    }
                }
 
                $date = preg_replace('/(?<=[\s\d])UT$/i', '+0000', trim($value));
                if (preg_match('/\d{1,2}:\d{2}(:\d{2})?\x20?(am|pm)?/i', $date) && strtotime($date) !== false)
                {
                    return true;
                }
 
                if (preg_match('/([a-z\d_](\.?[a-z\d_-]+)*)?[a-z\d_]@(([a-z\d][a-z\d-]*)?[a-z\d]\.?)+/i', $value))
                {
                    return true;
                }
 
                return false;
            };
 
            foreach ($matches as $item)
            {
                $score += (int) $isHeader($item[1], $item[2]);
            }
        }
 
        return $score;
    }

    protected function scoreShortHead(&$head)
    {
        $score = 0;
 
        $regex = '/(?:^|\n)
            (?<date>.{5,50}\d),?\x20
            [^\d\n]{0,20}(?<time>\d{1,2}\:\d{2}(?:\:\d{2})?\x20?(?:am|pm)?),?\x20
            (?<from>.+):\s*$
        /ixu';
        if (preg_match($regex, $head, $matches))
        {
            $matches['date'] = trim($matches['date']);
            if (strtotime($matches['date']) !== false)
            {
                $score++;
            }
            else if (preg_match('/^[^\x20]+\x20+((?:[^\x20]+\x20+)?(.+))$/', $matches['date'], $date))
            {
                if (strtotime($date[1]) !== false || strtotime($date[2]) !== false)
                    $score++;
            }
 
            if (preg_match('/([a-z\d_](\.?[a-z\d_-]+)*)?[a-z\d_]@(([a-z\d][a-z\d-]*)?[a-z\d]\.?)+/i', $matches['from']))
                $score++;
        }
 
        return $score;
    }

    protected function removeReplyHead(&$text)
    {
        list($before, $after) = explode(static::QUOTE_PLACEHOLDER, $text, 2);
 
        if (!trim($before))
            return $text;
 
        $data = static::reduceTags($before);

        $fullHeadRegex = '/(?:^|\n\n)
            (?<hr>_{20,}\n(?:[\t\x20]*\n)?)?
            (?<head>(?:[^\:\n]{1,20}:[\t\x20]+.+(?:\n|$)){2,6})\s*$
        /xu';
        if (preg_match($fullHeadRegex, $data, $matches))
        {
            $score  = (int) !empty($matches['hr']);
            $score += $this->scoreFullHead($matches['head']);
 
            if ($score > 1)
            {
                $pattern = preg_replace(array('/.+/', '/\n/'), array('.+', '\n'), $matches[0]);
                $before  = preg_replace_callback(
                    sprintf('/%s$/', $pattern),
                    function($matches)
                    {
                        return Message::reduceHead($matches[0]);
                    },
                    $before
                );
 
                return sprintf('%s%s%s', $before, static::QUOTE_PLACEHOLDER, $after);
            }
        }

        $shortHeadRegex = '/(?:^|\n)
            (?<date>.{5,50}\d),?\x20
            [^\d\n]{0,20}(?<time>\d{1,2}\:\d{2}(?:\:\d{2})?\x20?(?:am|pm)?),?\x20
            (?<from>.+):\s*$
        /ixu';
        if (preg_match($shortHeadRegex, $data, $matches))
        {
            $score = 0;
            $score += $this->scoreShortHead($matches[0]);
 
            if ($score > 0)
            {
                $pattern = preg_replace(array('/.+/', '/\n/'), array('.+', '\n'), $matches[0]);
                $before  = preg_replace_callback(
                    sprintf('/%s$/', $pattern),
                    function($matches)
                    {
                        return Message::reduceHead($matches[0]);
                    },
                    $before
                );
 
                return sprintf('%s%s%s', $before, static::QUOTE_PLACEHOLDER, $after);
            }
        }
 
        return $text;
    }

    protected function removeForwardHead(&$text)
    {
        if (!trim($text))
            return $text;
 
        $data = static::reduceTags($text);
 
        $shortHeadRegex = '/(?:^|\n)\s*
            -{3,}.{4,40}?-{3,}[\t\x20]*\n
            (?<head>(?:[\t\x20]*\n)?
            (?<date>.{5,50}\d),?\x20
            [^\d\n]{0,20}(?<time>\d{1,2}\:\d{2}(?:\:\d{2})?\x20?(?:am|pm)?),?\x20
            (?<from>.+):(?:\s*\n)?)
        /ixu';
 
        $hasMarker = preg_match($shortHeadRegex, $data);
        $fullHeadRegex = '/(?:^|\n\n)\s*
            (?<marker>-{3,}.{4,40}?-{3,}[\t\x20]*\n)'.($hasMarker ? '' : '?').'
            (?<head>(?:[\t\x20]*\n)?
            (?<lines>(?:[^\:\n]{1,20}:[\t\x20]+.+(?:\n|$)){2,6}))
            \s*(?:\n|$)
        /xu';
 
        if (preg_match($fullHeadRegex, $data, $matches, PREG_OFFSET_CAPTURE))
        {
            $score  = (int) !empty($matches['marker'][0]);
            $score += $this->scoreFullHead($matches['lines'][0]);
 
            if ($score > 1)
            {
                // @TODO: Main\Text\BinaryString::getSubstring()
                $pattern = preg_replace(
                    array('/.+/', '/\n/'), array('.+', '\n'),
                    array(substr($data, 0, $matches['head'][1]), $matches['head'][0])
                );
 
                return preg_replace_callback(
                    sprintf('/^(%s)(%s)/', $pattern[0], $pattern[1]),
                    function($matches)
                    {
                        return sprintf("%s\n\n%s", $matches[1], Message::reduceHead($matches[2]));
                    },
                    $text
                );
            }
        }
 
        if (preg_match($shortHeadRegex, $data, $matches, PREG_OFFSET_CAPTURE))
        {
            $score  = 0;
            $score += $this->scoreShortHead($matches['head'][0]);
 
            if ($score > 0)
            {
                // @TODO: Main\Text\BinaryString::getSubstring()
                $pattern = preg_replace(
                    array('/.+/', '/\n/'), array('.+', '\n'),
                    array(substr($data, 0, $matches['head'][1]), $matches['head'][0])
                );
 
                return preg_replace_callback(
                    sprintf('/^(%s)(%s)/', $pattern[0], $pattern[1]),
                    function($matches)
                    {
                        return sprintf("%s\n\n%s", $matches[1], Message::reduceHead($matches[2]));
                    },
                    $text
                );
            }
        }
 
        return $text;
    }

    protected static function reduceTags(&$text)
    {
        $data = $text;
 
        $data = preg_replace('/^(\[\/?(\*|[busi]|img|table|tr|td|th|quote|(url|size|color|font|list)(=.+?)?)\])+$/im', "\t", $data);
        $data = preg_replace('/\[\/?(\*|[busi]|img|table|tr|td|th|quote|(url|size|color|font|list)(=.+?)?)\]/i', '', $data);
 
        return $data;
    }

    public static function reduceHead(&$text)
    {
        preg_match_all('/\[\/?([busi]|img|table|tr|td|th|quote|(url|size|color|font|list)(=.+?)?)\]/is', $text, $tags);
 
        $result = join($tags[0]);
        unset($tags);
 
        do
        {
            $result = preg_replace('/\[([busi]|img|table|tr|td|th|quote|url|size|color|font|list)(=.+?)?\]\[\/\1\]/is', '', $result, -1, $n2);
        }
        while ($n2 > 0);
 
        return $result;
    }
 
}