. // }}} namespace Component\Link; use App\Core\DB\DB; use App\Core\Event; use App\Core\Modules\Component; use App\Entity\Actor; use App\Entity\Note; use App\Util\Common; use App\Util\HTML; use Component\Link\Entity\NoteToLink; use InvalidArgumentException; class Link extends Component { /** * Extract URLs from $content and create the appropriate Link and NoteToLink entities */ public function onProcessNoteContent(Note $note, string $content): bool { if (Common::config('attachments', 'process_links')) { $matched_urls = []; // TODO: This solution to ignore mentions when content is in html is far from ideal preg_match_all($this->getURLRegex(), preg_replace('##', '', $content), $matched_urls); $matched_urls = array_unique($matched_urls[1]); foreach ($matched_urls as $match) { try { $link_id = Entity\Link::getOrCreate($match)->getId(); DB::persist(NoteToLink::create(['link_id' => $link_id, 'note_id' => $note->getId()])); } catch (InvalidArgumentException) { continue; } } } return Event::next; } public function onRenderPlainTextNoteContent(string &$text): bool { $text = $this->replaceURLs($text); return Event::next; } public function getURLRegex(): string { $geouri_labeltext_regex = '\pN\pL\-'; $geouri_mark_regex = '\-\_\.\!\~\*\\\'\(\)'; // the \\\' is really pretty $geouri_unreserved_regex = '\pN\pL' . $geouri_mark_regex; $geouri_punreserved_regex = '\[\]\:\&\+\$'; $geouri_pctencoded_regex = '(?:\%[0-9a-fA-F][0-9a-fA-F])'; $geouri_paramchar_regex = $geouri_unreserved_regex . $geouri_punreserved_regex; //FIXME: add $geouri_pctencoded_regex here so it works return '#' . '(?:^|[\s\<\>\(\)\[\]\{\}\\\'\\\";]+)(?![\@\!\#])' . '(' . '(?:' . '(?:' //Known protocols . '(?:' . '(?:(?:' . implode('|', $this->URLSchemes(self::URL_SCHEME_COLON_DOUBLE_SLASH)) . ')://)' . '|' . '(?:(?:' . implode('|', $this->URLSchemes(self::URL_SCHEME_SINGLE_COLON)) . '):)' . ')' . '(?:[\pN\pL\-\_\+\%\~]+(?::[\pN\pL\-\_\+\%\~]+)?\@)?' //user:pass@ . '(?:' . '(?:' . '\[[\pN\pL\-\_\:\.]+(?URLSchemes(self::URL_SCHEME_COLON_COORDINATES)) . '):' // There's an order that must be followed here too, if ;crs= is used, it must precede ;u= // Also 'crsp' (;crs=$crsp) must match $geouri_labeltext_regex // Also 'uval' (;u=$uval) must be a pnum: \-?[0-9]+ . '(?:' . '(?:[0-9]+(?:\.[0-9]+)?(?:\,[0-9]+(?:\.[0-9]+)?){1,2})' // 1(.23)?(,4(.56)){1,2} . '(?:\;(?:[' . $geouri_labeltext_regex . ']+)(?:\=[' . $geouri_paramchar_regex . ']+)*)*' . ')' . ')' // URLs without domain name, like magnet:?xt=... . '|(?:(?:' . implode('|', $this->URLSchemes(self::URL_SCHEME_NO_DOMAIN)) . '):(?=\?))' // zero-length lookahead requires ? after : . (Common::config('linkify', 'ipv4') // Convert IPv4 addresses to hyperlinks ? '|(?:(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.){3}(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)' : '') . (Common::config('linkify', 'ipv6') // Convert IPv6 addresses to hyperlinks ? '|(?:' //IPv6 . '\[?(?:(?:(?:[0-9A-Fa-f]{1,4}:){7}(?:(?:[0-9A-Fa-f]{1,4})|:))|(?:(?:[0-9A-Fa-f]{1,4}:){6}(?::|(?:(?:25[0-5]|2[0-4]\d|[01]?\d{1,2})(?:\.(?:25[0-5]|2[0-4]\d|[01]?\d{1,2})){3})|(?::[0-9A-Fa-f]{1,4})))|(?:(?:[0-9A-Fa-f]{1,4}:){5}(?:(?::(?:(?:25[0-5]|2[0-4]\d|[01]?\d{1,2})(?:\.(?:25[0-5]|2[0-4]\d|[01]?\d{1,2})){3})?)|(?:(?::[0-9A-Fa-f]{1,4}){1,2})))|(?:(?:[0-9A-Fa-f]{1,4}:){4}(?::[0-9A-Fa-f]{1,4}){0,1}(?:(?::(?:(?:25[0-5]|2[0-4]\d|[01]?\d{1,2})(?:\.(?:25[0-5]|2[0-4]\d|[01]?\d{1,2})){3})?)|(?:(?::[0-9A-Fa-f]{1,4}){1,2})))|(?:(?:[0-9A-Fa-f]{1,4}:){3}(?::[0-9A-Fa-f]{1,4}){0,2}(?:(?::(?:(?:25[0-5]|2[0-4]\d|[01]?\d{1,2})(?:\.(?:25[0-5]|2[0-4]\d|[01]?\d{1,2})){3})?)|(?:(?::[0-9A-Fa-f]{1,4}){1,2})))|(?:(?:[0-9A-Fa-f]{1,4}:){2}(?::[0-9A-Fa-f]{1,4}){0,3}(?:(?::(?:(?:25[0-5]|2[0-4]\d|[01]?\d{1,2})(?:\.(?:25[0-5]|2[0-4]\d|[01]?\d{1,2})){3})?)|(?:(?::[0-9A-Fa-f]{1,4}){1,2})))|(?:(?:[0-9A-Fa-f]{1,4}:)(?::[0-9A-Fa-f]{1,4}){0,4}(?:(?::(?:(?:25[0-5]|2[0-4]\d|[01]?\d{1,2})(?:\.(?:25[0-5]|2[0-4]\d|[01]?\d{1,2})){3})?)|(?:(?::[0-9A-Fa-f]{1,4}){1,2})))|(?::(?::[0-9A-Fa-f]{1,4}){0,5}(?:(?::(?:(?:25[0-5]|2[0-4]\d|[01]?\d{1,2})(?:\.(?:25[0-5]|2[0-4]\d|[01]?\d{1,2})){3})?)|(?:(?::[0-9A-Fa-f]{1,4}){1,2})))|(?:(?:(?:25[0-5]|2[0-4]\d|[01]?\d{1,2})(?:\.(?:25[0-5]|2[0-4]\d|[01]?\d{1,2})){3})))\]?(? self::URL_SCHEME_COLON_DOUBLE_SLASH, 'https' => self::URL_SCHEME_COLON_DOUBLE_SLASH, 'ftp' => self::URL_SCHEME_COLON_DOUBLE_SLASH, 'ftps' => self::URL_SCHEME_COLON_DOUBLE_SLASH, 'mms' => self::URL_SCHEME_COLON_DOUBLE_SLASH, 'rtsp' => self::URL_SCHEME_COLON_DOUBLE_SLASH, 'gopher' => self::URL_SCHEME_COLON_DOUBLE_SLASH, 'news' => self::URL_SCHEME_COLON_DOUBLE_SLASH, 'nntp' => self::URL_SCHEME_COLON_DOUBLE_SLASH, 'telnet' => self::URL_SCHEME_COLON_DOUBLE_SLASH, 'wais' => self::URL_SCHEME_COLON_DOUBLE_SLASH, 'file' => self::URL_SCHEME_COLON_DOUBLE_SLASH, 'prospero' => self::URL_SCHEME_COLON_DOUBLE_SLASH, 'webcal' => self::URL_SCHEME_COLON_DOUBLE_SLASH, 'irc' => self::URL_SCHEME_COLON_DOUBLE_SLASH, 'ircs' => self::URL_SCHEME_COLON_DOUBLE_SLASH, 'aim' => self::URL_SCHEME_SINGLE_COLON, 'bitcoin' => self::URL_SCHEME_SINGLE_COLON, 'fax' => self::URL_SCHEME_SINGLE_COLON, 'jabber' => self::URL_SCHEME_SINGLE_COLON, 'mailto' => self::URL_SCHEME_SINGLE_COLON, 'tel' => self::URL_SCHEME_SINGLE_COLON, 'xmpp' => self::URL_SCHEME_SINGLE_COLON, 'magnet' => self::URL_SCHEME_NO_DOMAIN, 'geo' => self::URL_SCHEME_COLON_COORDINATES, ]; return array_keys(array_filter($schemes, fn ($scheme) => \is_null($filter) || ($scheme & $filter))); } /** * Find links in the given text and pass them to the given callback function. */ public function replaceURLs(string $text): string { $regex = $this->getURLRegex(); return preg_replace_callback($regex, fn ($matches) => $this->callbackHelper($matches, [$this, 'linkify']), $text); } /** * Intermediate callback for `replaceURLs()`, which helps resolve some * ambiguous link forms before passing on to the final callback. * * @param callable(string $text): string $callback: return replacement text */ private function callbackHelper(array $matches, callable $callback): string { $url = $matches[1]; $left = mb_strpos($matches[0], $url); $right = $left + mb_strlen($url); $groupSymbolSets = [ [ 'left' => '(', 'right' => ')', ], [ 'left' => '[', 'right' => ']', ], [ 'left' => '{', 'right' => '}', ], [ 'left' => '<', 'right' => '>', ], ]; $cannotEndWith = ['.', '?', ',', '#']; do { $original_url = $url; foreach ($groupSymbolSets as $groupSymbolSet) { if (mb_substr($url, -1) == $groupSymbolSet['right']) { $group_left_count = mb_substr_count($url, $groupSymbolSet['left']); $group_right_count = mb_substr_count($url, $groupSymbolSet['right']); if ($group_left_count < $group_right_count) { --$right; $url = mb_substr($url, 0, -1); } } } if (\in_array(mb_substr($url, -1), $cannotEndWith)) { --$right; $url = mb_substr($url, 0, -1); } } while ($original_url != $url); $result = $callback($url); return mb_substr($matches[0], 0, $left) . $result . mb_substr($matches[0], $right); } /** * Convert a plain text $url to HTML */ public function linkify(string $url): string { // It comes in special'd, so we unspecial it before passing to the stringifying // functions $url = htmlspecialchars_decode($url); if (str_contains($url, '@') && !str_contains($url, ':') && ($email = filter_var($url, \FILTER_VALIDATE_EMAIL)) !== false) { //url is an email address without the mailto: protocol $url = "mailto:{$email}"; } $attrs = ['href' => $url, 'title' => $url]; // TODO Check to see whether this is a known "attachment" URL. // Whether to nofollow $nf = Common::config('nofollow', 'external'); if ($nf == 'never') { $attrs['rel'] = 'external'; } else { $attrs['rel'] = 'noopener nofollow external noreferrer'; } return HTML::html(['a' => ['attrs' => $attrs, $url]], options: ['indent' => false]); } public function onNoteDeleteRelated(Note &$note, Actor $actor): bool { DB::wrapInTransaction(fn () => NoteToLink::removeWhereNoteId($note->getId())); return Event::next; } }