1<?php 2 3use dokuwiki\Extension\EventHandler; 4use dokuwiki\Extension\Event; 5use dokuwiki\Logger; 6 7/** 8 * Action Component for the Bot Monitoring Plugin 9 * 10 * @license GPL 3 (http://www.gnu.org/licenses/gpl.html) 11 * @author Sascha Leib <sascha.leib(at)kolmio.com> 12 */ 13 14class action_plugin_botmon extends DokuWiki_Action_Plugin { 15 16 public function __construct() { 17 18 // determine if a captcha should be loaded: 19 $this->showCaptcha = 'Z'; // Captcha unknown 20 21 $useCaptcha = $this->getConf('useCaptcha'); // should we show a captcha? 22 23 if ($useCaptcha !== 'disabled') { 24 if ($_SERVER['REQUEST_METHOD'] == 'HEAD') { 25 $this->showCaptcha = 'H'; // Method is HEAD, no need for captcha 26 } elseif ($this->captchaWhitelisted()) { 27 $this->showCaptcha = 'W'; // IP is whitelisted, no captcha 28 } elseif ($this->hasCaptchaCookie()) { 29 $this->showCaptcha = 'N'; // No, user already has a cookie, don't show the captcha 30 } else { 31 $this->showCaptcha = 'Y'; // Yes, show the captcha 32 } 33 } 34 } 35 36 /** 37 * Registers a callback functions 38 * 39 * @param EventHandler $controller DokuWiki's event controller object 40 * @return void 41 */ 42 public function register(EventHandler $controller) { 43 44 global $ACT; 45 46 // populate the session id and type: 47 $this->setSessionInfo(); 48 49 // insert header data into the page: 50 if ($ACT == 'show' || $ACT == 'edit' || $ACT == 'media') { 51 $controller->register_hook('TPL_METAHEADER_OUTPUT', 'BEFORE', $this, 'insertHeader'); 52 53 // Override the page rendering, if a captcha needs to be displayed: 54 $controller->register_hook('TPL_ACT_RENDER', 'BEFORE', $this, 'insertCaptchaCode'); 55 56 } else if ($ACT == 'admin' && isset($_REQUEST['page']) && $_REQUEST['page'] == 'botmon') { 57 $controller->register_hook('TPL_METAHEADER_OUTPUT', 'BEFORE', $this, 'insertAdminHeader'); 58 } 59 60 // also show a captcha before the image preview 61 $controller->register_hook('TPL_IMG_DISPLAY', 'BEFORE', $this, 'showImageCaptcha'); 62 63 // write to the log after the page content was displayed: 64 $controller->register_hook('TPL_CONTENT_DISPLAY', 'AFTER', $this, 'writeServerLog'); 65 66 } 67 68 /* session information */ 69 private $sessionId = null; 70 private $sessionType = ''; 71 private $showCaptcha = 'X'; 72 73 /** 74 * Inserts tracking code to the page header 75 * (only called on 'show' actions) 76 * 77 * @param Event $event event object by reference 78 * @return void 79 */ 80 public function insertHeader(Event $event, $param) { 81 82 global $INFO; 83 84 85 // build the tracker code: 86 $code = $this->getBMHeader(); 87 88 // add the deferred script loader:: 89 $code .= DOKU_TAB . DOKU_TAB . "addEventListener('DOMContentLoaded', function(){" . NL; 90 $code .= DOKU_TAB . DOKU_TAB . DOKU_TAB . "const e=document.createElement('script');" . NL; 91 $code .= DOKU_TAB . DOKU_TAB . DOKU_TAB . "e.async=true;e.defer=true;" . NL; 92 $code .= DOKU_TAB . DOKU_TAB . DOKU_TAB . "e.src='".DOKU_BASE."lib/plugins/botmon/client.js';" . NL; 93 $code .= DOKU_TAB . DOKU_TAB . DOKU_TAB . "document.getElementsByTagName('head')[0].appendChild(e);" . NL; 94 $code .= DOKU_TAB . DOKU_TAB . "});"; 95 $event->data['script'][] = ['_data' => $code]; 96 } 97 98 /* create the BM object code for insertion into a script element: */ 99 private function getBMHeader() { 100 101 // build the tracker code: 102 $code = DOKU_TAB . DOKU_TAB . "document._botmon = {t0: Date.now(), session: " . json_encode($this->sessionId) . ", seed: " . json_encode($this->getConf('captchaSeed')) . ", ip: " . json_encode($_SERVER['REMOTE_ADDR']) . "};" . NL; 103 104 // is there a user logged in? 105 $username = ( !empty($INFO['userinfo']) && !empty($INFO['userinfo']['name']) ? $INFO['userinfo']['name'] : ''); 106 if ($username) { 107 $code .= DOKU_TAB . DOKU_TAB . 'document._botmon.user = "' . $username . '";'. NL; 108 } 109 110 return $code; 111 112 } 113 114 /** 115 * Inserts tracking code to the page header 116 * (only called on 'show' actions) 117 * 118 * @param Event $event event object by reference 119 * @return void 120 */ 121 public function insertAdminHeader(Event $event, $param) { 122 123 $event->data['link'][] = ['rel' => 'stylesheet', 'href' => DOKU_BASE.'lib/plugins/botmon/admin.css', 'defer' => 'defer']; 124 $event->data['script'][] = ['src' => DOKU_BASE.'lib/plugins/botmon/admin.js', 'defer' => 'defer', '_data' => '']; 125 } 126 127 /** 128 * Writes data to the server log. 129 * 130 * @return void 131 */ 132 public function writeServerLog(Event $event, $param) { 133 134 global $conf; 135 global $INFO; 136 137 // is there a user logged in? 138 $username = ( !empty($INFO['userinfo']) && !empty($INFO['userinfo']['name']) 139 ? $INFO['userinfo']['name'] : ''); 140 141 // clean the page ID 142 $pageId = preg_replace('/[\x00-\x1F]/', "\u{FFFD}", $INFO['id'] ?? ''); 143 144 // create the log array: 145 $logArr = Array( 146 $_SERVER['REMOTE_ADDR'], /* remote IP */ 147 $pageId, /* page ID */ 148 $this->sessionId, /* Session ID */ 149 $this->sessionType, /* session ID type */ 150 $username, /* user name */ 151 $_SERVER['HTTP_USER_AGENT'] ?? '', /* User agent */ 152 $_SERVER['HTTP_REFERER'] ?? '', /* HTTP Referrer */ 153 substr($conf['lang'],0,2), /* page language */ 154 implode(',', array_unique(array_map( function($it) { return substr(trim($it),0,2); }, explode(',',trim($_SERVER['HTTP_ACCEPT_LANGUAGE'], " \t;,*"))))), /* accepted client languages */ 155 $this->getCountryCode(), /* GeoIP country code */ 156 $this->showCaptcha, /* show captcha? */ 157 $_SERVER['REQUEST_METHOD'] ?? '' /* request method */ 158 ); 159 160 //* create the log line */ 161 $filename = __DIR__ .'/logs/' . gmdate('Y-m-d') . '.srv.txt'; /* use GMT date for filename */ 162 $logline = gmdate('Y-m-d H:i:s'); /* use GMT time for log entries */ 163 foreach ($logArr as $tab) { 164 $logline .= "\t" . $tab; 165 }; 166 167 /* write the log line to the file */ 168 $logfile = fopen($filename, 'a'); 169 if (!$logfile) die(); 170 if (fwrite($logfile, $logline . "\n") === false) { 171 fclose($logfile); 172 die(); 173 } 174 175 /* Done */ 176 fclose($logfile); 177 } 178 179 private function getCountryCode() { 180 181 $country = ( $_SERVER['REMOTE_ADDR'] == '127.0.0.1' ? 'local' : 'ZZ' ); // default if no geoip is available! 182 183 $lib = $this->getConf('geoiplib'); /* which library to use? (can only be phpgeoip or disabled) */ 184 185 try { 186 187 // use GeoIP module? 188 if ($lib == 'phpgeoip' && extension_loaded('geoip') && geoip_db_avail(GEOIP_COUNTRY_EDITION)) { // Use PHP GeoIP module 189 $result = geoip_country_code_by_name($_SERVER['REMOTE_ADDR']); 190 $country = ($result ? $result : $country); 191 } 192 } catch (Exception $e) { 193 Logger::error('BotMon Plugin: GeoIP Error', $e->getMessage()); 194 } 195 196 return $country; 197 } 198 199 private function setSessionInfo() { 200 201 // what is the session identifier? 202 if (isset($_SESSION)) { 203 $sesKeys = array_keys($_SESSION); /* DokuWiki Session ID preferred */ 204 foreach ($sesKeys as $key) { 205 if (substr($key, 0, 2) == 'DW') { 206 $this->sessionId = $key; 207 $this->sessionType = 'dw'; 208 return; 209 } 210 } 211 } 212 if (!$this->sessionId) { /* no DokuWiki Session ID, try PHP session ID */ 213 $this->sessionId = session_id(); 214 $this->sessionType = 'php'; 215 } 216 if (!$this->sessionId) { /* no PHP session ID, try IP address */ 217 $this->sessionId = $_SERVER['REMOTE_ADDR']; 218 $this->sessionType = 'ip'; 219 } 220 221 if (!$this->sessionId) { /* if all fails, use random data */ 222 $this->sessionId = rand(100000000, 999999999); 223 $this->sessionType = 'rnd'; 224 } 225 226 } 227 228 public function insertCaptchaCode(Event $event) { 229 230 $useCaptcha = $this->getConf('useCaptcha'); // which background to show? 231 232 // only if we previously determined that we need a captcha: 233 if ($this->showCaptcha == 'Y') { 234 235 echo '<h1 class="sectionedit1">'; tpl_pagetitle(); echo "</h1>\n"; // always show the original page title 236 $event->preventDefault(); // don't show normal content 237 switch ($useCaptcha) { 238 case 'loremipsum': 239 $this->insertLoremIpsum(); // show dada filler instead of text 240 break; 241 case 'dada': 242 $this->insertDadaFiller(); // show dada filler instead of text 243 break; 244 } 245 246 // insert the captcha loader code: 247 echo '<script>' . NL; 248 249 // add the deferred script loader:: 250 echo DOKU_TAB . "addEventListener('DOMContentLoaded', function(){" . NL; 251 echo DOKU_TAB . DOKU_TAB . "const cj=document.createElement('script');" . NL; 252 echo DOKU_TAB . DOKU_TAB . "cj.async=true;cj.defer=true;cj.type='text/javascript';" . NL; 253 echo DOKU_TAB . DOKU_TAB . "cj.src='".DOKU_BASE."lib/plugins/botmon/captcha.js';" . NL; 254 echo DOKU_TAB . DOKU_TAB . "document.getElementsByTagName('head')[0].appendChild(cj);" . NL; 255 echo DOKU_TAB . "});" . NL; 256 257 // add the translated strings for the captcha: 258 echo DOKU_TAB . '$BMLocales = {' . NL; 259 echo DOKU_TAB . DOKU_TAB . '"dlgTitle": ' . json_encode($this->getLang('bm_dlgTitle')) . ',' . NL; 260 echo DOKU_TAB . DOKU_TAB . '"dlgSubtitle": ' . json_encode($this->getLang('bm_dlgSubtitle')) . ',' . NL; 261 echo DOKU_TAB . DOKU_TAB . '"dlgConfirm": ' . json_encode($this->getLang('bm_dlgConfirm')) . ',' . NL; 262 echo DOKU_TAB . DOKU_TAB . '"dlgChecking": ' . json_encode($this->getLang('bm_dlgChecking')) . ',' . NL; 263 echo DOKU_TAB . DOKU_TAB . '"dlgLoading": ' . json_encode($this->getLang('bm_dlgLoading')) . ',' . NL; 264 echo DOKU_TAB . DOKU_TAB . '"dlgError": ' . json_encode($this->getLang('bm_dlgError')) . ',' . NL; 265 echo DOKU_TAB . '};' . NL; 266 267 echo '</script>' . NL; 268 } 269 } 270 271 public function showImageCaptcha(Event $event, $param) { 272 273 $useCaptcha = $this->getConf('useCaptcha'); 274 275 echo '<script>' . $this->getBMHeader($event, $param) . '</script>'; 276 277 $cCode = '-'; 278 if ($useCaptcha !== 'disabled') { 279 if ($this->captchaWhitelisted()) { 280 $cCode = 'W'; // whitelisted 281 } 282 elseif ($this->hasCaptchaCookie()) { 283 $cCode = 'N'; // user already has a cookie 284 } 285 else { 286 $cCode = 'Y'; // show the captcha 287 288 echo '<svg width="100%" height="100%" viewBox="0 0 800 400" version="1.1" xmlns="http://www.w3.org/2000/svg"><path d="M1,1l798,398" style="fill:none;stroke:#f00;stroke-width:1px;"/><path d="M1,399l798,-398" style="fill:none;stroke:#f00;stroke-width:1px;"/><rect x="1" y="1" width="798" height="398" style="fill:none;stroke:#000;stroke-width:1px;"/></svg>'; // placeholder image 289 $event->preventDefault(); // don't show normal content 290 291 // TODO Insert dummy image 292 $this->insertCaptchaLoader(); // and load the captcha 293 } 294 }; 295 296 $this->showCaptcha = $cCode; // store the captcha code for the logfile 297 } 298 299 private function hasCaptchaCookie() { 300 301 $cookieVal = isset($_COOKIE['DWConfirm']) ? $_COOKIE['DWConfirm'] : null; 302 303 $today = substr((new DateTime())->format('c'), 0, 10); 304 305 $raw = $this->getConf('captchaSeed') . '|' . $_SERVER['SERVER_NAME'] . '|' . $_SERVER['REMOTE_ADDR'] . '|' . $today; 306 $expected = hash('sha256', $raw); 307 308 //echo '<ul><li>cookie: ' . $cookieVal . '</li><li>expected: ' . $expected . '</li><li>matches: ' .($cookieVal == $expected ? 'true' : 'false') . '</li></ul>'; 309 310 return $cookieVal == $expected; 311 } 312 313 // check if the visitor's IP is on a whitelist: 314 private function captchaWhitelisted() { 315 316 // normalise IP address: 317 $ip = inet_pton($_SERVER['REMOTE_ADDR']); 318 319 // find which file to open: 320 $prefixes = ['user', 'default']; 321 foreach ($prefixes as $pre) { 322 $filename = __DIR__ .'/config/' . $pre . '-whitelist.txt'; 323 if (file_exists($filename)) { 324 break; 325 } 326 } 327 328 if (file_exists($filename)) { 329 $lines = file($filename, FILE_SKIP_EMPTY_LINES); 330 foreach ($lines as $line) { 331 if (trim($line) !== '' && !str_starts_with($line, '#')) { 332 $col = explode("\t", $line); 333 if (count($col) >= 2) { 334 $from = inet_pton($col[0]); 335 $to = inet_pton($col[1]); 336 337 if ($ip >= $from && $ip <= $to) { 338 return true; /* IP whitelisted */ 339 } 340 } 341 } 342 } 343 } 344 return false; /* IP not found in whitelist */ 345 } 346 347 // inserts a blank box to ensure there is enough space for the captcha: 348 private function insertLoremIpsum() { 349 350 echo '<div class="level1">' . NL; 351 echo '<p>' . NL . 'Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.'. NL . '</p>' . NL; 352 echo '<p>' . NL . 'At vero eos et accusamus et iusto odio dignissimos ducimus, qui blanditiis praesentium voluptatum deleniti atque corrupti, quos dolores et quas molestias excepturi sint, obcaecati cupiditate non provident, similique sunt in culpa, qui officia deserunt mollitia animi, id est laborum et dolorum fuga.'. NL . '</p>' . NL; 353 echo '</div>' . NL; 354 355 } 356 357 /* Generates a few paragraphs of Dada text to show instead of the article content */ 358 private function insertDadaFiller() { 359 360 global $conf; 361 global $TOC; 362 global $ID; 363 364 // list of languages to search for the wordlist 365 $langs = array_unique([$conf['lang'], 'la']); 366 367 // find path to the first available wordlist: 368 foreach ($langs as $lang) { 369 $filename = __DIR__ .'/lang/' . $lang . '/wordlist.txt'; /* language-specific wordlist */ 370 if (file_exists($filename)) { 371 break; 372 } 373 } 374 375 // load the wordlist file: 376 if (file_exists($filename)) { 377 $words = array(); 378 $totalWeight = 0; 379 $lines = file($filename, FILE_SKIP_EMPTY_LINES); 380 foreach ($lines as $line) { 381 $arr = explode("\t", $line); 382 $arr[1] = ( count($arr) > 1 ? (int) trim($arr[1]) : 1 ); 383 $totalWeight += (int) $arr[1]; 384 array_push($words, $arr); 385 } 386 } else { 387 echo '<script> console.log("Can’t generate filler text: wordlist file not found!"); </script>'; 388 return; 389 } 390 391 // If a TOC exists, use it for the headlines: 392 if(is_array($TOC)) { 393 $toc = $TOC; 394 } else { 395 $meta = p_get_metadata($ID, '', METADATA_RENDER_USING_CACHE); 396 //$tocok = (isset($meta['internal']['toc']) ? $meta['internal']['toc'] : $tocok = true); 397 $toc = isset($meta['description']['tableofcontents']) ? $meta['description']['tableofcontents'] : null; 398 } 399 if (!$toc) { // no TOC, generate my own: 400 $hlCount = mt_rand(0, (int) $conf['tocminheads']); 401 $toc = array(); 402 for ($i=0; $i<$hlCount; $i++) { 403 array_push($toc, $this->dadaMakeHeadline($words, $totalWeight)); // $toc 404 } 405 } 406 407 // if H1 heading is not in the TOC, add a chappeau section: 408 $chapeauCount = mt_rand(1, 3); 409 if ((int) $conf['toptoclevel'] > 1) { 410 echo "<div class=\"level1\">\n"; 411 for ($i=0; $i<$chapeauCount; $i++) { 412 echo $this->dadaMakeParagraph($words, $totalWeight); 413 } 414 echo "</div>\n"; 415 } 416 417 // text sections for each sub-headline: 418 foreach ($toc as $hl) { 419 echo $this->dadaMakeSection($words, $totalWeight, $hl); 420 } 421 } 422 423 private function dadaMakeSection($words, $totalWeight, $hl) { 424 425 global $conf; 426 427 // how many paragraphs? 428 $paragraphCount = mt_rand(1, 4); 429 430 // section level 431 $topTocLevel = (int) $conf['toptoclevel']; 432 $secLevel = $hl['level'] + 1;; 433 434 // return value: 435 $sec = ""; 436 437 // make a headline: 438 if ($topTocLevel > 1 || $secLevel > 1) { 439 $sec .= "<h{$secLevel} id=\"{$hl['hid']}\">{$hl['title']}</h{$secLevel}>\n"; 440 } 441 442 // add the paragraphs: 443 $sec .= "<div class=\"level{$secLevel}\">\n"; 444 for ($i=0; $i<$paragraphCount; $i++) { 445 $sec .= $this->dadaMakeParagraph($words, $totalWeight); 446 } 447 $sec .= "</div>\n"; 448 449 return $sec; 450 } 451 452 private function dadaMakeHeadline($words, $totalWeight) { 453 454 // how many words to generate? 455 $wordCount = mt_rand(2, 5); 456 457 // function returns an array: 458 $r = Array(); 459 460 // generate the headline: 461 $hlArr = array(); 462 for ($i=0; $i<$wordCount; $i++) { 463 array_push($hlArr, $this->dadaSelectRandomWord($words, $totalWeight)); 464 } 465 466 $r['title'] = ucfirst(implode(' ', $hlArr)); 467 468 $r['hid'] = preg_replace('/[^\w\d\-]+/i', '_', strtolower($r['title'])); 469 $r['type'] = 'ul'; // always ul! 470 $r['level'] = 1; // always level 1 for now 471 472 return $r; 473 } 474 475 private function dadaMakeParagraph($words, $totalWeight) { 476 477 // how many words to generate? 478 $sentenceCount = mt_rand(2, 5); 479 480 $paragraph = array(); 481 for ($i=0; $i<$sentenceCount; $i++) { 482 array_push($paragraph, $this->dadaMakeSentence($words, $totalWeight)); 483 } 484 485 return "<p>\n" . implode(' ', $paragraph) . "\n</p>\n"; 486 487 } 488 489 private function dadaMakeSentence($words, $totalWeight) { 490 491 // how many words to generate? 492 $wordCount = mt_rand(4, 20); 493 494 // generate the sentence: 495 $sentence = array(); 496 for ($i=0; $i<$wordCount; $i++) { 497 array_push($sentence, $this->dadaSelectRandomWord($words, $totalWeight)); 498 } 499 500 return ucfirst(implode(' ', $sentence)) . '.'; 501 502 } 503 504 private function dadaSelectRandomWord($list, $totalWeight) { 505 506 // get a random selection: 507 $rand = mt_rand(0, $totalWeight); 508 509 // match the selection to the weighted list: 510 $cumulativeWeight = 0; 511 for ($i=0; $i<count($list); $i++) { 512 $cumulativeWeight += $list[$i][1]; 513 if ($cumulativeWeight >= $rand) { 514 return $list[$i][0]; 515 } 516 } 517 return '***'; 518 } 519 520}