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 // captcha configuration options 268 echo DOKU_TAB . '$BMConfig = {' . NL; 269 echo DOKU_TAB . DOKU_TAB . '"captchaBypass": ' . json_encode($this->getConf('captchaBypass')) . NL; 270 echo DOKU_TAB . '};' . NL; 271 272 echo '</script>' . NL; 273 274 // insert a warning message for users without JavaScript: 275 echo '<dialog open closedby="any" id="BM__NoJSWarning"><p>' . $this->getLang('bm_noJsWarning') . '</p></dialog>' . NL; 276 277 } 278 } 279 280 public function showImageCaptcha(Event $event, $param) { 281 282 $useCaptcha = $this->getConf('useCaptcha'); 283 284 echo '<script>' . $this->getBMHeader($event, $param) . '</script>'; 285 286 $cCode = '-'; 287 if ($useCaptcha !== 'disabled') { 288 if ($this->captchaWhitelisted()) { 289 $cCode = 'W'; // whitelisted 290 } 291 elseif ($this->hasCaptchaCookie()) { 292 $cCode = 'N'; // user already has a cookie 293 } 294 else { 295 $cCode = 'Y'; // show the captcha 296 297 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 298 $event->preventDefault(); // don't show normal content 299 300 // TODO Insert dummy image 301 $this->insertCaptchaLoader(); // and load the captcha 302 } 303 }; 304 305 $this->showCaptcha = $cCode; // store the captcha code for the logfile 306 } 307 308 /** 309 * Checks if the user has a valid captcha cookie. 310 * 311 * @return boolean 312 * @access private 313 * 314 **/ 315 private function hasCaptchaCookie() { 316 317 $cookieVal = isset($_COOKIE['DWConfirm']) ? $_COOKIE['DWConfirm'] : null; 318 319 $today = substr((new DateTime())->format('c'), 0, 10); 320 321 $raw = $this->getConf('captchaSeed') . ';' . $_SERVER['SERVER_NAME'] . ';' . $_SERVER['REMOTE_ADDR'] . ';' . $today; 322 $expected = hash('sha256', $raw); 323 324 // for debugging: write captcha data to the log: 325 $this->writeCaptchaLog($_SERVER['REMOTE_ADDR'], $cookieVal, $_SERVER['SERVER_NAME'], $expected); 326 327 return $cookieVal == $expected; 328 } 329 330 /** 331 * Writes data to the captcha log. 332 * 333 * @return void 334 */ 335 private function writeCaptchaLog($remote_addr, $cookieVal, $serverName, $expected) { 336 337 global $INFO; 338 339 $logArr = Array( 340 $remote_addr, /* remote IP */ 341 $cookieVal, /* cookie value */ 342 $this->getConf('captchaSeed'), /* seed */ 343 $serverName, /* server name */ 344 $expected, /* expected cookie value */ 345 ($cookieVal == $expected ? 'MATCH' : 'WRONG'), /* cookie matches expected value? */ 346 $_SERVER['REQUEST_URI'] /* request URI */ 347 ); 348 349 //* create the log line */ 350 $filename = __DIR__ .'/logs/' . gmdate('Y-m-d') . '.captcha.txt'; /* use GMT date for filename */ 351 $logline = gmdate('Y-m-d H:i:s'); /* use GMT time for log entries */ 352 foreach ($logArr as $tab) { 353 $logline .= "\t" . $tab; 354 }; 355 356 /* write the log line to the file */ 357 $logfile = fopen($filename, 'a'); 358 if (!$logfile) die(); 359 if (fwrite($logfile, $logline . "\n") === false) { 360 fclose($logfile); 361 die(); 362 } 363 364 // in case of errors, write the cookie data to the log: 365 if (!$cookieVal) { 366 $logline = print_r($_COOKIE, true); 367 if (fwrite($logfile, $logline . "\n") === false) { 368 fclose($logfile); 369 die(); 370 } 371 } 372 373 /* Done. close the file. */ 374 fclose($logfile); 375 } 376 377 378 // check if the visitor's IP is on a whitelist: 379 private function captchaWhitelisted() { 380 381 // normalise IP address: 382 $ip = inet_pton($_SERVER['REMOTE_ADDR']); 383 384 // find which file to open: 385 $prefixes = ['user', 'default']; 386 foreach ($prefixes as $pre) { 387 $filename = __DIR__ .'/config/' . $pre . '-whitelist.txt'; 388 if (file_exists($filename)) { 389 break; 390 } 391 } 392 393 if (file_exists($filename)) { 394 $lines = file($filename, FILE_SKIP_EMPTY_LINES); 395 foreach ($lines as $line) { 396 if (trim($line) !== '' && !str_starts_with($line, '#')) { 397 $col = explode("\t", $line); 398 if (count($col) >= 2) { 399 $from = inet_pton($col[0]); 400 $to = inet_pton($col[1]); 401 402 if ($ip >= $from && $ip <= $to) { 403 return true; /* IP whitelisted */ 404 } 405 } 406 } 407 } 408 } 409 return false; /* IP not found in whitelist */ 410 } 411 412 // inserts a blank box to ensure there is enough space for the captcha: 413 private function insertLoremIpsum() { 414 415 echo '<div class="level1">' . NL; 416 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; 417 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; 418 echo '</div>' . NL; 419 420 } 421 422 /* Generates a few paragraphs of Dada text to show instead of the article content */ 423 private function insertDadaFiller() { 424 425 global $conf; 426 global $TOC; 427 global $ID; 428 429 // list of languages to search for the wordlist 430 $langs = array_unique([$conf['lang'], 'la']); 431 432 // find path to the first available wordlist: 433 foreach ($langs as $lang) { 434 $filename = __DIR__ .'/lang/' . $lang . '/wordlist.txt'; /* language-specific wordlist */ 435 if (file_exists($filename)) { 436 break; 437 } 438 } 439 440 // load the wordlist file: 441 if (file_exists($filename)) { 442 $words = array(); 443 $totalWeight = 0; 444 $lines = file($filename, FILE_SKIP_EMPTY_LINES); 445 foreach ($lines as $line) { 446 $arr = explode("\t", $line); 447 $arr[1] = ( count($arr) > 1 ? (int) trim($arr[1]) : 1 ); 448 $totalWeight += (int) $arr[1]; 449 array_push($words, $arr); 450 } 451 } else { 452 echo '<script> console.log("Can’t generate filler text: wordlist file not found!"); </script>'; 453 return; 454 } 455 456 // If a TOC exists, use it for the headlines: 457 if(is_array($TOC)) { 458 $toc = $TOC; 459 } else { 460 $meta = p_get_metadata($ID, '', METADATA_RENDER_USING_CACHE); 461 //$tocok = (isset($meta['internal']['toc']) ? $meta['internal']['toc'] : $tocok = true); 462 $toc = isset($meta['description']['tableofcontents']) ? $meta['description']['tableofcontents'] : null; 463 } 464 if (!$toc) { // no TOC, generate my own: 465 $hlCount = mt_rand(0, (int) $conf['tocminheads']); 466 $toc = array(); 467 for ($i=0; $i<$hlCount; $i++) { 468 array_push($toc, $this->dadaMakeHeadline($words, $totalWeight)); // $toc 469 } 470 } 471 472 // if H1 heading is not in the TOC, add a chappeau section: 473 $chapeauCount = mt_rand(1, 3); 474 if ((int) $conf['toptoclevel'] > 1) { 475 echo "<div class=\"level1\">\n"; 476 for ($i=0; $i<$chapeauCount; $i++) { 477 echo $this->dadaMakeParagraph($words, $totalWeight); 478 } 479 echo "</div>\n"; 480 } 481 482 // text sections for each sub-headline: 483 foreach ($toc as $hl) { 484 echo $this->dadaMakeSection($words, $totalWeight, $hl); 485 } 486 } 487 488 private function dadaMakeSection($words, $totalWeight, $hl) { 489 490 global $conf; 491 492 // how many paragraphs? 493 $paragraphCount = mt_rand(1, 4); 494 495 // section level 496 $topTocLevel = (int) $conf['toptoclevel']; 497 $secLevel = $hl['level'] + 1;; 498 499 // return value: 500 $sec = ""; 501 502 // make a headline: 503 if ($topTocLevel > 1 || $secLevel > 1) { 504 $sec .= "<h{$secLevel} id=\"{$hl['hid']}\">{$hl['title']}</h{$secLevel}>\n"; 505 } 506 507 // add the paragraphs: 508 $sec .= "<div class=\"level{$secLevel}\">\n"; 509 for ($i=0; $i<$paragraphCount; $i++) { 510 $sec .= $this->dadaMakeParagraph($words, $totalWeight); 511 } 512 $sec .= "</div>\n"; 513 514 return $sec; 515 } 516 517 private function dadaMakeHeadline($words, $totalWeight) { 518 519 // how many words to generate? 520 $wordCount = mt_rand(2, 5); 521 522 // function returns an array: 523 $r = Array(); 524 525 // generate the headline: 526 $hlArr = array(); 527 for ($i=0; $i<$wordCount; $i++) { 528 array_push($hlArr, $this->dadaSelectRandomWord($words, $totalWeight)); 529 } 530 531 $r['title'] = ucfirst(implode(' ', $hlArr)); 532 533 $r['hid'] = preg_replace('/[^\w\d\-]+/i', '_', strtolower($r['title'])); 534 $r['type'] = 'ul'; // always ul! 535 $r['level'] = 1; // always level 1 for now 536 537 return $r; 538 } 539 540 private function dadaMakeParagraph($words, $totalWeight) { 541 542 // how many words to generate? 543 $sentenceCount = mt_rand(2, 5); 544 545 $paragraph = array(); 546 for ($i=0; $i<$sentenceCount; $i++) { 547 array_push($paragraph, $this->dadaMakeSentence($words, $totalWeight)); 548 } 549 550 return "<p>\n" . implode(' ', $paragraph) . "\n</p>\n"; 551 552 } 553 554 private function dadaMakeSentence($words, $totalWeight) { 555 556 // how many words to generate? 557 $wordCount = mt_rand(4, 20); 558 559 // generate the sentence: 560 $sentence = array(); 561 for ($i=0; $i<$wordCount; $i++) { 562 array_push($sentence, $this->dadaSelectRandomWord($words, $totalWeight)); 563 } 564 565 return ucfirst(implode(' ', $sentence)) . '.'; 566 567 } 568 569 private function dadaSelectRandomWord($list, $totalWeight) { 570 571 // get a random selection: 572 $rand = mt_rand(0, $totalWeight); 573 574 // match the selection to the weighted list: 575 $cumulativeWeight = 0; 576 for ($i=0; $i<count($list); $i++) { 577 $cumulativeWeight += $list[$i][1]; 578 if ($cumulativeWeight >= $rand) { 579 return $list[$i][0]; 580 } 581 } 582 return '***'; 583 } 584 585}