1<?php 2/** 3 * Search with Scopes 4 * 5 * @license GPL 2 (http://www.gnu.org/licenses/gpl.html) 6 * @author i-net software <tools@inetsoftware.de> 7 * @author Gerry Weissbach <gweissbach@inetsoftware.de> 8 */ 9 10// must be run within Dokuwiki 11if (!defined('DOKU_INC')) die(); 12if (!defined('DOKU_PLUGIN')) define('DOKU_PLUGIN', DOKU_INC . 'lib/plugins/'); 13 14require_once(DOKU_PLUGIN . 'syntax.php'); 15 16class syntax_plugin_siteexport_toc extends DokuWiki_Syntax_Plugin { 17 18 private $insideToc = false; 19 private $savedToc = array(); 20 private $options = array(); 21 22 private $mergedPages = array(); 23 private $includedPages = array(); 24 private $merghintIds = array(); 25 26 public function getType() { return 'protected'; } 27 public function getPType() { return 'block'; } 28 public function getAllowedTypes() { return array('container'); } 29 public function getSort() { return 100; } 30 31 /** 32 * Connect pattern to lexer 33 */ 34 public function connectTo($mode) { 35 $this->Lexer->addEntryPattern('<toc>(?=.*?</toc>)', $mode, 'plugin_siteexport_toc'); 36 $this->Lexer->addEntryPattern('<toc .+?>(?=.*?</toc>)', $mode, 'plugin_siteexport_toc'); 37 $this->Lexer->addSpecialPattern("\[\[.+?\]\]", $mode, 'plugin_siteexport_toc'); 38 } 39 40 public function postConnect() { 41 $this->Lexer->addExitPattern('</toc.*?>', 'plugin_siteexport_toc'); 42 } 43 44 public function handle($match, $state, $pos, Doku_Handler $handler) { 45 global $ID, $INFO; 46 47 switch ($state) { 48 case DOKU_LEXER_ENTER: 49 50 $this->insideToc = true; 51 $this->options = explode(' ', substr($match, 5, -1)?:""); 52 return array('start' => true, 'pos' => $pos, 'options' => $this->options); 53 54 case DOKU_LEXER_SPECIAL: 55 56 if ($this->insideToc) { 57 58 $link = preg_replace(array('/^\[\[/', '/\]\]$/u'), '', $match); 59 // Split title from URL 60 $link = explode('|', $link, 2); 61 if (!isset($link[1])) { 62 $link[1] = NULL; 63 } else if (preg_match('/^\{\{[^\}]+\}\}$/', $link[1])) { 64 // If the title is an image, convert it to an array containing the image details 65 $link[1] = Doku_Handler_Parse_Media($link[1]); 66 } 67 $link[0] = trim($link[0]); 68 69 if (!(preg_match('/^[a-zA-Z0-9\.]+>{1}.*$/u', $link[0]) || 70 preg_match('/^\\\\\\\\[\w.:?\-;,]+?\\\\/u', $link[0]) || 71 preg_match('#^([a-z0-9\-\.+]+?)://#i', $link[0]) || 72 preg_match('<' . PREG_PATTERN_VALID_EMAIL . '>', $link[0]) || 73 preg_match('!^#.+!', $link[0])) 74 ) { 75 76 // Get current depth from call stack 77 $depth = 1; 78 if ($handler->CallWriter instanceof Doku_Handler_List) { 79 80 $calls = array_reverse($handler->CallWriter->calls); 81 $call = $calls[0]; 82 foreach ($calls as $item) { 83 if (in_array($item[0], array('list_item', 'list_open'))) { $call = $item; break; } 84 } 85 86 $listType = null; 87 $depth = $handler->CallWriter->interpretSyntax($call[1][0], $listType)-1; // Minus one because of plus one inside the interpret function 88 } 89 90 if (empty($link[0])) { break; } // No empty elements. This would lead to problems 91 return array($link[0], $link[1], $depth); 92 } else { 93 // use parser! - but with another p 94 $handler->internallink($match, $state, $pos); 95 } 96 } else { 97 // use parser! 98 $handler->internallink($match, $state, $pos); 99 } 100 101 return false; 102 case DOKU_LEXER_UNMATCHED: 103 104 $handler->_addCall('cdata', array($match), $pos); 105 106 return false; 107 case DOKU_LEXER_EXIT: 108 109 $this->insideToc = false; 110 return 'save__meta'; 111 } 112 return false; 113 } 114 115 public function render($mode, Doku_Renderer $renderer, $data) { 116 global $ID, $lang, $INFO; 117 118 list($SID, $NAME, $DEPTH) = $data; 119 120 resolve_pageid(getNS($ID), $SID, $exists = null); 121// $SID = cleanID($SID); // hier kein cleanID, da sonst moeglicherweise der anker verloren geht 122 123 // Render XHTML and ODT 124 if ($mode == 'xhtml' || $mode == 'odt') { 125 126 // TOC Title 127 if (is_array($data) && $data['start'] == true) { 128 129 if (is_Array($data['options'])) { 130 foreach ($data['options'] as $opt) { 131 switch ($opt) { 132 case 'description' : $renderer->meta['sitetoc']['showDescription'] = true; break; 133 case 'notoc' : $renderer->meta['sitetoc']['noTOC'] = true; break; 134 case 'merge' : $renderer->meta['sitetoc']['mergeDoc'] = true; break; 135 case 'nohead' : $renderer->meta['sitetoc']['noTocHeader'] = true; break; 136 case 'mergeheader' : $renderer->meta['sitetoc']['mergeHeader'] = true; break; 137 case 'pagebreak' : $renderer->meta['sitetoc']['pagebreak'] = true; break; 138 case 'mergehint' : $renderer->meta['sitetoc']['mergehint'] = true; break; 139 } 140 } 141 } 142 143 $renderer->section_open("1 sitetoc"); 144 if ($renderer->meta['sitetoc']['noTocHeader'] === false) { 145 $renderer->header($lang['toc'], 1, $data['pos']); 146 } 147 148 return true; 149 } else 150 151 // All Output has been done 152 if (!is_array($data) && $data == 'save__meta') { 153 154 // Close TOC 155 $renderer->section_close(); 156 157 if ($renderer->meta['sitetoc']['noTOC'] === true) { 158 $renderer->doc = preg_replace("/<div.*?sitetoc.*?$/si", "", $renderer->doc); 159 } 160 161 // If this is not set, we may have it as Metadata 162 if (empty($this->mergedPages) && $renderer->meta['sitetoc']['mergeDoc']) { 163 $toc = $renderer->meta['sitetoc']['siteexportTOC']; 164 165 if (is_array($toc)) { 166 foreach ($toc as $tocItem) { 167 $this->mergedPages[] = array($tocItem['id'], $tocItem['depth']); 168 } 169 } 170 171 } 172 173 // If there is some data to be merged 174 if (count($this->mergedPages) > 0) { 175 176 $renderer->doc = ''; // Start fresh! 177 178 $renderer->section_open("1 mergedsite" . ($renderer->meta['sitetoc']['mergehint'] && count($this->mergedPages) > 1 ? ' mergehint' : '')); 179 180 // Prepare lookup Array 181 foreach ($this->mergedPages as $tocItem) { 182 list($this->includedPages[]) = explode('#', $tocItem[0]); 183 } 184 185 // Load the instructions 186 $instr = array(); 187 foreach ($this->mergedPages as $tocElement) { 188 189 list($tocItem, $depth) = $tocElement; 190 $file = wikiFN($tocItem); 191 192 if (@file_exists($file)) { 193 $instructions = p_cached_instructions($file, false, $tocItem); 194 } else { 195 $instructions = p_get_instructions(io_readWikiPage($file, $tocItem)); 196 } 197 198 // Convert Link and header instructions 199 $instructions = $this->_convertInstructions($instructions, $addID = null, $renderer, $depth); 200 201 if ($renderer->meta['sitetoc']['mergeHeader'] && count($this->mergedPages) > 1) { 202 // get a hint for merged pages 203 if ($renderer->meta['sitetoc']['mergehint']) { 204 // only if the first section is already there 205 $mergeHint = p_get_metadata($tocItem, 'mergehint', METADATA_RENDER_USING_SIMPLE_CACHE); 206 if (empty($mergeHint)) { $mergeHint = p_get_metadata($tocItem, 'thema', METADATA_RENDER_USING_SIMPLE_CACHE); } 207 if (empty($mergeHint)) { $mergeHint = tpl_pagetitle($tocItem, true); } 208 $instructions = $this->_mergeWithHeaders($this->_initialHeaderStructure($instructions), $instructions, 1, $mergeHint); 209 } 210 // Merge 211 $instr = $this->_mergeWithHeaders($instr, $instructions, 1); 212 } else 213 if ($renderer->meta['sitetoc']['pagebreak']) { 214 $sitepagebreak = array(array( 215 'plugin', 216 array( 217 'siteexport_toctools', 218 array( 219 'pagebreak', 220 null, 221 null 222 ) 223 ) 224 )); 225 $instr = array_merge($instr, $instructions, $sitepagebreak); 226 } else { 227 // Concat 228 $instr = array_merge($instr, $instructions); 229 } 230 } 231 232 if (!empty($instr)) { 233 $this->_cleanAllInstructions($instr); 234 235 // if the last element is a pagebreak, remove it. 236 if ($instr[count($instr)-1][1][0] == 'siteexport_toctools' && $instr[count($instr)-1][1][1][0] == 'pagebreak' ) { 237 $instr = array_slice($instr, 0, -1); 238 } 239 240 241 // print "<pre>"; print_r($instr); print "</pre>"; 242 $this->_render_output($renderer, $mode, $instr); 243 } 244 245 $renderer->section_close(); 246 } 247 return true; 248 } 249 250 // Save the current ID 251 $LNID = $SID; 252 253 // Add ID to flags['mergeDoc'] 254 if ($renderer->meta['sitetoc']['mergeDoc'] === true) { // || (count($renderer->meta['sitetoc']['siteexportTOC']) > 0 && $renderer->meta['sitetoc']['siteexportMergeDoc'] === true) ) { 255 $this->mergedPages[] = array($SID, $DEPTH); 256 resolve_pageid(getNS($ID), $SID, $exists); 257 } else { 258 // // print normal internal link (XHTML odt) 259 $renderer->internallink($LNID, $NAME, null); 260 261 // Display Description underneath 262 if ($renderer->meta['sitetoc']['showDescription'] === true) { 263 $renderer->cdata(p_get_metadata($SID, 'description abstract', true)); 264 } 265 } 266 267 // Render Metadata 268 } else if ($mode == 'metadata') { 269 if (!is_array($data) && $data == 'save__meta') { 270 $renderer->meta['sitetoc']['siteexportTOC'] = $this->savedToc; 271 272 foreach ($this->savedToc as $page) { 273 $renderer->meta['relation']['references'][$page['id']] = $page['exists']; 274 } 275 276 $this->savedToc = array(); 277 } else if (!isset($data['start']) && !isset($data['pos'])) { 278 $this->savedToc[] = $this->__addTocItem($SID, $NAME, $DEPTH, $renderer); 279 } 280 } else { 281 return false; 282 } 283 284 return true; 285 } 286 287 /* 288 * pull apart the ID and create an Entry for the TOC 289 */ 290 private function __addTocItem($id, $name, $depth, $renderer) { 291 global $conf; 292 global $ID; 293 294 // Render Title 295 $default = $renderer->_simpleTitle($id); 296 $exists = false; $isImage = false; $linktype = null; 297 resolve_pageid(getNS($ID), $id, $exists); 298 $name = $renderer->_getLinkTitle($name, $default, $isImage, $id, $linktype); 299 300 //keep hash anchor 301 list($id, $hash) = explode('#', $id, 2); 302 if (!empty($hash)) $hash = $renderer->_headerToLink($hash); 303 304 // Build Sitetoc Item 305 $item = array(); 306 $item['id'] = $id; 307 $item['name'] = $name; 308 $item['anchor'] = $hash; 309 $item['depth'] = $depth; 310 $item['exists'] = $exists; 311 if (!$conf['skipacl'] && auth_quickaclcheck($item['id']) < AUTH_READ) { 312 return false; 313 } 314 315 return $item; 316 } 317 318 /* 319 * Render the output of one page 320 */ 321 private function _render_output($renderer, $mode, $instr) { 322 global $ID; 323 324 // Section IDs 325 // $addID = sectionID($addID, $check); //not possible to use a:b:c for id 326 327 if ($mode == 'xhtml') { 328 329 //--------RENDER 330 //renderer information(TOC build / Cache used) 331 $info = array(); 332 $content = p_render($mode, $instr, $info); 333 334 //Remove TOC`s, section edit buttons and tags 335 $content = $this->_cleanXHTML($content); 336 337 // embed the included page 338 // $renderer->doc .= '<div class="include">'; 339 //add an anchor to find start of a inserted page 340 // $renderer->doc .= "<a name='$addID' id='$addID'>"; 341 $renderer->doc .= $content; 342 // $renderer->doc .= '</div>'; 343 } else if ($mode == 'odt') { 344 345 // Loop through the instructions 346 foreach ($instr as $instruction) { 347 // Execute the callback against the Renderer 348 call_user_func_array(array($renderer, $instruction[0]), $instruction[1]); 349 } 350 } 351 } 352 353 /* 354 * Corrects relative internal links and media and 355 * converts headers of included pages to subheaders of the current page 356 */ 357 private function _convertInstructions($instr, $id, &$renderer, $depth = 1) { 358 global $ID; 359 global $conf; 360 361 $n = count($instr); 362 363 for ($i = 0; $i < $n; $i++) { 364 //internal links(links inside this wiki) an relative links 365 if ((substr($instr[$i][0], 0, 12) == 'internallink')) { 366 $this->_convert_link($renderer, $instr[$i], $id); 367 } 368 else if ((substr($instr[$i][0], 0, 13) == 'internalmedia')) { 369 $this->_convert_media($renderer, $instr[$i], $id); 370 } 371 else if ((substr($instr[$i][0], 0, 6) == 'header')) { 372 $this->_convert_header($renderer, $instr[$i], $depth-1); // -1 because the depth starts at 1 373 } 374 else if ((substr($instr[$i][0], 0, 12) == 'section_open')) { 375 $this->_convert_section($renderer, $instr[$i], $depth-1); // -1 because the depth starts at 1 376 } 377 } 378 379 //if its the document start, cut off the first element(document information) 380 if ($instr[0][0] == 'document_start') 381 return array_slice($instr, 1, -1); 382 else 383 return $instr; 384 } 385 386 /* 387 * Convert link of given instruction 388 */ 389 private function _convert_link(&$renderer, &$instr, $id) { 390 global $ID; 391 392 $exists = false; 393 394 resolve_pageid(getNS($id), $instr[1][0], $exists); 395 list($pageID, $pageReference) = explode("#", $instr[1][0], 2); 396 397 if (in_array($pageID, $this->includedPages)) { 398 // Crate new internal Links 399 $check = null; 400 401 // Either get existing reference or create from first heading. If still not there take the alternate ID 402 $pageNameLink = empty($pageReference) ? sectionID($pageID, $check) : $pageReference; 403 404 $instr[1][0] = $ID . "#" . $pageNameLink; 405 406 } else { 407 // Convert external Links to plain Text 408 409 $instr = array( 410 "cdata", 411 array($instr[1][1]), 412 $instr[2] 413 ); 414 } 415 } 416 417 /* 418 * Convert internalmedia of given instruction 419 */ 420 private function _convert_media(&$renderer, &$instr, $id) { 421 global $ID; 422 423 // Resolvemedia returns the absolute path to media by reference 424 $exists = false; 425 resolve_mediaid(getNS($id), $instr[1][0], $exists); 426 } 427 428 /** 429 * @param integer $depth 430 */ 431 private function _convert_header(&$renderer, &$instr, $depth) { 432 // More Depth! 433 $instr[1][1] += $depth; 434 } 435 436 /** 437 * @param integer $depth 438 */ 439 private function _convert_section(&$renderer, &$instr, $depth) { 440 // More Depth! 441 $instr[1][0] += $depth; 442 } 443 444 private function _mergeWithHeaders($existing, $newInstructions, $level = 1, $mergeHint = array()) { 445 446 $returnInstructions = array(); 447 $preparedInstructions = array(); 448 $existingStart = $existingEnd = 0; 449 $firstRun = true; 450 451 while ($this->_findNextHeaderSection($existing, $level, $existingStart, $existingEnd)) { 452 453 if ($firstRun) { 454 $returnInstructions = array_merge($returnInstructions, array_slice($existing, 0, $existingStart)); 455 $firstRun = false; 456 } 457 458 $currentSlice = array_slice($existing, $existingStart, $existingEnd-$existingStart); 459 460 // Find matching part with headername 461 $newStart = $newEnd = 0; 462 if ($this->_findNextHeaderSection($newInstructions, $level, $newStart, $newEnd, $currentSlice[0][1][0])) { 463 464 $newSlice = array_slice($newInstructions, $newStart, $newEnd-$newStart); 465 if ($newSlice[0][0] == 'header') 466 array_shift($newSlice); // Remove Heading 467 468 // merge found parts on next level. 469 $returnedInstructions = $this->_mergeWithHeaders($currentSlice, $newSlice, $level+1, $mergeHint); 470 471 // Put them at the end! 472 $preparedInstructions = array_merge($preparedInstructions, $returnedInstructions); 473 474 // Remove from input 475 array_splice($newInstructions, $newStart, $newEnd-$newStart); 476 } else { 477 // Nothing else found 478 $preparedInstructions = array_merge($preparedInstructions, $currentSlice); 479 } 480 481 $existingStart = $existingEnd; 482 } 483 484 // Append the rest 485 $returnInstructions = array_merge($returnInstructions, array_slice($existing, $existingStart)); 486 487 // Check for section close inconsistencies and put one at the very end ... 488 $section_postpend = array(); 489 if ( 490 ( 491 ($tmp1 = array_slice($newInstructions, -1)) 492 && ($tmp1[0][0] == 'section_close') 493 ) 494 && 495 ( 496 ($tmp2 = array_slice($newInstructions, -2)) 497 && ($tmp2[0][0] == 'section_close') 498 ) 499 ) { 500 $section_postpend = array_splice($newInstructions, -1); 501 } 502 if ( 503 ( 504 ($tmp3 = array_slice($returnInstructions, -1)) 505 && ($tmp3[0][0] == 'section_close') 506 ) 507 && 508 ( 509 ($tmp4 = array_slice($returnInstructions, -2)) 510 && ($tmp4[0][0] == 'section_close') 511 ) 512 ) { 513 $section_postpend = array_merge($section_postpend, array_splice($returnInstructions, -1)); 514 } 515 516 // What if there are headings left inside the $newInstructions????? 517 // Find matching part with headername 518 $newStart = $newEnd = 0; 519 $section_prepend = array(); 520 if ($this->_findNextHeaderSection($newInstructions, $level, $newStart, $newEnd)) { 521 // If there are header in here, build a prepend and have the rest at the end 522 $section_prepend = array_splice($newInstructions, 0, $newStart); 523 } else { 524 // If not, prepend all of it. 525 $section_prepend = $newInstructions; 526 $newInstructions = array(); 527 } 528 529 $this->_insertMergeHint($section_prepend, $mergeHint); 530 531 $returnInstructions = array_merge($returnInstructions, $section_prepend, $preparedInstructions, $newInstructions, $section_postpend); 532 533 return $returnInstructions; 534 } 535 536 /** 537 * @param integer $level 538 */ 539 private function _findNextHeaderSection($section, $level, &$start, &$end, $headerName = null) { 540 541 $inCount = count($section); 542 $currentSlice = -1; 543 544 // Find Level 1 Header that matches. 545 for ($i = $start; $i < $inCount; $i++) { 546 547 $instruction = $section[$i]; 548 $end = $i; // Or it will be lost and a section close will be missing. 549 550 // First Level Header 551 if ($instruction[0] == 'header' && $instruction[1][1] == $level) { 552 553 if ($currentSlice > 0) { 554 return true; 555 } 556 557 if ($headerName == null || ($headerName == $instruction[1][0])) { 558 // Begin of new slice ... 559 $start = $currentSlice = $i; 560 } 561 } 562 } 563 564 // Nothing found 565 $end = $i; // Or it will be lost and a section close will be missing. 566 return $currentSlice > 0; 567 } 568 569 private function _cleanAllInstructions(&$instr) { 570 $this->_cleanInstructions($instr, '/section_(close|open)/'); 571 $this->_cleanInstructions($instr, '/listu_(close|open)/'); 572 $this->_cleanInstructions($instr, '/listo_(close|open)/'); 573 } 574 575 /** 576 * @param string $tag 577 */ 578 private function _cleanInstructions(&$instructions, $tag) { 579 580 $inCount = count($instructions); 581 for ($i = 0; $i < $inCount; $i++) { 582 583 // Last instruction 584 if ($i == $inCount-1) { 585 break; 586 } 587 588 if (preg_match($tag, $instructions[$i][0]) && preg_match($tag, $instructions[$i+1][0]) && $instructions[$i][0] != $instructions[$i+1][0]) { 589 590 // found different tags, but both match the expression and follow each other - so they can be elliminated 591 array_splice($instructions, $i, 2); 592 $inCount -= 2; 593 $i--; 594 } 595 596 } 597 } 598 599 /** 600 * Strip everything except for the headers 601 */ 602 private function _initialHeaderStructure($instructions) { 603 $inCount = count($instructions); 604 for ($i = 0; $i < $inCount; $i++) { 605 606 // Last instruction 607 if ($i == $inCount-1) { 608 break; 609 } 610 611 if (!in_array($instructions[$i][0], array('header', 'section_open', 'section_close', 'p_open', 'p_close'))) { 612 // found non-matching 613 array_splice($instructions, $i, 1); 614 $inCount--; 615 $i--; 616 } 617 } 618 return $instructions; 619 } 620 621 private function _insertMergeHint(&$instructions, $mergeHint) { 622 623 // Surround new slice with a mergehint 624 if (empty($mergeHint)) { return; } 625 626 // No emtpy insruction sets. 627 $this->_cleanAllInstructions($instructions); 628 629 if (empty($instructions)) { return; } 630 631 $mergeHintPrepend = array(array( 632 'plugin', 633 array( 634 'siteexport_toctools', 635 array( 636 'mergehint', 637 'start', 638 $mergeHint, 639 $mergeHintId 640 ) 641 ) 642 )); 643 644 $mergeHintPostpend = array(array( 645 'plugin', 646 array( 647 'siteexport_toctools', 648 array( 649 'mergehint', 650 'end', 651 $mergeHint 652 ) 653 ) 654 )); 655 656 if ($instructions[0][0] == 'plugin' && $instructions[0][1][0] == 'siteexport_toctools' && $instructions[0][1][1][1] == 'start' ) { 657/* 658 print "<pre>"; print_r($instructions); print "</pre>"; 659//*/ 660 // This is already section merge hint ... but it will have a section at its end ... hopefully 661 $mergeHintPrependPrepend = array(); 662 do { 663 array_push( $mergeHintPrependPrepend, array_shift( $instructions ) ); 664 } while( $instructions[0][0] != 'section_open' ); 665 array_splice($mergeHintPrepend, 0, 0, $mergeHintPrependPrepend); 666 667 } 668 669 // only section content should be surrounded. 670 if ($instructions[0][0] != 'section_open') { return; } 671 672 // save for later use 673 $mergeHints = array(); 674 $mergeHintId = sectionid($mergeHint, $mergeHints); 675 $this->merghintIds[$mergeHintId] = $mergeHint; 676 677/* 678 print "<pre>"; print_r($instructions); print "</pre>"; 679//*/ 680 681 $instructions = array_merge($mergeHintPrepend, $instructions, $mergeHintPostpend); 682 } 683 684 /** 685 * Remove TOC, section edit buttons and tags 686 */ 687 private function _cleanXHTML($xhtml) { 688 $replace = array( 689 '!<div class="toc">.*?(</div>\n</div>)!s' => '', // remove TOCs 690 '#<!-- SECTION \[(\d*-\d*)\] -->#s' => '', // remove section edit buttons 691 '!<div id="tags">.*?(</div>)!s' => '' // remove category tags 692 ); 693 $xhtml = preg_replace(array_keys($replace), array_values($replace), $xhtml); 694 return $xhtml; 695 } 696 697 /** 698 * Allow the plugin to prevent DokuWiki creating a second instance of itself 699 * 700 * @return bool true if the plugin can not be instantiated more than once 701 */ 702 public function isSingleton() { 703 return true; 704 } 705} 706// vim:ts=4:sw=4:et:enc=utf-8: 707