Para 1
Para 2
Para 1
Para 2
' 21 | * echo "\n\n"; 22 | * 23 | * // set innerHTML 24 | * $elem->innerHTML = 'FiveFilters.org'; 25 | * echo $elem->innerHTML; // prints 'FiveFilters.org' 26 | * echo "\n\n"; 27 | * 28 | * // print document (with our changes) 29 | * echo $doc->saveXML(); 30 | * @endcode 31 | * 32 | * @author Keyvan Minoukadeh - http://www.keyvan.net - keyvan@keyvan.net 33 | * @see http://fivefilters.org (the project this was written for) 34 | */ 35 | class JSLikeHTMLElement extends DOMElement 36 | { 37 | /** 38 | * Used for setting innerHTML like it's done in JavaScript: 39 | * @code 40 | * $div->innerHTML = 'The story begins...
'; 41 | * @endcode 42 | */ 43 | public function __set($name, $value) { 44 | if ($name == 'innerHTML') { 45 | // first, empty the element 46 | for ($x=$this->childNodes->length-1; $x>=0; $x--) { 47 | $this->removeChild($this->childNodes->item($x)); 48 | } 49 | // $value holds our new inner HTML 50 | if ($value != '') { 51 | $f = $this->ownerDocument->createDocumentFragment(); 52 | // appendXML() expects well-formed markup (XHTML) 53 | $result = @$f->appendXML($value); // @ to suppress PHP warnings 54 | if ($result) { 55 | if ($f->hasChildNodes()) $this->appendChild($f); 56 | } else { 57 | // $value is probably ill-formed 58 | $f = new DOMDocument(); 59 | $value = mb_convert_encoding($value, 'HTML-ENTITIES', 'UTF-8'); 60 | // Using', $html); 118 | $html = preg_replace($this->regexps['replaceFonts'], '<$1span>', $html); 119 | $html = mb_convert_encoding($html, 'HTML-ENTITIES', "UTF-8"); 120 | if (trim($html) == '') $html = ''; 121 | if ($parser=='html5lib' && ($this->dom = HTML5_Parser::parse($html))) { 122 | // all good 123 | } else { 124 | $this->dom = new DOMDocument(); 125 | $this->dom->preserveWhiteSpace = false; 126 | @$this->dom->loadHTML($html); 127 | } 128 | $this->dom->registerNodeClass('DOMElement', 'JSLikeHTMLElement'); 129 | } 130 | 131 | /** 132 | * Get article title element 133 | * @return DOMElement 134 | */ 135 | public function getTitle() { 136 | return $this->articleTitle; 137 | } 138 | 139 | /** 140 | * Get article content element 141 | * @return DOMElement 142 | */ 143 | public function getContent() { 144 | return $this->articleContent; 145 | } 146 | 147 | /** 148 | * Runs readability. 149 | * 150 | * Workflow: 151 | * 1. Prep the document by removing script tags, css, etc. 152 | * 2. Build readability's DOM tree. 153 | * 3. Grab the article content from the current dom tree. 154 | * 4. Replace the current DOM tree with the new one. 155 | * 5. Read peacefully. 156 | * 157 | * @return boolean true if we found content, false otherwise 158 | **/ 159 | public function init() 160 | { 161 | if (!isset($this->dom->documentElement)) return false; 162 | $this->removeScripts($this->dom); 163 | //die($this->getInnerHTML($this->dom->documentElement)); 164 | 165 | // Assume successful outcome 166 | $this->success = true; 167 | 168 | $bodyElems = $this->dom->getElementsByTagName('body'); 169 | if ($bodyElems->length > 0) { 170 | if ($this->bodyCache == null) { 171 | $this->bodyCache = $bodyElems->item(0)->innerHTML; 172 | } 173 | if ($this->body == null) { 174 | $this->body = $bodyElems->item(0); 175 | } 176 | } 177 | 178 | $this->prepDocument(); 179 | 180 | //die($this->dom->documentElement->parentNode->nodeType); 181 | //$this->setInnerHTML($this->dom->documentElement, $this->getInnerHTML($this->dom->documentElement)); 182 | //die($this->getInnerHTML($this->dom->documentElement)); 183 | 184 | /* Build readability's DOM tree */ 185 | $overlay = $this->dom->createElement('div'); 186 | $innerDiv = $this->dom->createElement('div'); 187 | $articleTitle = $this->getArticleTitle(); 188 | $articleContent = $this->grabArticle(); 189 | 190 | if (!$articleContent) { 191 | $this->success = false; 192 | $articleContent = $this->dom->createElement('div'); 193 | $articleContent->setAttribute('id', 'readability-content'); 194 | $articleContent->innerHTML = '
Sorry, Readability was unable to parse this page for content.
'; 195 | } 196 | 197 | $overlay->setAttribute('id', 'readOverlay'); 198 | $innerDiv->setAttribute('id', 'readInner'); 199 | 200 | /* Glue the structure of our document together. */ 201 | $innerDiv->appendChild($articleTitle); 202 | $innerDiv->appendChild($articleContent); 203 | $overlay->appendChild($innerDiv); 204 | 205 | /* Clear the old HTML, insert the new content. */ 206 | $this->body->innerHTML = ''; 207 | $this->body->appendChild($overlay); 208 | //document.body.insertBefore(overlay, document.body.firstChild); 209 | $this->body->removeAttribute('style'); 210 | 211 | $this->postProcessContent($articleContent); 212 | 213 | // Set title and content instance variables 214 | $this->articleTitle = $articleTitle; 215 | $this->articleContent = $articleContent; 216 | 217 | return $this->success; 218 | } 219 | 220 | /** 221 | * Debug 222 | */ 223 | protected function dbg($msg) { 224 | if ($this->debug) echo '* ',$msg, "\n"; 225 | } 226 | 227 | /** 228 | * Run any post-process modifications to article content as necessary. 229 | * 230 | * @param DOMElement 231 | * @return void 232 | */ 233 | public function postProcessContent($articleContent) { 234 | if ($this->convertLinksToFootnotes && !preg_match('/wikipedia\.org/', @$this->url)) { 235 | $this->addFootnotes($articleContent); 236 | } 237 | } 238 | 239 | /** 240 | * Get the article title as an H1. 241 | * 242 | * @return DOMElement 243 | */ 244 | protected function getArticleTitle() { 245 | $curTitle = ''; 246 | $origTitle = ''; 247 | 248 | try { 249 | $curTitle = $origTitle = $this->getInnerText($this->dom->getElementsByTagName('title')->item(0)); 250 | } catch(Exception $e) {} 251 | 252 | if (preg_match('/ [\|\-] /', $curTitle)) 253 | { 254 | $curTitle = preg_replace('/(.*)[\|\-] .*/i', '$1', $origTitle); 255 | 256 | if (count(explode(' ', $curTitle)) < 3) { 257 | $curTitle = preg_replace('/[^\|\-]*[\|\-](.*)/i', '$1', $origTitle); 258 | } 259 | } 260 | else if (strpos($curTitle, ': ') !== false) 261 | { 262 | $curTitle = preg_replace('/.*:(.*)/i', '$1', $origTitle); 263 | 264 | if (count(explode(' ', $curTitle)) < 3) { 265 | $curTitle = preg_replace('/[^:]*[:](.*)/i','$1', $origTitle); 266 | } 267 | } 268 | else if(strlen($curTitle) > 150 || strlen($curTitle) < 15) 269 | { 270 | $hOnes = $this->dom->getElementsByTagName('h1'); 271 | if($hOnes->length == 1) 272 | { 273 | $curTitle = $this->getInnerText($hOnes->item(0)); 274 | } 275 | } 276 | 277 | $curTitle = trim($curTitle); 278 | 279 | if (count(explode(' ', $curTitle)) <= 4) { 280 | $curTitle = $origTitle; 281 | } 282 | 283 | $articleTitle = $this->dom->createElement('h1'); 284 | $articleTitle->innerHTML = $curTitle; 285 | 286 | return $articleTitle; 287 | } 288 | 289 | /** 290 | * Prepare the HTML document for readability to scrape it. 291 | * This includes things like stripping javascript, CSS, and handling terrible markup. 292 | * 293 | * @return void 294 | **/ 295 | protected function prepDocument() { 296 | /** 297 | * In some cases a body element can't be found (if the HTML is totally hosed for example) 298 | * so we create a new body node and append it to the document. 299 | */ 300 | if ($this->body == null) 301 | { 302 | $this->body = $this->dom->createElement('body'); 303 | $this->dom->documentElement->appendChild($this->body); 304 | } 305 | $this->body->setAttribute('id', 'readabilityBody'); 306 | 307 | /* Remove all style tags in head */ 308 | $styleTags = $this->dom->getElementsByTagName('style'); 309 | for ($i = $styleTags->length-1; $i >= 0; $i--) 310 | { 311 | $styleTags->item($i)->parentNode->removeChild($styleTags->item($i)); 312 | } 313 | 314 | /* Turn all double br's into p's */ 315 | /* Note, this is pretty costly as far as processing goes. Maybe optimize later. */ 316 | //document.body.innerHTML = document.body.innerHTML.replace(readability.regexps.replaceBrs, '').replace(readability.regexps.replaceFonts, '<$1span>'); 317 | // We do this in the constructor for PHP as that's when we have raw HTML - before parsing it into a DOM tree. 318 | // Manipulating innerHTML as it's done in JS is not possible in PHP. 319 | } 320 | 321 | /** 322 | * For easier reading, convert this document to have footnotes at the bottom rather than inline links. 323 | * @see http://www.roughtype.com/archives/2010/05/experiments_in.php 324 | * 325 | * @return void 326 | **/ 327 | public function addFootnotes($articleContent) { 328 | $footnotesWrapper = $this->dom->createElement('div'); 329 | $footnotesWrapper->setAttribute('id', 'readability-footnotes'); 330 | $footnotesWrapper->innerHTML = '
tags, etc.
411 | *
412 | * @param DOMElement
413 | * @return void
414 | */
415 | function prepArticle($articleContent) {
416 | $this->cleanStyles($articleContent);
417 | $this->killBreaks($articleContent);
418 | if ($this->revertForcedParagraphElements) {
419 | $this->revertReadabilityStyledElements($articleContent);
420 | }
421 |
422 | /* Clean out junk from the article content */
423 | $this->cleanConditionally($articleContent, 'form');
424 | $this->clean($articleContent, 'object');
425 | $this->clean($articleContent, 'h1');
426 |
427 | /**
428 | * If there is only one h2, they are probably using it
429 | * as a header and not a subheader, so remove it since we already have a header.
430 | ***/
431 | if (!$this->lightClean && ($articleContent->getElementsByTagName('h2')->length == 1)) {
432 | $this->clean($articleContent, 'h2');
433 | }
434 | $this->clean($articleContent, 'iframe');
435 |
436 | $this->cleanHeaders($articleContent);
437 |
438 | /* Do these last as the previous stuff may have removed junk that will affect these */
439 | $this->cleanConditionally($articleContent, 'table');
440 | $this->cleanConditionally($articleContent, 'ul');
441 | $this->cleanConditionally($articleContent, 'div');
442 |
443 | /* Remove extra paragraphs */
444 | $articleParagraphs = $articleContent->getElementsByTagName('p');
445 | for ($i = $articleParagraphs->length-1; $i >= 0; $i--)
446 | {
447 | $imgCount = $articleParagraphs->item($i)->getElementsByTagName('img')->length;
448 | $embedCount = $articleParagraphs->item($i)->getElementsByTagName('embed')->length;
449 | $objectCount = $articleParagraphs->item($i)->getElementsByTagName('object')->length;
450 | $iframeCount = $articleParagraphs->item($i)->getElementsByTagName('iframe')->length;
451 |
452 | if ($imgCount === 0 && $embedCount === 0 && $objectCount === 0 && $iframeCount === 0 && $this->getInnerText($articleParagraphs->item($i), false) == '')
453 | {
454 | $articleParagraphs->item($i)->parentNode->removeChild($articleParagraphs->item($i));
455 | }
456 | }
457 |
458 | try {
459 | $articleContent->innerHTML = preg_replace('/ innerHTML);
460 | //articleContent.innerHTML = articleContent.innerHTML.replace(/ dbg("Cleaning innerHTML of breaks failed. This is an IE strict-block-elements bug. Ignoring.: " . $e);
464 | }
465 | }
466 |
467 | /**
468 | * Initialize a node with the readability object. Also checks the
469 | * className/id for special names to add to its score.
470 | *
471 | * @param Element
472 | * @return void
473 | **/
474 | protected function initializeNode($node) {
475 | $readability = $this->dom->createAttribute('readability');
476 | $readability->value = 0; // this is our contentScore
477 | $node->setAttributeNode($readability);
478 |
479 | switch (strtoupper($node->tagName)) { // unsure if strtoupper is needed, but using it just in case
480 | case 'DIV':
481 | $readability->value += 5;
482 | break;
483 |
484 | case 'PRE':
485 | case 'TD':
486 | case 'BLOCKQUOTE':
487 | $readability->value += 3;
488 | break;
489 |
490 | case 'ADDRESS':
491 | case 'OL':
492 | case 'UL':
493 | case 'DL':
494 | case 'DD':
495 | case 'DT':
496 | case 'LI':
497 | case 'FORM':
498 | $readability->value -= 3;
499 | break;
500 |
501 | case 'H1':
502 | case 'H2':
503 | case 'H3':
504 | case 'H4':
505 | case 'H5':
506 | case 'H6':
507 | case 'TH':
508 | $readability->value -= 5;
509 | break;
510 | }
511 | $readability->value += $this->getClassWeight($node);
512 | }
513 |
514 | /***
515 | * grabArticle - Using a variety of metrics (content score, classname, element types), find the content that is
516 | * most likely to be the stuff a user wants to read. Then return it wrapped up in a div.
517 | *
518 | * @return DOMElement
519 | **/
520 | protected function grabArticle($page=null) {
521 | $stripUnlikelyCandidates = $this->flagIsActive(self::FLAG_STRIP_UNLIKELYS);
522 | if (!$page) $page = $this->dom;
523 | $allElements = $page->getElementsByTagName('*');
524 | /**
525 | * First, node prepping. Trash nodes that look cruddy (like ones with the class name "comment", etc), and turn divs
526 | * into P tags where they have been used inappropriately (as in, where they contain no other block level elements.)
527 | *
528 | * Note: Assignment from index for performance. See http://www.peachpit.com/articles/article.aspx?p=31567&seqNum=5
529 | * TODO: Shouldn't this be a reverse traversal?
530 | **/
531 | $node = null;
532 | $nodesToScore = array();
533 | for ($nodeIndex = 0; ($node = $allElements->item($nodeIndex)); $nodeIndex++) {
534 | //for ($nodeIndex=$targetList->length-1; $nodeIndex >= 0; $nodeIndex--) {
535 | //$node = $targetList->item($nodeIndex);
536 | $tagName = strtoupper($node->tagName);
537 | /* Remove unlikely candidates */
538 | if ($stripUnlikelyCandidates) {
539 | $unlikelyMatchString = $node->getAttribute('class') . $node->getAttribute('id');
540 | if (
541 | preg_match($this->regexps['unlikelyCandidates'], $unlikelyMatchString) &&
542 | !preg_match($this->regexps['okMaybeItsACandidate'], $unlikelyMatchString) &&
543 | $tagName != 'BODY'
544 | )
545 | {
546 | $this->dbg('Removing unlikely candidate - ' . $unlikelyMatchString);
547 | //$nodesToRemove[] = $node;
548 | $node->parentNode->removeChild($node);
549 | $nodeIndex--;
550 | continue;
551 | }
552 | }
553 |
554 | if ($tagName == 'P' || $tagName == 'TD' || $tagName == 'PRE') {
555 | $nodesToScore[] = $node;
556 | }
557 |
558 | /* Turn all divs that don't have children block level elements into p's */
559 | if ($tagName == 'DIV') {
560 | if (!preg_match($this->regexps['divToPElements'], $node->innerHTML)) {
561 | //$this->dbg('Altering div to p');
562 | $newNode = $this->dom->createElement('p');
563 | try {
564 | $newNode->innerHTML = $node->innerHTML;
565 | //$nodesToReplace[] = array('new'=>$newNode, 'old'=>$node);
566 | $node->parentNode->replaceChild($newNode, $node);
567 | $nodeIndex--;
568 | $nodesToScore[] = $node; // or $newNode?
569 | }
570 | catch(Exception $e) {
571 | $this->dbg('Could not alter div to p, reverting back to div.: ' . $e);
572 | }
573 | }
574 | else
575 | {
576 | /* EXPERIMENTAL */
577 | // TODO: change these p elements back to text nodes after processing
578 | for ($i = 0, $il = $node->childNodes->length; $i < $il; $i++) {
579 | $childNode = $node->childNodes->item($i);
580 | if ($childNode->nodeType == 3) { // XML_TEXT_NODE
581 | //$this->dbg('replacing text node with a p tag with the same content.');
582 | $p = $this->dom->createElement('p');
583 | $p->innerHTML = $childNode->nodeValue;
584 | $p->setAttribute('style', 'display: inline;');
585 | $p->setAttribute('class', 'readability-styled');
586 | $childNode->parentNode->replaceChild($p, $childNode);
587 | }
588 | }
589 | }
590 | }
591 | }
592 |
593 | /**
594 | * Loop through all paragraphs, and assign a score to them based on how content-y they look.
595 | * Then add their score to their parent node.
596 | *
597 | * A score is determined by things like number of commas, class names, etc. Maybe eventually link density.
598 | **/
599 | $candidates = array();
600 | for ($pt=0; $pt < count($nodesToScore); $pt++) {
601 | $parentNode = $nodesToScore[$pt]->parentNode;
602 | // $grandParentNode = $parentNode ? $parentNode->parentNode : null;
603 | $grandParentNode = !$parentNode ? null : (($parentNode->parentNode instanceof DOMElement) ? $parentNode->parentNode : null);
604 | $innerText = $this->getInnerText($nodesToScore[$pt]);
605 |
606 | if (!$parentNode || !isset($parentNode->tagName)) {
607 | continue;
608 | }
609 |
610 | /* If this paragraph is less than 25 characters, don't even count it. */
611 | if(strlen($innerText) < 25) {
612 | continue;
613 | }
614 |
615 | /* Initialize readability data for the parent. */
616 | if (!$parentNode->hasAttribute('readability'))
617 | {
618 | $this->initializeNode($parentNode);
619 | $candidates[] = $parentNode;
620 | }
621 |
622 | /* Initialize readability data for the grandparent. */
623 | if ($grandParentNode && !$grandParentNode->hasAttribute('readability') && isset($grandParentNode->tagName))
624 | {
625 | $this->initializeNode($grandParentNode);
626 | $candidates[] = $grandParentNode;
627 | }
628 |
629 | $contentScore = 0;
630 |
631 | /* Add a point for the paragraph itself as a base. */
632 | $contentScore++;
633 |
634 | /* Add points for any commas within this paragraph */
635 | $contentScore += count(explode(',', $innerText));
636 |
637 | /* For every 100 characters in this paragraph, add another point. Up to 3 points. */
638 | $contentScore += min(floor(strlen($innerText) / 100), 3);
639 |
640 | /* Add the score to the parent. The grandparent gets half. */
641 | $parentNode->getAttributeNode('readability')->value += $contentScore;
642 |
643 | if ($grandParentNode) {
644 | $grandParentNode->getAttributeNode('readability')->value += $contentScore/2;
645 | }
646 | }
647 |
648 | /**
649 | * After we've calculated scores, loop through all of the possible candidate nodes we found
650 | * and find the one with the highest score.
651 | **/
652 | $topCandidate = null;
653 | for ($c=0, $cl=count($candidates); $c < $cl; $c++)
654 | {
655 | /**
656 | * Scale the final candidates score based on link density. Good content should have a
657 | * relatively small link density (5% or less) and be mostly unaffected by this operation.
658 | **/
659 | $readability = $candidates[$c]->getAttributeNode('readability');
660 | $readability->value = $readability->value * (1-$this->getLinkDensity($candidates[$c]));
661 |
662 | $this->dbg('Candidate: ' . $candidates[$c]->tagName . ' (' . $candidates[$c]->getAttribute('class') . ':' . $candidates[$c]->getAttribute('id') . ') with score ' . $readability->value);
663 |
664 | if (!$topCandidate || $readability->value > (int)$topCandidate->getAttribute('readability')) {
665 | $topCandidate = $candidates[$c];
666 | }
667 | }
668 |
669 | /**
670 | * If we still have no top candidate, just use the body as a last resort.
671 | * We also have to copy the body node so it is something we can modify.
672 | **/
673 | if ($topCandidate === null || strtoupper($topCandidate->tagName) == 'BODY')
674 | {
675 | $topCandidate = $this->dom->createElement('div');
676 | if ($page instanceof DOMDocument) {
677 | if (!isset($page->documentElement)) {
678 | // we don't have a body either? what a mess! :)
679 | } else {
680 | $topCandidate->innerHTML = $page->documentElement->innerHTML;
681 | $page->documentElement->innerHTML = '';
682 | $page->documentElement->appendChild($topCandidate);
683 | }
684 | } else {
685 | $topCandidate->innerHTML = $page->innerHTML;
686 | $page->innerHTML = '';
687 | $page->appendChild($topCandidate);
688 | }
689 | $this->initializeNode($topCandidate);
690 | }
691 |
692 | /**
693 | * Now that we have the top candidate, look through its siblings for content that might also be related.
694 | * Things like preambles, content split by ads that we removed, etc.
695 | **/
696 | $articleContent = $this->dom->createElement('div');
697 | $articleContent->setAttribute('id', 'readability-content');
698 | $siblingScoreThreshold = max(10, ((int)$topCandidate->getAttribute('readability')) * 0.2);
699 | $siblingNodes = $topCandidate->parentNode->childNodes;
700 | if (!isset($siblingNodes)) {
701 | $siblingNodes = new stdClass;
702 | $siblingNodes->length = 0;
703 | }
704 |
705 | for ($s=0, $sl=$siblingNodes->length; $s < $sl; $s++)
706 | {
707 | $siblingNode = $siblingNodes->item($s);
708 | $append = false;
709 |
710 | $this->dbg('Looking at sibling node: ' . $siblingNode->nodeName . (($siblingNode->nodeType === XML_ELEMENT_NODE && $siblingNode->hasAttribute('readability')) ? (' with score ' . $siblingNode->getAttribute('readability')) : ''));
711 |
712 | //dbg('Sibling has score ' . ($siblingNode->readability ? siblingNode.readability.contentScore : 'Unknown'));
713 |
714 | if ($siblingNode === $topCandidate)
715 | // or if ($siblingNode->isSameNode($topCandidate))
716 | {
717 | $append = true;
718 | }
719 |
720 | $contentBonus = 0;
721 | /* Give a bonus if sibling nodes and top candidates have the example same classname */
722 | if ($siblingNode->nodeType === XML_ELEMENT_NODE && $siblingNode->getAttribute('class') == $topCandidate->getAttribute('class') && $topCandidate->getAttribute('class') != '') {
723 | $contentBonus += ((int)$topCandidate->getAttribute('readability')) * 0.2;
724 | }
725 |
726 | if ($siblingNode->nodeType === XML_ELEMENT_NODE && $siblingNode->hasAttribute('readability') && (((int)$siblingNode->getAttribute('readability')) + $contentBonus) >= $siblingScoreThreshold)
727 | {
728 | $append = true;
729 | }
730 |
731 | if (strtoupper($siblingNode->nodeName) == 'P') {
732 | $linkDensity = $this->getLinkDensity($siblingNode);
733 | $nodeContent = $this->getInnerText($siblingNode);
734 | $nodeLength = strlen($nodeContent);
735 |
736 | if ($nodeLength > 80 && $linkDensity < 0.25)
737 | {
738 | $append = true;
739 | }
740 | else if ($nodeLength < 80 && $linkDensity === 0 && preg_match('/\.( |$)/', $nodeContent))
741 | {
742 | $append = true;
743 | }
744 | }
745 |
746 | if ($append)
747 | {
748 | $this->dbg('Appending node: ' . $siblingNode->nodeName);
749 |
750 | $nodeToAppend = null;
751 | $sibNodeName = strtoupper($siblingNode->nodeName);
752 | if ($sibNodeName != 'DIV' && $sibNodeName != 'P') {
753 | /* We have a node that isn't a common block level element, like a form or td tag. Turn it into a div so it doesn't get filtered out later by accident. */
754 |
755 | $this->dbg('Altering siblingNode of ' . $sibNodeName . ' to div.');
756 | $nodeToAppend = $this->dom->createElement('div');
757 | try {
758 | $nodeToAppend->setAttribute('id', $siblingNode->getAttribute('id'));
759 | $nodeToAppend->innerHTML = $siblingNode->innerHTML;
760 | }
761 | catch(Exception $e)
762 | {
763 | $this->dbg('Could not alter siblingNode to div, reverting back to original.');
764 | $nodeToAppend = $siblingNode;
765 | $s--;
766 | $sl--;
767 | }
768 | } else {
769 | $nodeToAppend = $siblingNode;
770 | $s--;
771 | $sl--;
772 | }
773 |
774 | /* To ensure a node does not interfere with readability styles, remove its classnames */
775 | $nodeToAppend->removeAttribute('class');
776 |
777 | /* Append sibling and subtract from our list because it removes the node when you append to another node */
778 | $articleContent->appendChild($nodeToAppend);
779 | }
780 | }
781 |
782 | /**
783 | * So we have all of the content that we need. Now we clean it up for presentation.
784 | **/
785 | $this->prepArticle($articleContent);
786 |
787 | /**
788 | * Now that we've gone through the full algorithm, check to see if we got any meaningful content.
789 | * If we didn't, we may need to re-run grabArticle with different flags set. This gives us a higher
790 | * likelihood of finding the content, and the sieve approach gives us a higher likelihood of
791 | * finding the -right- content.
792 | **/
793 | if (strlen($this->getInnerText($articleContent, false)) < 250)
794 | {
795 | // TODO: find out why element disappears sometimes, e.g. for this URL http://www.businessinsider.com/6-hedge-fund-etfs-for-average-investors-2011-7
796 | // in the meantime, we check and create an empty element if it's not there.
797 | if (!isset($this->body->childNodes)) $this->body = $this->dom->createElement('body');
798 | $this->body->innerHTML = $this->bodyCache;
799 |
800 | if ($this->flagIsActive(self::FLAG_STRIP_UNLIKELYS)) {
801 | $this->removeFlag(self::FLAG_STRIP_UNLIKELYS);
802 | return $this->grabArticle($this->body);
803 | }
804 | else if ($this->flagIsActive(self::FLAG_WEIGHT_CLASSES)) {
805 | $this->removeFlag(self::FLAG_WEIGHT_CLASSES);
806 | return $this->grabArticle($this->body);
807 | }
808 | else if ($this->flagIsActive(self::FLAG_CLEAN_CONDITIONALLY)) {
809 | $this->removeFlag(self::FLAG_CLEAN_CONDITIONALLY);
810 | return $this->grabArticle($this->body);
811 | }
812 | else {
813 | return false;
814 | }
815 | }
816 | return $articleContent;
817 | }
818 |
819 | /**
820 | * Remove script tags from document
821 | *
822 | * @param DOMElement
823 | * @return void
824 | */
825 | public function removeScripts($doc) {
826 | $scripts = $doc->getElementsByTagName('script');
827 | for($i = $scripts->length-1; $i >= 0; $i--)
828 | {
829 | $scripts->item($i)->parentNode->removeChild($scripts->item($i));
830 | }
831 | }
832 |
833 | /**
834 | * Get the inner text of a node.
835 | * This also strips out any excess whitespace to be found.
836 | *
837 | * @param DOMElement $
838 | * @param boolean $normalizeSpaces (default: true)
839 | * @return string
840 | **/
841 | public function getInnerText($e, $normalizeSpaces=true) {
842 | $textContent = '';
843 |
844 | if (!isset($e->textContent) || $e->textContent == '') {
845 | return '';
846 | }
847 |
848 | $textContent = trim($e->textContent);
849 |
850 | if ($normalizeSpaces) {
851 | return preg_replace($this->regexps['normalize'], ' ', $textContent);
852 | } else {
853 | return $textContent;
854 | }
855 | }
856 |
857 | /**
858 | * Get the number of times a string $s appears in the node $e.
859 | *
860 | * @param DOMElement $e
861 | * @param string - what to count. Default is ","
862 | * @return number (integer)
863 | **/
864 | public function getCharCount($e, $s=',') {
865 | return substr_count($this->getInnerText($e), $s);
866 | }
867 |
868 | /**
869 | * Remove the style attribute on every $e and under.
870 | *
871 | * @param DOMElement $e
872 | * @return void
873 | */
874 | public function cleanStyles($e) {
875 | if (!is_object($e)) return;
876 | $elems = $e->getElementsByTagName('*');
877 | foreach ($elems as $elem) {
878 | $elem->removeAttribute('style');
879 | }
880 | }
881 |
882 | /**
883 | * Get the density of links as a percentage of the content
884 | * This is the amount of text that is inside a link divided by the total text in the node.
885 | *
886 | * @param DOMElement $e
887 | * @return number (float)
888 | */
889 | public function getLinkDensity($e) {
890 | $links = $e->getElementsByTagName('a');
891 | $textLength = strlen($this->getInnerText($e));
892 | $linkLength = 0;
893 | for ($i=0, $il=$links->length; $i < $il; $i++)
894 | {
895 | $linkLength += strlen($this->getInnerText($links->item($i)));
896 | }
897 | if ($textLength > 0) {
898 | return $linkLength / $textLength;
899 | } else {
900 | return 0;
901 | }
902 | }
903 |
904 | /**
905 | * Get an elements class/id weight. Uses regular expressions to tell if this
906 | * element looks good or bad.
907 | *
908 | * @param DOMElement $e
909 | * @return number (Integer)
910 | */
911 | public function getClassWeight($e) {
912 | if(!$this->flagIsActive(self::FLAG_WEIGHT_CLASSES)) {
913 | return 0;
914 | }
915 |
916 | $weight = 0;
917 |
918 | /* Look for a special classname */
919 | if ($e->hasAttribute('class') && $e->getAttribute('class') != '')
920 | {
921 | if (preg_match($this->regexps['negative'], $e->getAttribute('class'))) {
922 | $weight -= 25;
923 | }
924 | if (preg_match($this->regexps['positive'], $e->getAttribute('class'))) {
925 | $weight += 25;
926 | }
927 | }
928 |
929 | /* Look for a special ID */
930 | if ($e->hasAttribute('id') && $e->getAttribute('id') != '')
931 | {
932 | if (preg_match($this->regexps['negative'], $e->getAttribute('id'))) {
933 | $weight -= 25;
934 | }
935 | if (preg_match($this->regexps['positive'], $e->getAttribute('id'))) {
936 | $weight += 25;
937 | }
938 | }
939 | return $weight;
940 | }
941 |
942 | /**
943 | * Remove extraneous break tags from a node.
944 | *
945 | * @param DOMElement $node
946 | * @return void
947 | */
948 | public function killBreaks($node) {
949 | $html = $node->innerHTML;
950 | $html = preg_replace($this->regexps['killBreaks'], ' Para 1 Para 2 Para 1 Para 2
20 | Last update: {{lastupdate}}
21 |
20 | Last update: {{lastupdate}}
21 |
', $html);
951 | $node->innerHTML = $html;
952 | }
953 |
954 | /**
955 | * Clean a node of all elements of type "tag".
956 | * (Unless it's a youtube/vimeo video. People love movies.)
957 | *
958 | * Updated 2012-09-18 to preserve youtube/vimeo iframes
959 | *
960 | * @param DOMElement $e
961 | * @param string $tag
962 | * @return void
963 | */
964 | public function clean($e, $tag) {
965 | $targetList = $e->getElementsByTagName($tag);
966 | $isEmbed = ($tag == 'iframe' || $tag == 'object' || $tag == 'embed');
967 |
968 | for ($y=$targetList->length-1; $y >= 0; $y--) {
969 | /* Allow youtube and vimeo videos through as people usually want to see those. */
970 | if ($isEmbed) {
971 | $attributeValues = '';
972 | for ($i=0, $il=$targetList->item($y)->attributes->length; $i < $il; $i++) {
973 | $attributeValues .= $targetList->item($y)->attributes->item($i)->value . '|'; // DOMAttr? (TODO: test)
974 | }
975 |
976 | /* First, check the elements attributes to see if any of them contain youtube or vimeo */
977 | if (preg_match($this->regexps['video'], $attributeValues)) {
978 | continue;
979 | }
980 |
981 | /* Then check the elements inside this element for the same. */
982 | if (preg_match($this->regexps['video'], $targetList->item($y)->innerHTML)) {
983 | continue;
984 | }
985 | }
986 | $targetList->item($y)->parentNode->removeChild($targetList->item($y));
987 | }
988 | }
989 |
990 | /**
991 | * Clean an element of all tags of type "tag" if they look fishy.
992 | * "Fishy" is an algorithm based on content length, classnames,
993 | * link density, number of images & embeds, etc.
994 | *
995 | * @param DOMElement $e
996 | * @param string $tag
997 | * @return void
998 | */
999 | public function cleanConditionally($e, $tag) {
1000 | if (!$this->flagIsActive(self::FLAG_CLEAN_CONDITIONALLY)) {
1001 | return;
1002 | }
1003 |
1004 | $tagsList = $e->getElementsByTagName($tag);
1005 | $curTagsLength = $tagsList->length;
1006 |
1007 | /**
1008 | * Gather counts for other typical elements embedded within.
1009 | * Traverse backwards so we can remove nodes at the same time without effecting the traversal.
1010 | *
1011 | * TODO: Consider taking into account original contentScore here.
1012 | */
1013 | for ($i=$curTagsLength-1; $i >= 0; $i--) {
1014 | $weight = $this->getClassWeight($tagsList->item($i));
1015 | $contentScore = ($tagsList->item($i)->hasAttribute('readability')) ? (int)$tagsList->item($i)->getAttribute('readability') : 0;
1016 |
1017 | $this->dbg('Cleaning Conditionally ' . $tagsList->item($i)->tagName . ' (' . $tagsList->item($i)->getAttribute('class') . ':' . $tagsList->item($i)->getAttribute('id') . ')' . (($tagsList->item($i)->hasAttribute('readability')) ? (' with score ' . $tagsList->item($i)->getAttribute('readability')) : ''));
1018 |
1019 | if ($weight + $contentScore < 0) {
1020 | $tagsList->item($i)->parentNode->removeChild($tagsList->item($i));
1021 | }
1022 | else if ( $this->getCharCount($tagsList->item($i), ',') < 10) {
1023 | /**
1024 | * If there are not very many commas, and the number of
1025 | * non-paragraph elements is more than paragraphs or other ominous signs, remove the element.
1026 | **/
1027 | $p = $tagsList->item($i)->getElementsByTagName('p')->length;
1028 | $img = $tagsList->item($i)->getElementsByTagName('img')->length;
1029 | $li = $tagsList->item($i)->getElementsByTagName('li')->length-100;
1030 | $input = $tagsList->item($i)->getElementsByTagName('input')->length;
1031 | $a = $tagsList->item($i)->getElementsByTagName('a')->length;
1032 |
1033 | $embedCount = 0;
1034 | $embeds = $tagsList->item($i)->getElementsByTagName('embed');
1035 | for ($ei=0, $il=$embeds->length; $ei < $il; $ei++) {
1036 | if (preg_match($this->regexps['video'], $embeds->item($ei)->getAttribute('src'))) {
1037 | $embedCount++;
1038 | }
1039 | }
1040 | $embeds = $tagsList->item($i)->getElementsByTagName('iframe');
1041 | for ($ei=0, $il=$embeds->length; $ei < $il; $ei++) {
1042 | if (preg_match($this->regexps['video'], $embeds->item($ei)->getAttribute('src'))) {
1043 | $embedCount++;
1044 | }
1045 | }
1046 |
1047 | $linkDensity = $this->getLinkDensity($tagsList->item($i));
1048 | $contentLength = strlen($this->getInnerText($tagsList->item($i)));
1049 | $toRemove = false;
1050 |
1051 | if ($this->lightClean) {
1052 | $this->dbg('Light clean...');
1053 | if ( ($img > $p) && ($img > 4) ) {
1054 | $this->dbg(' more than 4 images and more image elements than paragraph elements');
1055 | $toRemove = true;
1056 | } else if ($li > $p && $tag != 'ul' && $tag != 'ol') {
1057 | $this->dbg(' too many or
');
1058 | $toRemove = true;
1059 | } else if ( $input > floor($p/3) ) {
1060 | $this->dbg(' too many elements');
1061 | $toRemove = true;
1062 | } else if ($contentLength < 25 && ($embedCount === 0 && ($img === 0 || $img > 2))) {
1063 | $this->dbg(' content length less than 25 chars, 0 embeds and either 0 images or more than 2 images');
1064 | $toRemove = true;
1065 | } else if($weight < 25 && $linkDensity > 0.2) {
1066 | $this->dbg(' weight smaller than 25 and link density above 0.2');
1067 | $toRemove = true;
1068 | } else if($a > 2 && ($weight >= 25 && $linkDensity > 0.5)) {
1069 | $this->dbg(' more than 2 links and weight above 25 but link density greater than 0.5');
1070 | $toRemove = true;
1071 | } else if($embedCount > 3) {
1072 | $this->dbg(' more than 3 embeds');
1073 | $toRemove = true;
1074 | }
1075 | } else {
1076 | $this->dbg('Standard clean...');
1077 | if ( $img > $p ) {
1078 | $this->dbg(' more image elements than paragraph elements');
1079 | $toRemove = true;
1080 | } else if ($li > $p && $tag != 'ul' && $tag != 'ol') {
1081 | $this->dbg(' too many
or
');
1082 | $toRemove = true;
1083 | } else if ( $input > floor($p/3) ) {
1084 | $this->dbg(' too many elements');
1085 | $toRemove = true;
1086 | } else if ($contentLength < 25 && ($img === 0 || $img > 2) ) {
1087 | $this->dbg(' content length less than 25 chars and 0 images, or more than 2 images');
1088 | $toRemove = true;
1089 | } else if($weight < 25 && $linkDensity > 0.2) {
1090 | $this->dbg(' weight smaller than 25 and link density above 0.2');
1091 | $toRemove = true;
1092 | } else if($weight >= 25 && $linkDensity > 0.5) {
1093 | $this->dbg(' weight above 25 but link density greater than 0.5');
1094 | $toRemove = true;
1095 | } else if(($embedCount == 1 && $contentLength < 75) || $embedCount > 1) {
1096 | $this->dbg(' 1 embed and content length smaller than 75 chars, or more than one embed');
1097 | $toRemove = true;
1098 | }
1099 | }
1100 |
1101 | if ($toRemove) {
1102 | //$this->dbg('Removing: '.$tagsList->item($i)->innerHTML);
1103 | $tagsList->item($i)->parentNode->removeChild($tagsList->item($i));
1104 | }
1105 | }
1106 | }
1107 | }
1108 |
1109 | /**
1110 | * Clean out spurious headers from an Element. Checks things like classnames and link density.
1111 | *
1112 | * @param DOMElement $e
1113 | * @return void
1114 | */
1115 | public function cleanHeaders($e) {
1116 | for ($headerIndex = 1; $headerIndex < 3; $headerIndex++) {
1117 | $headers = $e->getElementsByTagName('h' . $headerIndex);
1118 | for ($i=$headers->length-1; $i >=0; $i--) {
1119 | if ($this->getClassWeight($headers->item($i)) < 0 || $this->getLinkDensity($headers->item($i)) > 0.33) {
1120 | $headers->item($i)->parentNode->removeChild($headers->item($i));
1121 | }
1122 | }
1123 | }
1124 | }
1125 |
1126 | public function flagIsActive($flag) {
1127 | return ($this->flags & $flag) > 0;
1128 | }
1129 |
1130 | public function addFlag($flag) {
1131 | $this->flags = $this->flags | $flag;
1132 | }
1133 |
1134 | public function removeFlag($flag) {
1135 | $this->flags = $this->flags & ~$flag;
1136 | }
1137 | }
--------------------------------------------------------------------------------
/lib/fivefilters-php-readability/composer.json:
--------------------------------------------------------------------------------
1 | {
2 | "name": "fivefilters/php-readability",
3 | "type": "library",
4 | "description": "Automatic article extraction from HTML",
5 | "keywords": ["article extraction","content extraction","extraction","article","content","html"],
6 | "homepage": "http://code.fivefilters.org/php-readability/",
7 | "license": "Apache-2.0",
8 | "authors": [
9 | {
10 | "name": "Keyvan Minoukadeh",
11 | "email": "keyvan@keyvan.net",
12 | "homepage": "http://keyvan.net",
13 | "role": "Developer (ported original JS code to PHP)"
14 | },
15 | {
16 | "name": "Arc90",
17 | "homepage": "http://arc90.com",
18 | "role": "Developer (original JS version)"
19 | }
20 | ],
21 | "require": {
22 | "php": ">=5.2"
23 | },
24 | "autoload": {
25 | "psr-0": { "Readability": "" }
26 | }
27 | }
--------------------------------------------------------------------------------
/lib/fivefilters-php-readability/examples/JSLikeHTMLElement.php:
--------------------------------------------------------------------------------
1 | registerNodeClass('DOMElement', 'JSLikeHTMLElement');
6 | $doc->loadHTML('
');
7 | $elem = $doc->getElementsByTagName('div')->item(0);
8 |
9 | // print innerHTML
10 | echo $elem->innerHTML; // prints '" . htmlentities($strHtml, ENT_QUOTES, "UTF-8") . "
";
47 | }
48 |
49 | $out = $strHtml;
50 |
51 | //check common site
52 | $commonSiteExtraction = Parser::extractFromCommonSite($url, $strHtml);
53 | if ($commonSiteExtraction !== false) {
54 | $out = $commonSiteExtraction;
55 | }
56 |
57 | // Purify
58 | $preConfig = \HTMLPurifier_Config::createDefault();
59 | $preConfig->set('HTML.TidyLevel', 'heavy');
60 | $preConfig->set('HTML.ForbiddenElements', array('style', 'script', 'link'));
61 | $preConfig->set('URI.Base', $url);
62 | $preConfig->set('URI.DefaultScheme', 'https');
63 | $preConfig->set('URI.MakeAbsolute', true);
64 | $prePurifier = new \HTMLPurifier($preConfig);
65 | $out = $prePurifier->purify($out);
66 |
67 | $readability = new \Readability($out, $url);
68 | $result = $readability->init();
69 | if ($result) {
70 | error_log("Extracting: readability success");
71 | $out = $readability->getContent()->innerHTML;
72 | } else {
73 | // error_log("Extracting: readability error");
74 | error_log(Color::set("Extracting: readability error", "red"));
75 | $out = '';
76 | }
77 |
78 | // Reformat bad HTML from Readability
79 | $postConfig = \HTMLPurifier_Config::createDefault();
80 | $postConfig->set('HTML.TidyLevel', 'none');
81 | $postPurifier = new \HTMLPurifier($postConfig);
82 | $out = $postPurifier->purify($out);
83 |
84 | return $out;
85 | }
86 |
87 | public static function extractImage($strHtml)
88 | {
89 | $reader = new Opengraph\Reader();
90 | $image = "";
91 | try {
92 | $reader->parse($strHtml);
93 | $og = $reader->getArrayCopy();
94 | if (array_key_exists('og:image', $og) && is_array($og['og:image']) && count($og['og:image']) > 0) {
95 | $image = $og['og:image'][0]["og:image:url"];
96 | }
97 | } catch (\RuntimeException $e) {
98 | error_log($e->getMessage());
99 | }
100 | return $image;
101 | }
102 | }
103 |
--------------------------------------------------------------------------------
/src/Uploader.php:
--------------------------------------------------------------------------------
1 | 'latest',
14 | 'region' => 'eu-west-1',
15 | ]);
16 |
17 | $files = [
18 | [
19 | "Bucket" => getenv('AWS_BUCKET'),
20 | "Body" => file_get_contents(__DIR__ . "/../www/index.html"),
21 | 'Key' => "index.html",
22 | "ContentType" => "text/html",
23 | "ACL" => "public-read",
24 | ],
25 | [
26 | "Bucket" => getenv('AWS_BUCKET'),
27 | "Body" => file_get_contents(__DIR__ . "/../www/latest.html"),
28 | "Key" => "latest.html",
29 | "ContentType" => "text/html",
30 | "ACL" => "public-read",
31 | ],
32 | [
33 | "Bucket" => getenv('AWS_BUCKET'),
34 | "Body" => file_get_contents(__DIR__ . "/../www/cache.manifest"),
35 | "Key" => "cache.manifest",
36 | "ContentType" => "text/cache-manifest",
37 | "ACL" => "public-read",
38 | ],
39 | ];
40 |
41 | foreach ($files as $object) {
42 | $s3->putObject($object);
43 | }
44 | }
45 | }
46 |
--------------------------------------------------------------------------------
/src/templates/fullhn.manifest.mustache:
--------------------------------------------------------------------------------
1 | CACHE MANIFEST
2 |
3 | # {{ version }}
4 |
5 | CACHE:
6 | {{ cachedfiles }}
7 |
8 | NETWORK:
9 | latest.html
10 | *
11 |
--------------------------------------------------------------------------------
/src/templates/index.mustache:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
15 | Reload to view new articles
16 |
17 | Full {{ title }}
19 | 
24 |
61 |
73 |
74 |
79 |
80 |
81 |
82 |
83 |
84 |
87 |
88 |
89 |
--------------------------------------------------------------------------------
/src/templates/latest.mustache:
--------------------------------------------------------------------------------
1 |
2 |
3 |
4 |
5 |
25 | {{# articles }}
26 |
42 |
43 | {{# articles }}
44 |
35 | {{ title }}
36 | {{ domain }}
37 |
38 |
39 |
46 |
47 | {{ title }}
48 |
49 |
50 |
55 |
56 | {{{ parsed }}}
57 |
58 |
15 | Reload to view new articles
16 |
17 | Full {{ title }}
19 |
24 |
48 |
60 |
61 |
66 |
67 |
68 |
69 |
70 |
71 |
74 |
75 |
76 |
--------------------------------------------------------------------------------
/src/templates/partials/head.mustache:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mauricesvay/FullHackerNews/73117bd1f81ddb163048e5828ee6da1edabd157e/src/templates/partials/head.mustache
--------------------------------------------------------------------------------
/www/apple-touch-icon.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mauricesvay/FullHackerNews/73117bd1f81ddb163048e5828ee6da1edabd157e/www/apple-touch-icon.png
--------------------------------------------------------------------------------
/www/css/img/loader.gif:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mauricesvay/FullHackerNews/73117bd1f81ddb163048e5828ee6da1edabd157e/www/css/img/loader.gif
--------------------------------------------------------------------------------
/www/css/style.css:
--------------------------------------------------------------------------------
1 | /*! normalize.css v2.1.2 | MIT License | git.io/normalize */
2 | article,
3 | aside,
4 | details,
5 | figcaption,
6 | figure,
7 | footer,
8 | header,
9 | hgroup,
10 | main,
11 | nav,
12 | section,
13 | summary {
14 | display: block;
15 | }
16 | audio,
17 | canvas,
18 | video {
19 | display: inline-block;
20 | }
21 | audio:not([controls]) {
22 | display: none;
23 | height: 0;
24 | }
25 | [hidden] {
26 | display: none;
27 | }
28 | html {
29 | font-family: sans-serif;
30 | -ms-text-size-adjust: 100%;
31 | -webkit-text-size-adjust: 100%;
32 | }
33 | body {
34 | margin: 0;
35 | }
36 | a:focus {
37 | outline: thin dotted;
38 | }
39 | a:active,
40 | a:hover {
41 | outline: 0;
42 | }
43 | h1 {
44 | font-size: 2em;
45 | margin: 0.67em 0;
46 | }
47 | abbr[title] {
48 | border-bottom: 1px dotted;
49 | }
50 | b,
51 | strong {
52 | font-weight: bold;
53 | }
54 | dfn {
55 | font-style: italic;
56 | }
57 | hr {
58 | -moz-box-sizing: content-box;
59 | box-sizing: content-box;
60 | height: 0;
61 | }
62 | mark {
63 | background: #ff0;
64 | color: #000;
65 | }
66 | code,
67 | kbd,
68 | pre,
69 | samp {
70 | font-family: monospace, serif;
71 | font-size: 1em;
72 | }
73 | pre {
74 | white-space: pre-wrap;
75 | }
76 | q {
77 | quotes: "\201C" "\201D" "\2018" "\2019";
78 | }
79 | small {
80 | font-size: 80%;
81 | }
82 | sub,
83 | sup {
84 | font-size: 75%;
85 | line-height: 0;
86 | position: relative;
87 | vertical-align: baseline;
88 | }
89 | sup {
90 | top: -0.5em;
91 | }
92 | sub {
93 | bottom: -0.25em;
94 | }
95 | img {
96 | border: 0;
97 | }
98 | svg:not(:root) {
99 | overflow: hidden;
100 | }
101 | figure {
102 | margin: 0;
103 | }
104 | fieldset {
105 | border: 1px solid silver;
106 | margin: 0 2px;
107 | padding: 0.35em 0.625em 0.75em;
108 | }
109 | legend {
110 | border: 0;
111 | padding: 0;
112 | }
113 | button,
114 | input,
115 | select,
116 | textarea {
117 | font-family: inherit;
118 | font-size: 100%;
119 | margin: 0;
120 | }
121 | button,
122 | input {
123 | line-height: normal;
124 | }
125 | button,
126 | select {
127 | text-transform: none;
128 | }
129 | button,
130 | html input[type="button"],
131 | input[type="reset"],
132 | input[type="submit"] {
133 | -webkit-appearance: button;
134 | cursor: pointer;
135 | }
136 | button[disabled],
137 | html input[disabled] {
138 | cursor: default;
139 | }
140 | input[type="checkbox"],
141 | input[type="radio"] {
142 | box-sizing: border-box;
143 | padding: 0;
144 | }
145 | input[type="search"] {
146 | -webkit-appearance: textfield;
147 | -moz-box-sizing: content-box;
148 | -webkit-box-sizing: content-box;
149 | box-sizing: content-box;
150 | }
151 | input[type="search"]::-webkit-search-cancel-button,
152 | input[type="search"]::-webkit-search-decoration {
153 | -webkit-appearance: none;
154 | }
155 | button::-moz-focus-inner,
156 | input::-moz-focus-inner {
157 | border: 0;
158 | padding: 0;
159 | }
160 | textarea {
161 | overflow: auto;
162 | vertical-align: top;
163 | }
164 | table {
165 | border-collapse: collapse;
166 | border-spacing: 0;
167 | }
168 |
169 | /**
170 | * General
171 | ******************************************************************************/
172 | body {
173 | margin: 0;
174 | font-family: Avenir, sans-serif;
175 | word-wrap: break-word;
176 | line-height: 1.5;
177 | background: #fff;
178 | color: #2c3e50;
179 | border-top: 6px solid #f60;
180 | }
181 |
182 | header,
183 | .main {
184 | max-width: 45em;
185 | margin: 0 auto;
186 | padding: 1em;
187 | }
188 |
189 | /**
190 | * Header
191 | ******************************************************************************/
192 | header {
193 | overflow: hidden;
194 | }
195 | header h1 {
196 | float: left;
197 | font-size: 1.5em;
198 | }
199 | header p {
200 | float: right;
201 | margin-top: 1.25em;
202 | color: #999;
203 | }
204 |
205 | a {
206 | color: #2980b9;
207 | }
208 | a:visited {
209 | color: #8e44ad;
210 | }
211 | #reload,
212 | .loader {
213 | display: none;
214 | }
215 | #reload a {
216 | display: block;
217 | text-decoration: none;
218 | background: #f60;
219 | color: #fff;
220 | text-align: center;
221 | padding: 0.5em;
222 | }
223 | #reload a:hover {
224 | background: #f40;
225 | }
226 |
227 | /**
228 | * Table of contents
229 | ******************************************************************************/
230 | .toc {
231 | margin: 0 0 1em 0;
232 | padding: 0;
233 | }
234 | .toc li {
235 | padding: 0.5em 0;
236 | list-style: none;
237 | border-top: 1px solid #eee;
238 | }
239 | .toc a {
240 | color: #2980b9;
241 | text-decoration: none;
242 | display: flex;
243 | }
244 |
245 | .toc .preview {
246 | flex: 0 0 48px;
247 | width: 48px;
248 | height: 48px;
249 | background-size: cover;
250 | background-position: center center;
251 | background-color: #fff;
252 | align-items: center;
253 | border-radius: 2px;
254 | }
255 |
256 | .toc .info {
257 | flex: 1 1 auto;
258 | margin-left: 16px;
259 | }
260 |
261 | .toc .title {
262 | display: block;
263 | line-height: 24px;
264 | }
265 |
266 | .toc .site {
267 | line-height: 24px;
268 | text-decoration: none;
269 | color: #999;
270 | font-size: small;
271 | }
272 |
273 | /**
274 | * Articles
275 | ******************************************************************************/
276 | .hn-article {
277 | border-top: 3px solid #ecf0f1;
278 | padding: 1em 0;
279 | }
280 | .title {
281 | line-height: 1.2em;
282 | }
283 | .title a {
284 | color: #2c3e50;
285 | text-decoration: none;
286 | font-size: 1em;
287 | }
288 |
289 | .meta {
290 | font-size: 0.8em;
291 | margin: 1em 0;
292 | }
293 | .meta a {
294 | color: #7f8c8d;
295 | }
296 | blockquote {
297 | margin: 1em 0 1em 0;
298 | padding-left: 1em;
299 | border-left: 3px solid #eee;
300 | }
301 |
302 | table {
303 | border: 1px solid #eee;
304 | width: 100%;
305 | }
306 | tr {
307 | border-top: 1px solid #eee;
308 | }
309 | td {
310 | padding: 1ex;
311 | }
312 |
313 | img {
314 | max-width: 100%;
315 | height: auto;
316 | }
317 |
318 | pre {
319 | border: 1px solid #ccc;
320 | padding: 1em;
321 | background: #f3f3f3;
322 | }
323 |
324 | iframe {
325 | max-width: 100%;
326 | }
327 |
328 | hr {
329 | border: 0;
330 | border-top: 1px solid #eee;
331 | }
332 |
333 | /**
334 | * Navigation
335 | ******************************************************************************/
336 | .toolbar {
337 | position: fixed;
338 | bottom: 0;
339 | left: 0;
340 | right: 0;
341 | background: rgba(0, 0, 0, 0.2);
342 | color: #fff;
343 | text-align: center;
344 | padding: 0.4em 0.4em 1em 0.4em;
345 | font-family: Avenir;
346 | }
347 | .toolbar a {
348 | text-transform: uppercase;
349 | color: #fff;
350 | display: inline-block;
351 | padding: 0.3em 1em;
352 | text-decoration: none;
353 | background: #000;
354 | border-radius: 24px;
355 | }
356 | .toolbar a:active {
357 | -o-transform: translateY(1px);
358 | -ms-transform: translateY(1px);
359 | -moz-transform: translateY(1px);
360 | -webkit-transform: translateY(1px);
361 | transform: translateY(1px);
362 | }
363 |
364 | /**
365 | * Footer
366 | ******************************************************************************/
367 | footer {
368 | background: #eee;
369 | padding: 1em 1em 3em 1em;
370 | margin: 0;
371 | }
372 | footer div {
373 | max-width: 45em;
374 | margin: 0 auto;
375 | }
376 |
377 | /**
378 | * Responsive
379 | ******************************************************************************/
380 | @media only screen and (max-width: 767px) and (orientation: portrait) {
381 | header h1,
382 | header p {
383 | float: none;
384 | text-align: center;
385 | margin: 0.5em 0;
386 | }
387 | header h1 {
388 | margin-bottom: 0;
389 | }
390 | pre {
391 | font-size: 0.9em;
392 | }
393 |
394 | h1 {
395 | font-size: 1.25em;
396 | }
397 | h2 {
398 | font-size: 1.15em;
399 | }
400 | h3 {
401 | font-size: 1em;
402 | }
403 | }
404 |
--------------------------------------------------------------------------------
/www/favicon.ico:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mauricesvay/FullHackerNews/73117bd1f81ddb163048e5828ee6da1edabd157e/www/favicon.ico
--------------------------------------------------------------------------------
/www/js/app.js:
--------------------------------------------------------------------------------
1 | window.App = {
2 |
3 | current : 0,
4 | nbArticles : 0,
5 |
6 | init: function() {
7 | App.nbArticles = $('.hn-article').length;
8 | this.makeLastUpdateReadable();
9 | this.buildNavigation();
10 | this.checkCache();
11 | },
12 |
13 | checkCache : function() {
14 | var appCache = window.applicationCache;
15 | appCache.addEventListener('cached', App.handleCacheEvent, false);
16 | appCache.addEventListener('checking', App.handleCacheEvent, false);
17 | appCache.addEventListener('downloading', App.handleCacheEvent, false);
18 | appCache.addEventListener('error', App.handleCacheError, false);
19 | appCache.addEventListener('noupdate', App.handleCacheEvent, false);
20 | appCache.addEventListener('obsolete', App.handleCacheEvent, false);
21 | appCache.addEventListener('progress', App.handleCacheEvent, false);
22 | appCache.addEventListener('updateready', App.handleCacheEvent, false);
23 | },
24 |
25 | handleCacheError: function() {
26 | $('.loader').hide();
27 | },
28 | handleCacheEvent: function() {
29 | switch (window.applicationCache.status) {
30 | case window.applicationCache.CHECKING:
31 | case window.applicationCache.DOWNLOADING:
32 | $('.loader').show();
33 | break;
34 | case window.applicationCache.UPDATEREADY:
35 | $('.loader').hide();
36 | $('#reload').show();
37 | break;
38 | default :
39 | $('.loader').hide();
40 | }
41 | },
42 |
43 | scrollToIndex : function(idx) {
44 | var selector = '.hn-article[data-index='+idx+']';
45 | var focusSelector = selector + ' > .title a';
46 | if ($('.hn-article[data-index='+idx+']').length) {
47 | var scroll = $(selector).position().top;
48 | $(window).scrollTop(scroll);
49 | $(focusSelector).focus();
50 | setTimeout(function(){
51 | App.current = idx;
52 | }, 100);
53 | }
54 | },
55 |
56 | makeLastUpdateReadable: function() {
57 | var $lastUpdate = $('#lastupdate');
58 | var lastupdate = moment($lastUpdate.attr("title"));
59 | $lastUpdate.html(lastupdate.fromNow());
60 | },
61 |
62 | buildNavigation: function() {
63 | $('.toc a').on('click', function(e){
64 | e.preventDefault();
65 | var index = $(this).attr('href').replace('#article-','');
66 | App.scrollToIndex(index);
67 | });
68 | $('.hn-article').waypoint(function() {
69 | App.current = parseInt($(this).attr('data-index'),10);
70 | });
71 | $('#next').on('click', function(e){
72 | e.preventDefault();
73 | var next = Math.min(App.nbArticles, App.current + 1);
74 | App.scrollToIndex(next);
75 | });
76 | $('#prev').on('click', function(e){
77 | e.preventDefault();
78 | var prev = Math.max(0, (App.current - 1));
79 | App.scrollToIndex(prev);
80 | });
81 | $('#index').on('click', function(e) {
82 | e.preventDefault();
83 | $(window).scrollTop(0);
84 | $('#toc a').first().focus();
85 | });
86 | }
87 | };
--------------------------------------------------------------------------------
/www/js/moment.min.js:
--------------------------------------------------------------------------------
1 | // moment.js
2 | // version : 2.0.0
3 | // author : Tim Wood
4 | // license : MIT
5 | // momentjs.com
6 | (function(e){function O(e,t){return function(n){return j(e.call(this,n),t)}}function M(e){return function(t){return this.lang().ordinal(e.call(this,t))}}function _(){}function D(e){H(this,e)}function P(e){var t=this._data={},n=e.years||e.year||e.y||0,r=e.months||e.month||e.M||0,i=e.weeks||e.week||e.w||0,s=e.days||e.day||e.d||0,o=e.hours||e.hour||e.h||0,u=e.minutes||e.minute||e.m||0,a=e.seconds||e.second||e.s||0,f=e.milliseconds||e.millisecond||e.ms||0;this._milliseconds=f+a*1e3+u*6e4+o*36e5,this._days=s+i*7,this._months=r+n*12,t.milliseconds=f%1e3,a+=B(f/1e3),t.seconds=a%60,u+=B(a/60),t.minutes=u%60,o+=B(u/60),t.hours=o%24,s+=B(o/24),s+=i*7,t.days=s%30,r+=B(s/30),t.months=r%12,n+=B(r/12),t.years=n}function H(e,t){for(var n in t)t.hasOwnProperty(n)&&(e[n]=t[n]);return e}function B(e){return e<0?Math.ceil(e):Math.floor(e)}function j(e,t){var n=e+"";while(n.length
25 | {{# articles }}
26 |
29 |
30 | {{# articles }}
31 |
33 |
34 | {{ title }}
35 |
36 |
37 |
42 |
43 | {{{ content }}}
44 |
45 |