Merge pull request #14 from j0k3r/revert-bc

Revert BC changes

.scrutinizer.yml

@@ -1,3 +1,2 @@
 tools:
-    external_code_coverage:
+    external_code_coverage: false
-        timeout: 600

.travis.yml

@@ -6,29 +6,37 @@ php:
     - 5.4
     - 5.5
     - 5.6
-    - nightly
+    - 7.0
     - hhvm
 
-# run build against nightly but allow them to fail
 matrix:
     fast_finish: true
     allow_failures:
-        - php: nightly
         - php: hhvm
 
 # faster builds on new travis setup not using sudo
 sudo: false
 
+# cache vendor dirs
+cache:
+    directories:
+        - vendor
+        - $HOME/.composer/cache
+
 install:
     - composer self-update
 
 before_script:
+    - if [ -n "$GH_TOKEN" ]; then composer config github-oauth.github.com ${GH_TOKEN}; fi;
+    # disable TLS for composer because openssl is disabled for PHP 5.3.3 on travis
+    # see: https://blog.travis-ci.com/upcoming_ubuntu_11_10_migration/
+    - if [[ $TRAVIS_PHP_VERSION = 5.3.3 ]]; then composer config -g -- disable-tls true; fi;
+    - if [[ $TRAVIS_PHP_VERSION = 5.3.3 ]]; then composer config -g -- secure-http false; fi;
     - composer install --prefer-dist --no-interaction
 
 script:
-    - phpunit --coverage-clover=coverage.clover
+    - mkdir -p build/logs
+    - phpunit -v --coverage-clover build/logs/clover.xml
 
 after_script:
-    - |
+    - php vendor/bin/coveralls -v
-        wget https://scrutinizer-ci.com/ocular.phar
-        php ocular.phar code-coverage:upload --format=php-clover coverage.clover

README.md

@@ -1,7 +1,7 @@
 # Readability
 
 [![Build Status](https://travis-ci.org/j0k3r/php-readability.svg?branch=master)](https://travis-ci.org/j0k3r/php-readability)
-[![Code Coverage](https://scrutinizer-ci.com/g/j0k3r/php-readability/badges/coverage.png?b=master)](https://scrutinizer-ci.com/g/j0k3r/php-readability/?branch=master)
+[![Coverage Status](https://coveralls.io/repos/j0k3r/php-readability/badge.svg?branch=master&service=github)](https://coveralls.io/github/j0k3r/php-readability/?branch=master)
 
 This is an extract of the Readability class from this [full-text-rss](https://github.com/Dither/full-text-rss) fork. It can be defined as a better version of the original [php-readability](https://bitbucket.org/fivefilters/php-readability/overview).
 
@@ -45,3 +45,20 @@ if ($result) {
     echo 'Looks like we couldn\'t find the content. :(';
 }
 ```
+
+If you want to debug it, or check what's going on, you can inject a logger (which must follow `Psr\Log\LoggerInterface`, Monolog for example):
+
+```php
+use Readability\Readability;
+use Monolog\Logger;
+use Monolog\Handler\StreamHandler;
+
+$url = 'http://www.medialens.org/index.php/alerts/alert-archive/alerts-2013/729-thatcher.html';
+$html = file_get_contents($url);
+
+$logger = new Logger('readability');
+$logger->pushHandler(new StreamHandler('path/to/your.log', Logger::DEBUG));
+
+$readability = new Readability($html, $url);
+$readability->setLogger($logger);
+```

composer.json

@@ -24,7 +24,11 @@
         "role": "Developer (original JS version)"
     }],
     "require": {
-        "php": ">=5.3.3"
+        "php": ">=5.3.3",
+        "monolog/monolog": "^1.13.1"
+    },
+    "require-dev": {
+        "satooshi/php-coveralls": "~0.6"
     },
     "autoload": {
         "psr-4": { "Readability\\": "src/" }

phpunit.xml.dist

@@ -26,7 +26,7 @@
         </whitelist>
     </filter>
 
-    <logging>
+    <!-- <logging>
         <log type="coverage-html" target="coverage" title="Readability" charset="UTF-8" yui="true" highlight="true" lowUpperBound="35" highLowerBound="70"/>
-    </logging>
+    </logging> -->
 </phpunit>

src/JSLikeHTMLElement.php

@@ -56,7 +56,8 @@ class JSLikeHTMLElement extends \DOMElement
                 $f = $this->ownerDocument->createDocumentFragment();
 
                 // appendXML() expects well-formed markup (XHTML)
-                $result = @$f->appendXML($value); // @ to suppress PHP warnings
+                // @ to suppress PHP warnings
+                $result = @$f->appendXML($value);
                 if ($result) {
                     if ($f->hasChildNodes()) {
                         $this->appendChild($f);
@@ -75,6 +76,7 @@ class JSLikeHTMLElement extends \DOMElement
 
                     if ($result) {
                         $import = $f->getElementsByTagName('htmlfragment')->item(0);
+
                         foreach ($import->childNodes as $child) {
                             $importedNode = $this->ownerDocument->importNode($child, true);
                             $this->appendChild($importedNode);
@@ -102,6 +104,7 @@ class JSLikeHTMLElement extends \DOMElement
     {
         if ($name == 'innerHTML') {
             $inner = '';
+
             foreach ($this->childNodes as $child) {
                 $inner .= $this->ownerDocument->saveXML($child);
             }

src/Readability.php

@@ -2,6 +2,10 @@
 
 namespace Readability;
 
+use Psr\Log\LoggerAwareInterface;
+use Psr\Log\LoggerInterface;
+use Psr\Log\NullLogger;
+
 /**
  * Arc90's Readability ported to PHP for FiveFilters.org
  * Based on readability.js version 1.7.1 (without multi-page support)
@@ -45,7 +49,7 @@ namespace Readability;
  * existing DOMElement objects without passing an entire HTML document to
  * be parsed.
  */
-class Readability
+class Readability implements LoggerAwareInterface
 {
     public $convertLinksToFootnotes = false;
     public $revertForcedParagraphElements = true;
@@ -53,23 +57,33 @@ class Readability
     public $articleContent;
     public $original_html;
     public $dom;
-    public $url = null; // optional - URL where HTML was retrieved
+    // optional - URL where HTML was retrieved
-    public $lightClean = true; // preserves more content (experimental)
+    public $url = null;
+    // preserves more content (experimental)
+    public $lightClean = true;
+    // no more used, keept to avoid BC
     public $debug = false;
     public $tidied = false;
-    protected $debugText = ''; // error text for one time output
+    // article domain regexp for calibration
-    protected $domainRegExp = null; // article domain regexp for calibration
+    protected $domainRegExp = null;
     protected $body = null; //
-    protected $bodyCache = null; // Cache the body HTML in case we need to re-use it later
+    // Cache the body HTML in case we need to re-use it later
-    protected $flags = 7; // 1 | 2 | 4;   // Start with all processing flags set.
+    protected $bodyCache = null;
-    protected $success = false; // indicates whether we were able to extract or not
+    // 1 | 2 | 4;   // Start with all processing flags set.
+    protected $flags = 7;
+    // indicates whether we were able to extract or not
+    protected $success = false;
+    protected $logger;
+    protected $parser;
+    protected $html;
+    protected $useTidy;
 
     /**
      * All of the regular expressions in use within readability.
      * Defined up here so we don't instantiate them repeatedly in loops.
      */
     public $regexps = array(
-        'unlikelyCandidates' => '/display\s*:\s*none|ignore|\binfo|annoy|clock|date|time|author|intro|links|hidd?e|about|archive|\bprint|bookmark|tags|share|search|social|robot|published|combx|comment|mast(?:head)|subscri|community|category|disqus|extra|head|head(?:er|note)|floor|foot(?:er|note)|menu|tool|function|nav|remark|rss|shoutbox|tool|widget|meta|banner|sponsor|adsense|inner-?ad|ad-|sponsor|\badv\b|\bads\b|agr?egate?|pager|sidebar|popup|tweet|twitter/i',
+        'unlikelyCandidates' => '/display\s*:\s*none|ignore|\binfo|annoy|clock|date|time|author|intro|links|hidd?e|about|archive|\bprint|bookmark|tags|tag-list|share|search|social|robot|published|combx|comment|mast(?:head)|subscri|community|category|disqus|extra|head|head(?:er|note)|floor|foot(?:er|note)|menu|tool|function|nav|remark|rss|shoutbox|tool|widget|meta|banner|sponsor|adsense|inner-?ad|ad-|sponsor|\badv\b|\bads\b|agr?egate?|pager|sidebar|popup|tweet|twitter/i',
         'okMaybeItsACandidate' => '/article\b|contain|\bcontent|column|general|detail|shadow|lightbox|blog|body|entry|main|page/i',
         'positive' => '/read|full|article|body|\bcontent|contain|entry|main|markdown|page|attach|pagination|post|text|blog|story/i',
         'negative' => '/bottom|stat|info|discuss|e[\-]?mail|comment|reply|log.{2}(n|ed)|sign|single|combx|com-|contact|_nav|link|media|\bout|promo|\bad-|related|scroll|shoutbox|sidebar|sponsor|shopping|teaser|recommend/i',
@@ -105,21 +119,33 @@ class Readability
     );
     // raw HTML filters
     protected $pre_filters = array(
-        '!<script[^>]*>(.*?)</script>!is' => '', // remove obvious scripts
+        // remove obvious scripts
-        '!<style[^>]*>(.*?)</style>!is' => '', // remove obvious styles
+        '!<script[^>]*>(.*?)</script>!is' => '',
-        '!</?span[^>]*>!is' => '', // remove spans as we redefine styles and they're probably special-styled
+        // remove obvious styles
-        '!<font[^>]*>\s*\[AD\]\s*</font>!is' => '', // HACK: firewall-filtered content
+        '!<style[^>]*>(.*?)</style>!is' => '',
-        '!(<br[^>]*>[ \r\n\s]*){2,}!i' => '</p><p>', // HACK: replace linebreaks plus br's with p's
+        // remove spans as we redefine styles and they're probably special-styled
-        //'!</?noscript>!is' => '', // replace noscripts
+        '!</?span[^>]*>!is' => '',
-        '!<(/?)font[^>]*>!is' => '<\\1span>', // replace fonts to spans
+        // HACK: firewall-filtered content
+        '!<font[^>]*>\s*\[AD\]\s*</font>!is' => '',
+        // HACK: replace linebreaks plus br's with p's
+        '!(<br[^>]*>[ \r\n\s]*){2,}!i' => '</p><p>',
+        // replace noscripts
+        //'!</?noscript>!is' => '',
+        // replace fonts to spans
+        '!<(/?)font[^>]*>!is' => '<\\1span>',
     );
     // output HTML filters
     protected $post_filters = array(
-        '/<br\s*\/?>\s*<p/i' => '<p', // replace excessive br's
+        // replace excessive br's
-        '!<(?:a|div|p)[^>]+/>!is' => '', // replace empty tags that break layouts
+        '/<br\s*\/?>\s*<p/i' => '<p',
-        //'!<(\s*/?\s*(?:blockquote|br|hr|code|div|article|span|footer|aside|p|pre|dl|li|ul|ol)) [^>]+>!is' => "<\\1>", // remove all attributes on text tags
+        // replace empty tags that break layouts
-        "/\n+/" => "\n", //single newlines cleanup
+        '!<(?:a|div|p)[^>]+/>!is' => '',
-        '!<pre[^>]*>\s*<code!is' => '<pre', // modern web...
+        // remove all attributes on text tags
+        //'!<(\s*/?\s*(?:blockquote|br|hr|code|div|article|span|footer|aside|p|pre|dl|li|ul|ol)) [^>]+>!is' => "<\\1>",
+        //single newlines cleanup
+        "/\n+/" => "\n",
+        // modern web...
+        '!<pre[^>]*>\s*<code!is' => '<pre',
         '!</code>\s*</pre>!is' => '</pre>',
         '!<[hb]r>!is' => '<\\1 />',
     );
@@ -151,70 +177,23 @@ class Readability
     public function __construct($html, $url = null, $parser = 'libxml', $use_tidy = true)
     {
         $this->url = $url;
-        $this->debugText = 'Parsing URL: '.$url."\n";
+        $this->html = $html;
+        $this->parser = $parser;
+        $this->useTidy = $use_tidy && function_exists('tidy_parse_string');
 
-        if ($url) {
+        $this->logger = new NullLogger();
-            $this->domainRegExp = '/'.strtr(preg_replace('/www\d*\./', '', parse_url($url, PHP_URL_HOST)), array('.' => '\.')).'/';
+        $this->loadHtml();
     }
 
-        mb_internal_encoding('UTF-8');
+    public function setLogger(LoggerInterface $logger)
-        mb_http_output('UTF-8');
+    {
-        mb_regex_encoding('UTF-8');
+        $this->logger = $logger;
-
-        // HACK: dirty cleanup to replace some stuff; shouldn't use regexps with HTML but well...
-        if (!$this->flagIsActive(self::FLAG_DISABLE_PREFILTER)) {
-            foreach ($this->pre_filters as $search => $replace) {
-                $html = preg_replace($search, $replace, $html);
-            }
-            unset($search, $replace);
-        }
-
-        if (trim($html) === '') {
-            $html = '<html></html>';
-        }
-
-        /*
-         * Use tidy (if it exists).
-         * This fixes problems with some sites which would otherwise trouble DOMDocument's HTML parsing.
-         * Although sometimes it makes matters worse, which is why there is an option to disable it.
-         *
-         */
-        if ($use_tidy && function_exists('tidy_parse_string')) {
-            $this->debugText .= 'Tidying document'."\n";
-            $tidy = tidy_parse_string($html, $this->tidy_config, 'UTF8');
-            if (tidy_clean_repair($tidy)) {
-                $this->original_html = $html;
-                $this->tidied = true;
-                $html = $tidy->value;
-                $html = preg_replace('/[\r\n]+/is', "\n", $html);
-            }
-            unset($tidy);
-        }
-
-        $html = mb_convert_encoding($html, 'HTML-ENTITIES', 'UTF-8');
-
-        if (!($parser == 'html5lib' && ($this->dom = \HTML5_Parser::parse($html)))) {
-            libxml_use_internal_errors(true);
-
-            $this->dom = new \DOMDocument();
-            $this->dom->preserveWhiteSpace = false;
-
-            if (PHP_VERSION_ID >= 50400) {
-                $this->dom->loadHTML($html, LIBXML_NOBLANKS | LIBXML_COMPACT | LIBXML_NOERROR);
-            } else {
-                $this->dom->loadHTML($html);
-            }
-
-            libxml_use_internal_errors(false);
-        }
-
-        $this->dom->registerNodeClass('DOMElement', 'Readability\JSLikeHTMLElement');
     }
 
     /**
      * Get article title element.
      *
-     * @return DOMElement
+     * @return \DOMElement
      */
     public function getTitle()
     {
@@ -224,7 +203,7 @@ class Readability
     /**
      * Get article content element.
      *
-     * @return DOMElement
+     * @return \DOMElement
      */
     public function getContent()
     {
@@ -253,6 +232,76 @@ class Readability
         $this->post_filters[$filter] = $replacer;
     }
 
+    /**
+     * Load HTML in a DOMDocument.
+     * Apply Pre filters
+     * Cleanup HTML using Tidy (or not).
+     *
+     * @todo This should be called in init() instead of from __construct
+     */
+    private function loadHtml()
+    {
+        $this->original_html = $this->html;
+
+        $this->logger->debug('Parsing URL: '.$this->url);
+
+        if ($this->url) {
+            $this->domainRegExp = '/'.strtr(preg_replace('/www\d*\./', '', parse_url($this->url, PHP_URL_HOST)), array('.' => '\.')).'/';
+        }
+
+        mb_internal_encoding('UTF-8');
+        mb_http_output('UTF-8');
+        mb_regex_encoding('UTF-8');
+
+        // HACK: dirty cleanup to replace some stuff; shouldn't use regexps with HTML but well...
+        if (!$this->flagIsActive(self::FLAG_DISABLE_PREFILTER)) {
+            foreach ($this->pre_filters as $search => $replace) {
+                $this->html = preg_replace($search, $replace, $this->html);
+            }
+            unset($search, $replace);
+        }
+
+        if (trim($this->html) === '') {
+            $this->html = '<html></html>';
+        }
+
+        /*
+         * Use tidy (if it exists).
+         * This fixes problems with some sites which would otherwise trouble DOMDocument's HTML parsing.
+         * Although sometimes it makes matters worse, which is why there is an option to disable it.
+         */
+        if ($this->useTidy) {
+            $this->logger->debug('Tidying document');
+
+            $tidy = tidy_parse_string($this->html, $this->tidy_config, 'UTF8');
+            if (tidy_clean_repair($tidy)) {
+                $this->tidied = true;
+                $this->html = $tidy->value;
+                $this->html = preg_replace('/[\r\n]+/is', "\n", $this->html);
+            }
+            unset($tidy);
+        }
+
+        $this->html = mb_convert_encoding($this->html, 'HTML-ENTITIES', 'UTF-8');
+
+        if (!($this->parser == 'html5lib' && ($this->dom = \HTML5_Parser::parse($this->html)))) {
+            libxml_use_internal_errors(true);
+
+            $this->dom = new \DOMDocument();
+            $this->dom->preserveWhiteSpace = false;
+
+            if (PHP_VERSION_ID >= 50400) {
+                $this->dom->loadHTML($this->html, LIBXML_NOBLANKS | LIBXML_COMPACT | LIBXML_NOERROR);
+            } else {
+                $this->dom->loadHTML($this->html);
+            }
+
+            libxml_use_internal_errors(false);
+        }
+
+        $this->dom->registerNodeClass('DOMElement', 'Readability\JSLikeHTMLElement');
+    }
+
     /**
      * Runs readability.
      *
@@ -283,7 +332,7 @@ class Readability
             }
         }
 
-        if ($bodyElems->length > 0 && $this->body == null) {
+        if ($bodyElems->length > 0 && $this->body === null) {
             $this->body = $bodyElems->item(0);
         }
 
@@ -319,40 +368,37 @@ class Readability
         // Set title and content instance variables.
         $this->articleTitle = $articleTitle;
         $this->articleContent = $articleContent;
-        $this->dump_dbg();
 
         return $this->success;
     }
 
     /**
      * Debug.
+     *
+     * @deprecated use $this->logger->debug() instead
      */
-    protected function dbg($msg) //, $error=false)
+    protected function dbg($msg)
     {
-        if ($this->debug) {
+        $this->logger->debug($msg);
-            $this->debugText .= $msg."\n";
-        }
     }
 
     /**
      * Dump debug info.
+     *
+     * @deprecated since Monolog gather log, we don't need it
      */
     protected function dump_dbg()
     {
-        if ($this->debug) {
-            openlog('Readability PHP ', LOG_PID | LOG_PERROR, 0);
-            syslog(6, $this->debugText); // 1 - error 6 - info
-        }
     }
 
     /**
      * Run any post-process modifications to article content as necessary.
      *
-     * @param DOMElement
+     * @param \DOMElement $articleContent
      */
     public function postProcessContent($articleContent)
     {
-        if ($this->convertLinksToFootnotes && !preg_match('/\bwiki/', @$this->url)) {
+        if ($this->convertLinksToFootnotes && !preg_match('/\bwiki/', $this->url)) {
             $this->addFootnotes($articleContent);
         }
     }
@@ -360,16 +406,15 @@ class Readability
     /**
      * Get the article title as an H1.
      *
-     * @return DOMElement
+     * @return \DOMElement
      */
     protected function getArticleTitle()
     {
-        $curTitle = '';
-        $origTitle = '';
-
         try {
             $curTitle = $origTitle = $this->getInnerText($this->dom->getElementsByTagName('title')->item(0));
-        } catch (Exception $e) {
+        } catch (\Exception $e) {
+            $curTitle = '';
+            $origTitle = '';
         }
 
         if (preg_match('/ [\|\-] /', $curTitle)) {
@@ -410,7 +455,7 @@ class Readability
          * In some cases a body element can't be found (if the HTML is totally hosed for example)
          * so we create a new body node and append it to the document.
          */
-        if ($this->body == null) {
+        if ($this->body === null) {
             $this->body = $this->dom->createElement('body');
             $this->dom->documentElement->appendChild($this->body);
         }
@@ -433,6 +478,8 @@ class Readability
      * For easier reading, convert this document to have footnotes at the bottom rather than inline links.
      *
      * @see http://www.roughtype.com/archives/2010/05/experiments_in.php
+     *
+     * @param \DOMElement $articleContent
      */
     public function addFootnotes($articleContent)
     {
@@ -496,18 +543,15 @@ class Readability
      * Prepare the article node for display. Clean out any inline styles,
      * iframes, forms, strip extraneous <p> tags, etc.
      *
-     * @param DOMElement
+     * @param \DOMElement $articleContent
      */
     public function prepArticle($articleContent)
     {
-        if ($this->lightClean) {
+        $this->logger->debug($this->lightClean ? 'Light clean enabled.' : 'Standard clean enabled.');
-            $this->dbg('Light clean enabled.');
-        } else {
-            $this->dbg('Standard clean enabled.');
-        }
 
         $this->cleanStyles($articleContent);
         $this->killBreaks($articleContent);
+
         $xpath = new \DOMXPath($articleContent->ownerDocument);
 
         if ($this->revertForcedParagraphElements) {
@@ -552,30 +596,31 @@ class Readability
         $this->cleanConditionally($articleContent, 'form');
         $this->cleanConditionally($articleContent, 'table');
         $this->cleanConditionally($articleContent, 'ul');
-        //if (!$this->lightClean)
         $this->cleanConditionally($articleContent, 'div');
 
         // Remove extra paragraphs.
         $articleParagraphs = $articleContent->getElementsByTagName('p');
 
         for ($i = $articleParagraphs->length - 1; $i >= 0; --$i) {
-            $imgCount = $articleParagraphs->item($i)->getElementsByTagName('img')->length;
+            $item = $articleParagraphs->item($i);
-            $embedCount = $articleParagraphs->item($i)->getElementsByTagName('embed')->length;
-            $objectCount = $articleParagraphs->item($i)->getElementsByTagName('object')->length;
-            $videoCount = $articleParagraphs->item($i)->getElementsByTagName('video')->length;
-            $audioCount = $articleParagraphs->item($i)->getElementsByTagName('audio')->length;
-            $iframeCount = $articleParagraphs->item($i)->getElementsByTagName('iframe')->length;
 
-            if ($iframeCount === 0 && $imgCount === 0 && $embedCount === 0 && $objectCount === 0 && $videoCount === 0 && $audioCount === 0 && mb_strlen(preg_replace('/\s+/is', '', $this->getInnerText($articleParagraphs->item($i), false, false))) === 0) {
+            $imgCount = $item->getElementsByTagName('img')->length;
-                $articleParagraphs->item($i)->parentNode->removeChild($articleParagraphs->item($i));
+            $embedCount = $item->getElementsByTagName('embed')->length;
+            $objectCount = $item->getElementsByTagName('object')->length;
+            $videoCount = $item->getElementsByTagName('video')->length;
+            $audioCount = $item->getElementsByTagName('audio')->length;
+            $iframeCount = $item->getElementsByTagName('iframe')->length;
+
+            if ($iframeCount === 0 && $imgCount === 0 && $embedCount === 0 && $objectCount === 0 && $videoCount === 0 && $audioCount === 0 && mb_strlen(preg_replace('/\s+/is', '', $this->getInnerText($item, false, false))) === 0) {
+                $item->parentNode->removeChild($item);
             }
 
             // add extra text to iframe tag to avoid an auto-closing iframe and then break the html code
             if ($iframeCount) {
-                $iframe = $articleParagraphs->item($i)->getElementsByTagName('iframe');
+                $iframe = $item->getElementsByTagName('iframe');
                 $iframe->item(0)->nodeValue = ' ';
 
-                $articleParagraphs->item($i)->parentNode->replaceChild($iframe->item(0), $articleParagraphs->item($i));
+                $item->parentNode->replaceChild($iframe->item(0), $item);
             }
         }
 
@@ -585,8 +630,8 @@ class Readability
                     $articleContent->innerHTML = preg_replace($search, $replace, $articleContent->innerHTML);
                 }
                 unset($search, $replace);
-            } catch (Exception $e) {
+            } catch (\Exception $e) {
-                $this->dbg('Cleaning output HTML failed. Ignoring: '.$e->getMessage());
+                $this->logger->error('Cleaning output HTML failed. Ignoring: '.$e->getMessage());
             }
         }
     }
@@ -595,7 +640,7 @@ class Readability
      * Initialize a node with the readability object. Also checks the
      * className/id for special names to add to its score.
      *
-     * @param Element
+     * @param \DOMElement $node
      */
     protected function initializeNode($node)
     {
@@ -622,10 +667,10 @@ class Readability
             case 'FIGURE':
                 $readability->value += 3;
                 break;
-/*          case 'SECTION': // often misused
+            case 'SECTION':
-                $readability->value += 2;
+                // often misused
+                // $readability->value += 2;
                 break;
-*/
             case 'OL':
             case 'UL':
             case 'DL':
@@ -661,10 +706,12 @@ class Readability
     }
 
     /**
-     * grabArticle - Using a variety of metrics (content score, classname, element types), find the content that is
+     * Using a variety of metrics (content score, classname, element types), find the content that is
      * most likely to be the stuff a user wants to read. Then return it wrapped up in a div.
      *
-     * @return DOMElement
+     * @param \DOMElement $page
+     *
+     * @return \DOMElement|bool
      */
     protected function grabArticle($page = null)
     {
@@ -692,32 +739,30 @@ class Readability
             //  (as in, where they contain no other block level elements).
             if (strcasecmp($tagName, 'div') === 0 || strcasecmp($tagName, 'article') === 0 || strcasecmp($tagName, 'section') === 0) {
                 if (!preg_match($this->regexps['divToPElements'], $node->innerHTML)) {
-                    //$this->dbg('Altering '.$node->getNodePath().' to p');
                     $newNode = $this->dom->createElement('p');
 
                     try {
                         $newNode->innerHTML = $node->innerHTML;
-                        // It's easier to debug using original attributes.
+
-                        //$newNode->setAttribute('class', $node->getAttribute('class'));
+                        $node->parentNode->replaceChild($newNode, $node);
-                        //$newNode->setAttribute('id', $node->getAttribute('id'));
-                        $node = $node->parentNode->replaceChild($newNode, $node);
                         --$nodeIndex;
                         $nodesToScore[] = $newNode;
-                    } catch (Exception $e) {
+                    } catch (\Exception $e) {
-                        $this->dbg('Could not alter div/article to p, reverting back to div: '.$e->getMessage());
+                        $this->logger->error('Could not alter div/article to p, reverting back to div: '.$e->getMessage());
                     }
                 } else {
                     // Will change these P elements back to text nodes after processing.
                     for ($i = 0, $il = $node->childNodes->length; $i < $il; ++$i) {
                         $childNode = $node->childNodes->item($i);
 
-                        if (is_object($childNode) && get_class($childNode) === 'DOMProcessingInstruction') { //executable tags (<?php or <?xml) warning
+                        // executable tags (<?php or <?xml) warning
+                        if (is_object($childNode) && get_class($childNode) === 'DOMProcessingInstruction') {
                             $childNode->parentNode->removeChild($childNode);
+
                             continue;
                         }
 
-                        if ($childNode->nodeType == 3) { // XML_TEXT_NODE
+                        if ($childNode->nodeType === XML_TEXT_NODE) {
-                            //$this->dbg('replacing text node with a P tag with the same content.');
                             $p = $this->dom->createElement('p');
                             $p->innerHTML = $childNode->nodeValue;
                             $p->setAttribute('data-readability-styled', 'true');
@@ -737,12 +782,13 @@ class Readability
          */
         for ($pt = 0, $scored = count($nodesToScore); $pt < $scored; ++$pt) {
             $parentNode = $nodesToScore[$pt]->parentNode;
+
             // No parent node? Move on...
             if (!$parentNode) {
                 continue;
             }
 
-            $grandParentNode = ($parentNode->parentNode instanceof DOMElement) ? $parentNode->parentNode : null;
+            $grandParentNode = $parentNode->parentNode instanceof \DOMElement ? $parentNode->parentNode : null;
             $innerText = $this->getInnerText($nodesToScore[$pt]);
 
             // If this paragraph is less than MIN_PARAGRAPH_LENGTH (default:20) characters, don't even count it.
@@ -772,7 +818,7 @@ class Readability
             /* TEST: For every positive/negative parent tag, add/substract half point. Up to 3 points. *\/
             $up = $nodesToScore[$pt];
             $score = 0;
-            while ($up->parentNode instanceof DOMElement) {
+            while ($up->parentNode instanceof \DOMElement) {
                 $up = $up->parentNode;
                 if (preg_match($this->regexps['positive'], $up->getAttribute('class') . ' ' . $up->getAttribute('id'))) {
                     $score += 0.5;
@@ -789,35 +835,38 @@ class Readability
                 $grandParentNode->getAttributeNode('readability')->value += $contentScore / self::GRANDPARENT_SCORE_DIVISOR;
             }
         }
+
         /*
          * Node prepping: trash nodes that look cruddy (like ones with the class name "comment", etc).
          * This is faster to do before scoring but safer after.
          */
         if ($this->flagIsActive(self::FLAG_STRIP_UNLIKELYS) && $xpath) {
             $candidates = $xpath->query('.//*[(self::footer and count(//footer)<2) or (self::aside and count(//aside)<2)]', $page->documentElement);
+            $node = null;
 
-            for ($node = null, $c = $candidates->length - 1; $c >= 0; --$c) {
+            for ($c = $candidates->length - 1; $c >= 0; --$c) {
                 $node = $candidates->item($c);
                 // node should be readable but not inside of an article otherwise it's probably non-readable block
                 if ($node->hasAttribute('readability') && (int) $node->getAttributeNode('readability')->value < 40 && ($node->parentNode ? strcasecmp($node->parentNode->tagName, 'article') !== 0 : true)) {
-                    $this->dbg('Removing unlikely candidate '.$node->getNodePath().' by "'.$node->tagName.'" with readability '.($node->hasAttribute('readability') ? (int) $node->getAttributeNode('readability')->value : 0));
+                    $this->logger->debug('Removing unlikely candidate (using note) '.$node->getNodePath().' by "'.$node->tagName.'" with readability '.($node->hasAttribute('readability') ? (int) $node->getAttributeNode('readability')->value : 0));
                     $node->parentNode->removeChild($node);
                 }
             }
 
             $candidates = $xpath->query('.//*[not(self::body) and (@class or @id or @style) and ((number(@readability) < 40) or not(@readability))]', $page->documentElement);
+            $node = null;
 
-            for ($node = null, $c = $candidates->length - 1; $c >= 0; --$c) {
+            for ($c = $candidates->length - 1; $c >= 0; --$c) {
                 $node = $candidates->item($c);
-                $tagName = $node->tagName;
+
-                /* Remove unlikely candidates */
+                // Remove unlikely candidates
                 $unlikelyMatchString = $node->getAttribute('class').' '.$node->getAttribute('id').' '.$node->getAttribute('style');
-                //$this->dbg('Processing '.$node->getNodePath().' by "'. $unlikelyMatchString.'" with readability '.($node->hasAttribute('readability') ? (int)$node->getAttributeNode('readability')->value : 0));
+
                 if (mb_strlen($unlikelyMatchString) > 3 && // don't process "empty" strings
                     preg_match($this->regexps['unlikelyCandidates'], $unlikelyMatchString) &&
                     !preg_match($this->regexps['okMaybeItsACandidate'], $unlikelyMatchString)
                 ) {
-                    $this->dbg('Removing unlikely candidate '.$node->getNodePath().' by "'.$unlikelyMatchString.'" with readability '.($node->hasAttribute('readability') ? (int) $node->getAttributeNode('readability')->value : 0));
+                    $this->logger->debug('Removing unlikely candidate (using conf) '.$node->getNodePath().' by "'.$unlikelyMatchString.'" with readability '.($node->hasAttribute('readability') ? (int) $node->getAttributeNode('readability')->value : 0));
                     $node->parentNode->removeChild($node);
                     --$nodeIndex;
                 }
@@ -835,15 +884,17 @@ class Readability
             $candidates = $xpath->query('.//*[@data-candidate]', $page->documentElement);
 
             for ($c = $candidates->length - 1; $c >= 0; --$c) {
+                $item = $candidates->item($c);
+
                 // Scale the final candidates score based on link density. Good content should have a
                 // relatively small link density (5% or less) and be mostly unaffected by this operation.
                 // If not for this we would have used XPath to find maximum @readability.
-                $readability = $candidates->item($c)->getAttributeNode('readability');
+                $readability = $item->getAttributeNode('readability');
-                $readability->value = round($readability->value * (1 - $this->getLinkDensity($candidates->item($c))), 0, PHP_ROUND_HALF_UP);
+                $readability->value = round($readability->value * (1 - $this->getLinkDensity($item)), 0, PHP_ROUND_HALF_UP);
 
                 if (!$topCandidate || $readability->value > (int) $topCandidate->getAttribute('readability')) {
-                    $this->dbg('Candidate: '.$candidates->item($c)->getNodePath().' ('.$candidates->item($c)->getAttribute('class').':'.$candidates->item($c)->getAttribute('id').') with score '.$readability->value);
+                    $this->logger->debug('Candidate: '.$item->getNodePath().' ('.$item->getAttribute('class').':'.$item->getAttribute('id').') with score '.$readability->value);
-                    $topCandidate = $candidates->item($c);
+                    $topCandidate = $item;
                 }
             }
 
@@ -860,9 +911,9 @@ class Readability
             if ($page instanceof \DOMDocument) {
                 if (!isset($page->documentElement)) {
                     // we don't have a body either? what a mess! :)
-                    $this->dbg('The page has no body!');
+                    $this->logger->debug('The page has no body!');
                 } else {
-                    $this->dbg('Setting body to a raw HTML of original page!');
+                    $this->logger->debug('Setting body to a raw HTML of original page!');
                     $topCandidate->innerHTML = $page->documentElement->innerHTML;
                     $page->documentElement->innerHTML = '';
                     $this->reinitBody();
@@ -882,7 +933,7 @@ class Readability
         if (strcasecmp($tagName, 'td') === 0 || strcasecmp($tagName, 'tr') === 0) {
             $up = $topCandidate;
 
-            if ($up->parentNode instanceof DOMElement) {
+            if ($up->parentNode instanceof \DOMElement) {
                 $up = $up->parentNode;
 
                 if (strcasecmp($up->tagName, 'table') === 0) {
@@ -891,7 +942,7 @@ class Readability
             }
         }
 
-        $this->dbg('Top candidate: '.$topCandidate->getNodePath());
+        $this->logger->debug('Top candidate: '.$topCandidate->getNodePath());
 
         /*
          * Now that we have the top candidate, look through its siblings for content that might also be related.
@@ -911,9 +962,8 @@ class Readability
             $siblingNode = $siblingNodes->item($s);
             $siblingNodeName = $siblingNode->nodeName;
             $append = false;
-            $this->dbg('Looking at sibling node: '.$siblingNode->getNodePath().(($siblingNode->nodeType === XML_ELEMENT_NODE && $siblingNode->hasAttribute('readability')) ? (' with score '.$siblingNode->getAttribute('readability')) : ''));
+            $this->logger->debug('Looking at sibling node: '.$siblingNode->getNodePath().(($siblingNode->nodeType === XML_ELEMENT_NODE && $siblingNode->hasAttribute('readability')) ? (' with score '.$siblingNode->getAttribute('readability')) : ''));
 
-            //$this->dbg('Sibling has score ' . ($siblingNode->readability ? siblingNode.readability.contentScore : 'Unknown'));
             if ($siblingNode->isSameNode($topCandidate)) {
                 $append = true;
             }
@@ -934,31 +984,25 @@ class Readability
                 $nodeContent = $this->getInnerText($siblingNode, true, true);
                 $nodeLength = mb_strlen($nodeContent);
 
-                if ($nodeLength > self::MIN_NODE_LENGTH && $linkDensity < self::MAX_LINK_DENSITY) {
+                if (($nodeLength > self::MIN_NODE_LENGTH && $linkDensity < self::MAX_LINK_DENSITY)
-                    $append = true;
+                    || ($nodeLength < self::MIN_NODE_LENGTH && $linkDensity === 0 && preg_match('/\.( |$)/', $nodeContent))) {
-                } elseif ($nodeLength < self::MIN_NODE_LENGTH && $linkDensity === 0 && preg_match('/\.( |$)/', $nodeContent)) {
                     $append = true;
                 }
             }
 
             if ($append) {
-                $this->dbg('Appending node: '.$siblingNode->getNodePath());
+                $this->logger->debug('Appending node: '.$siblingNode->getNodePath());
-                $nodeToAppend = null;
 
                 if (strcasecmp($siblingNodeName, 'div') !== 0 && strcasecmp($siblingNodeName, 'p') !== 0) {
-                    /* 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. */
+                    // 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.
-                    $this->dbg('Altering siblingNode '.$siblingNodeName.' to div.');
+                    $this->logger->debug('Altering siblingNode "'.$siblingNodeName.'" to "div".');
                     $nodeToAppend = $this->dom->createElement('div');
 
                     try {
-                        if ($siblingNode->getAttribute('id')) {
-                            $nodeToAppend->setAttribute('id', $siblingNode->getAttribute('id'));
-                        }
-
                         $nodeToAppend->setAttribute('alt', $siblingNodeName);
                         $nodeToAppend->innerHTML = $siblingNode->innerHTML;
-                    } catch (Exception $e) {
+                    } catch (\Exception $e) {
-                        $this->dbg('Could not alter siblingNode '.$siblingNodeName.' to div, reverting to original.');
+                        $this->logger->debug('Could not alter siblingNode "'.$siblingNodeName.'" to "div", reverting to original.');
                         $nodeToAppend = $siblingNode;
                         --$s;
                         --$sl;
@@ -994,17 +1038,17 @@ class Readability
 
             if ($this->flagIsActive(self::FLAG_STRIP_UNLIKELYS)) {
                 $this->removeFlag(self::FLAG_STRIP_UNLIKELYS);
-                $this->dbg('...content is shorter than '.self::MIN_ARTICLE_LENGTH." letters, trying not to strip unlikely content.\n");
+                $this->logger->debug('...content is shorter than '.self::MIN_ARTICLE_LENGTH." letters, trying not to strip unlikely content.\n");
 
                 return $this->grabArticle($this->body);
             } elseif ($this->flagIsActive(self::FLAG_WEIGHT_ATTRIBUTES)) {
                 $this->removeFlag(self::FLAG_WEIGHT_ATTRIBUTES);
-                $this->dbg('...content is shorter than '.self::MIN_ARTICLE_LENGTH." letters, trying not to weight attributes.\n");
+                $this->logger->debug('...content is shorter than '.self::MIN_ARTICLE_LENGTH." letters, trying not to weight attributes.\n");
 
                 return $this->grabArticle($this->body);
             } elseif ($this->flagIsActive(self::FLAG_CLEAN_CONDITIONALLY)) {
                 $this->removeFlag(self::FLAG_CLEAN_CONDITIONALLY);
-                $this->dbg('...content is shorter than '.self::MIN_ARTICLE_LENGTH." letters, trying not to clean at all.\n");
+                $this->logger->debug('...content is shorter than '.self::MIN_ARTICLE_LENGTH." letters, trying not to clean at all.\n");
 
                 return $this->grabArticle($this->body);
             }
@@ -1019,7 +1063,7 @@ class Readability
      * Get the inner text of a node.
      * This also strips out any excess whitespace to be found.
      *
-     * @param DOMElement $e
+     * @param \DOMElement $e
      * @param bool        $normalizeSpaces (default: true)
      * @param bool        $flattenLines    (default: false)
      *
@@ -1027,7 +1071,7 @@ class Readability
      */
     public function getInnerText($e, $normalizeSpaces = true, $flattenLines = false)
     {
-        if (!isset($e->textContent) || $e->textContent === '') {
+        if (null === $e || !isset($e->textContent) || $e->textContent === '') {
             return '';
         }
 
@@ -1045,7 +1089,7 @@ class Readability
     /**
      * Remove the style attribute on every $e and under.
      *
-     * @param DOMElement $e
+     * @param \DOMElement $e
      */
     public function cleanStyles($e)
     {
@@ -1065,7 +1109,7 @@ class Readability
      *
      * @param string $text
      *
-     * @return number (integer)
+     * @return int
      */
     public function getCommaCount($text)
     {
@@ -1078,7 +1122,7 @@ class Readability
      *
      * @param string $text
      *
-     * @return number (integer)
+     * @return int
      */
     public function getWordCount($text)
     {
@@ -1090,10 +1134,10 @@ class Readability
      * This is the amount of text that is inside a link divided by the total text in the node.
      * Can exclude external references to differentiate between simple text and menus/infoblocks.
      *
-     * @param DOMElement $e
+     * @param \DOMElement $e
      * @param string      $excludeExternal
      *
-     * @return number (float)
+     * @return int
      */
     public function getLinkDensity($e, $excludeExternal = false)
     {
@@ -1119,10 +1163,10 @@ class Readability
      * Get an element weight by attribute.
      * Uses regular expressions to tell if this element looks good or bad.
      *
-     * @param DOMElement $element
+     * @param \DOMElement $element
      * @param string      $attribute
      *
-     * @return number (Integer)
+     * @return int
      */
     protected function weightAttribute($element, $attribute)
     {
@@ -1131,19 +1175,20 @@ class Readability
         }
         $weight = 0;
 
-        //$attribute_val = trim($element->getAttribute('class')." ".$element->getAttribute('id'));
+        // $attributeValue = trim($element->getAttribute('class')." ".$element->getAttribute('id'));
-        $attribute_val = trim($element->getAttribute($attribute));
+        $attributeValue = trim($element->getAttribute($attribute));
-        if ($attribute_val != '') {
+
-            if (preg_match($this->regexps['negative'], $attribute_val)) {
+        if ($attributeValue != '') {
+            if (preg_match($this->regexps['negative'], $attributeValue)) {
                 $weight -= 25;
             }
-            if (preg_match($this->regexps['positive'], $attribute_val)) {
+            if (preg_match($this->regexps['positive'], $attributeValue)) {
                 $weight += 25;
             }
-            if (preg_match($this->regexps['unlikelyCandidates'], $attribute_val)) {
+            if (preg_match($this->regexps['unlikelyCandidates'], $attributeValue)) {
                 $weight -= 5;
             }
-            if (preg_match($this->regexps['okMaybeItsACandidate'], $attribute_val)) {
+            if (preg_match($this->regexps['okMaybeItsACandidate'], $attributeValue)) {
                 $weight += 5;
             }
         }
@@ -1154,9 +1199,9 @@ class Readability
     /**
      * Get an element relative weight.
      *
-     * @param DOMElement $e
+     * @param \DOMElement $e
      *
-     * @return number (Integer)
+     * @return int
      */
     public function getWeight($e)
     {
@@ -1165,9 +1210,9 @@ class Readability
         }
 
         $weight = 0;
-        /* Look for a special classname */
+        // Look for a special classname
         $weight += $this->weightAttribute($e, 'class');
-        /* Look for a special ID */
+        // Look for a special ID
         $weight += $this->weightAttribute($e, 'id');
 
         return $weight;
@@ -1176,7 +1221,7 @@ class Readability
     /**
      * Remove extraneous break tags from a node.
      *
-     * @param DOMElement $node
+     * @param \DOMElement $node
      */
     public function killBreaks($node)
     {
@@ -1191,33 +1236,34 @@ class Readability
      *
      * Updated 2012-09-18 to preserve youtube/vimeo iframes
      *
-     * @param DOMElement $e
+     * @param \DOMElement $e
      * @param string      $tag
      */
     public function clean($e, $tag)
     {
+        $currentItem = null;
         $targetList = $e->getElementsByTagName($tag);
         $isEmbed = ($tag === 'audio' || $tag === 'video' || $tag === 'iframe' || $tag === 'object' || $tag === 'embed');
 
-        for ($cur_item = null, $y = $targetList->length - 1; $y >= 0; --$y) {
+        for ($y = $targetList->length - 1; $y >= 0; --$y) {
-            /* Allow youtube and vimeo videos through as people usually want to see those. */
+            // Allow youtube and vimeo videos through as people usually want to see those.
-            $cur_item = $targetList->item($y);
+            $currentItem = $targetList->item($y);
 
             if ($isEmbed) {
-                $attributeValues = $cur_item->getAttribute('src').' '.$cur_item->getAttribute('href');
+                $attributeValues = $currentItem->getAttribute('src').' '.$currentItem->getAttribute('href');
 
-                /* First, check the elements attributes to see if any of them contain known media hosts */
+                // First, check the elements attributes to see if any of them contain known media hosts
                 if (preg_match($this->regexps['media'], $attributeValues)) {
                     continue;
                 }
 
-                /* Then check the elements inside this element for the same. */
+                // Then check the elements inside this element for the same.
                 if (preg_match($this->regexps['media'], $targetList->item($y)->innerHTML)) {
                     continue;
                 }
             }
 
-            $cur_item->parentNode->removeChild($cur_item);
+            $currentItem->parentNode->removeChild($currentItem);
         }
     }
 
@@ -1226,7 +1272,7 @@ class Readability
      * "Fishy" is an algorithm based on content length, classnames,
      * link density, number of images & embeds, etc.
      *
-     * @param DOMElement $e
+     * @param \DOMElement $e
      * @param string      $tag
      */
     public function cleanConditionally($e, $tag)
@@ -1237,6 +1283,7 @@ class Readability
 
         $tagsList = $e->getElementsByTagName($tag);
         $curTagsLength = $tagsList->length;
+        $node = null;
 
         /*
          * Gather counts for other typical elements embedded within.
@@ -1244,15 +1291,14 @@ class Readability
          *
          * TODO: Consider taking into account original contentScore here.
          */
-        for ($node = null, $i = $curTagsLength - 1; $i >= 0; --$i) {
+        for ($i = $curTagsLength - 1; $i >= 0; --$i) {
             $node = $tagsList->item($i);
-            //$class = $node->getAttribute('class').' '.$node->getAttribute('id'); //debug
             $weight = $this->getWeight($node);
             $contentScore = ($node->hasAttribute('readability')) ? (int) $node->getAttribute('readability') : 0;
-            $this->dbg('Start conditional cleaning of '.$node->getNodePath().' (class='.$node->getAttribute('class').'; id='.$node->getAttribute('id').')'.(($node->hasAttribute('readability')) ? (' with score '.$node->getAttribute('readability')) : ''));
+            $this->logger->debug('Start conditional cleaning of '.$node->getNodePath().' (class='.$node->getAttribute('class').'; id='.$node->getAttribute('id').')'.(($node->hasAttribute('readability')) ? (' with score '.$node->getAttribute('readability')) : ''));
 
             if ($weight + $contentScore < 0) {
-                $this->dbg('Removing...');
+                $this->logger->debug('Removing...');
                 $node->parentNode->removeChild($node);
             } elseif ($this->getCommaCount($this->getInnerText($node)) < self::MIN_COMMAS_IN_PARAGRAPH) {
                 /*
@@ -1286,52 +1332,51 @@ class Readability
 
                 if ($this->lightClean) {
                     if ($li > $p && $tag != 'ul' && $tag != 'ol') {
-                        $this->dbg(' too many <li> elements, and parent is not <ul> or <ol>');
+                        $this->logger->debug(' too many <li> elements, and parent is not <ul> or <ol>');
                         $toRemove = true;
                     } elseif ($input > floor($p / 3)) {
-                        $this->dbg(' too many <input> elements');
+                        $this->logger->debug(' too many <input> elements');
                         $toRemove = true;
                     } elseif ($contentLength < 6 && ($embedCount === 0 && ($img === 0 || $img > 2))) {
-                        $this->dbg(' content length less than 6 chars, 0 embeds and either 0 images or more than 2 images');
+                        $this->logger->debug(' content length less than 6 chars, 0 embeds and either 0 images or more than 2 images');
                         $toRemove = true;
                     } elseif ($weight < 25 && $linkDensity > 0.25) {
-                        $this->dbg(' weight is '.$weight.' < 25 and link density is '.sprintf('%.2f', $linkDensity).' > 0.25');
+                        $this->logger->debug(' weight is '.$weight.' < 25 and link density is '.sprintf('%.2f', $linkDensity).' > 0.25');
                         $toRemove = true;
                     } elseif ($a > 2 && ($weight >= 25 && $linkDensity > 0.5)) {
-                        $this->dbg('  more than 2 links and weight is '.$weight.' > 25 but link density is '.sprintf('%.2f', $linkDensity).' > 0.5');
+                        $this->logger->debug('  more than 2 links and weight is '.$weight.' > 25 but link density is '.sprintf('%.2f', $linkDensity).' > 0.5');
                         $toRemove = true;
                     } elseif ($embedCount > 3) {
-                        $this->dbg(' more than 3 embeds');
+                        $this->logger->debug(' more than 3 embeds');
                         $toRemove = true;
                     }
                 } else {
                     if ($img > $p) {
-                        $this->dbg(' more image elements than paragraph elements');
+                        $this->logger->debug(' more image elements than paragraph elements');
                         $toRemove = true;
                     } elseif ($li > $p && $tag != 'ul' && $tag != 'ol') {
-                        $this->dbg('  too many <li> elements, and parent is not <ul> or <ol>');
+                        $this->logger->debug('  too many <li> elements, and parent is not <ul> or <ol>');
                         $toRemove = true;
                     } elseif ($input > floor($p / 3)) {
-                        $this->dbg('  too many <input> elements');
+                        $this->logger->debug('  too many <input> elements');
                         $toRemove = true;
                     } elseif ($contentLength < 10 && ($img === 0 || $img > 2)) {
-                        $this->dbg('  content length less than 10 chars and 0 images, or more than 2 images');
+                        $this->logger->debug('  content length less than 10 chars and 0 images, or more than 2 images');
                         $toRemove = true;
                     } elseif ($weight < 25 && $linkDensity > 0.2) {
-                        $this->dbg('  weight is '.$weight.' lower than 0 and link density is '.sprintf('%.2f', $linkDensity).' > 0.2');
+                        $this->logger->debug('  weight is '.$weight.' lower than 0 and link density is '.sprintf('%.2f', $linkDensity).' > 0.2');
                         $toRemove = true;
                     } elseif ($weight >= 25 && $linkDensity > 0.5) {
-                        $this->dbg('  weight above 25 but link density is '.sprintf('%.2f', $linkDensity).' > 0.5');
+                        $this->logger->debug('  weight above 25 but link density is '.sprintf('%.2f', $linkDensity).' > 0.5');
                         $toRemove = true;
                     } elseif (($embedCount == 1 && $contentLength < 75) || $embedCount > 1) {
-                        $this->dbg('  1 embed and content length smaller than 75 chars, or more than one embed');
+                        $this->logger->debug('  1 embed and content length smaller than 75 chars, or more than one embed');
                         $toRemove = true;
                     }
                 }
 
                 if ($toRemove) {
-                    //$this->dbg('Removing: '.$node->innerHTML);
+                    $this->logger->debug('Removing...');
-                    $this->dbg('Removing...');
                     $node->parentNode->removeChild($node);
                 }
             }
@@ -1341,12 +1386,13 @@ class Readability
     /**
      * Clean out spurious headers from an Element. Checks things like classnames and link density.
      *
-     * @param DOMElement $e
+     * @param \DOMElement $e
      */
     public function cleanHeaders($e)
     {
         for ($headerIndex = 1; $headerIndex < 3; ++$headerIndex) {
             $headers = $e->getElementsByTagName('h'.$headerIndex);
+
             for ($i = $headers->length - 1; $i >= 0; --$i) {
                 if ($this->getWeight($headers->item($i)) < 0 || $this->getLinkDensity($headers->item($i)) > 0.33) {
                     $headers->item($i)->parentNode->removeChild($headers->item($i));
@@ -1355,16 +1401,33 @@ class Readability
         }
     }
 
+    /**
+     * Check if the given flag is active.
+     *
+     * @param int $flag
+     *
+     * @return bool
+     */
     public function flagIsActive($flag)
     {
         return ($this->flags & $flag) > 0;
     }
 
+    /**
+     * Add a flag.
+     *
+     * @param int $flag
+     */
     public function addFlag($flag)
     {
         $this->flags = $this->flags | $flag;
     }
 
+    /**
+     * Remove a flag.
+     *
+     * @param int $flag
+     */
     public function removeFlag($flag)
     {
         $this->flags = $this->flags & ~$flag;

tests/ReadabilityTest.php

@@ -3,33 +3,33 @@
 namespace Tests\Readability;
 
 use Readability\Readability;
-
+use Monolog\Logger;
-class ReadabilityTested extends Readability
+use Monolog\Handler\TestHandler;
-{
-    public function getDebugText()
-    {
-        return $this->debugText;
-    }
-
-    public function getDomainRegexp()
-    {
-        return $this->domainRegExp;
-    }
-}
 
 class ReadabilityTest extends \PHPUnit_Framework_TestCase
 {
+    public $logHandler;
+    public $logger;
+
+    private function getReadability($html, $url = null, $parser = 'libxml', $useTidy = true)
+    {
+        $readability = new Readability($html, $url, $parser, $useTidy);
+
+        $this->logHandler = new TestHandler();
+        $this->logger = new Logger('test', array($this->logHandler));
+        $readability->setLogger($this->logger);
+
+        return $readability;
+    }
+
     /**
      * @requires extension tidy
      */
     public function testConstructDefault()
     {
-        $readability = new ReadabilityTested('');
+        $readability = $this->getReadability('');
 
         $this->assertNull($readability->url);
-        $this->assertContains('Parsing URL', $readability->getDebugText());
-        $this->assertContains('Tidying document', $readability->getDebugText());
-        $this->assertNull($readability->getDomainRegexp());
         $this->assertInstanceOf('DomDocument', $readability->dom);
     }
 
@@ -38,40 +38,38 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
      */
     public function testConstructSimple()
     {
-        $readability = new ReadabilityTested('<html/>', 'http://0.0.0.0');
+        $readability = $this->getReadability('<html/>', 'http://0.0.0.0');
 
         $this->assertEquals('http://0.0.0.0', $readability->url);
-        $this->assertContains('Parsing URL', $readability->getDebugText());
-        $this->assertContains('Tidying document', $readability->getDebugText());
-        $this->assertEquals('/0\.0\.0\.0/', $readability->getDomainRegexp());
         $this->assertInstanceOf('DomDocument', $readability->dom);
+        $this->assertEquals('<html/>', $readability->original_html);
+        $this->assertTrue($readability->tidied);
     }
 
     public function testConstructDefaultWithoutTidy()
     {
-        $readability = new ReadabilityTested('', null, 'libxml', false);
+        $readability = $this->getReadability('', null, 'libxml', false);
 
         $this->assertNull($readability->url);
-        $this->assertContains('Parsing URL', $readability->getDebugText());
+        $this->assertEquals('', $readability->original_html);
-        $this->assertNotContains('Tidying document', $readability->getDebugText());
+        $this->assertFalse($readability->tidied);
-        $this->assertNull($readability->getDomainRegexp());
+
         $this->assertInstanceOf('DomDocument', $readability->dom);
     }
 
     public function testConstructSimpleWithoutTidy()
     {
-        $readability = new ReadabilityTested('<html/>', 'http://0.0.0.0', 'libxml', false);
+        $readability = $this->getReadability('<html/>', 'http://0.0.0.0', 'libxml', false);
 
         $this->assertEquals('http://0.0.0.0', $readability->url);
-        $this->assertContains('Parsing URL', $readability->getDebugText());
-        $this->assertNotContains('Tidying document', $readability->getDebugText());
-        $this->assertEquals('/0\.0\.0\.0/', $readability->getDomainRegexp());
         $this->assertInstanceOf('DomDocument', $readability->dom);
+        $this->assertEquals('<html/>', $readability->original_html);
+        $this->assertFalse($readability->tidied);
     }
 
     public function testInitNoContent()
     {
-        $readability = new ReadabilityTested('<html/>', 'http://0.0.0.0');
+        $readability = $this->getReadability('<html/>', 'http://0.0.0.0');
         $res = $readability->init();
 
         $this->assertFalse($res);
@@ -83,7 +81,7 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
 
     public function testInitP()
     {
-        $readability = new ReadabilityTested(str_repeat('<p>This is the awesome content :)</p>', 7), 'http://0.0.0.0');
+        $readability = $this->getReadability(str_repeat('<p>This is the awesome content :)</p>', 7), 'http://0.0.0.0');
         $res = $readability->init();
 
         $this->assertTrue($res);
@@ -96,7 +94,7 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
 
     public function testInitDivP()
     {
-        $readability = new ReadabilityTested('<div>'.str_repeat('<p>This is the awesome content :)</p>', 7).'</div>', 'http://0.0.0.0');
+        $readability = $this->getReadability('<div>'.str_repeat('<p>This is the awesome content :)</p>', 7).'</div>', 'http://0.0.0.0');
         $res = $readability->init();
 
         $this->assertTrue($res);
@@ -109,7 +107,7 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
 
     public function testInitDiv()
     {
-        $readability = new ReadabilityTested('<div>'.str_repeat('This is the awesome content :)', 7).'</div>', 'http://0.0.0.0');
+        $readability = $this->getReadability('<div>'.str_repeat('This is the awesome content :)', 7).'</div>', 'http://0.0.0.0');
         $readability->debug = true;
         $res = $readability->init();
 
@@ -123,7 +121,7 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
 
     public function testWithFootnotes()
     {
-        $readability = new ReadabilityTested('<div>'.str_repeat('<p>This is an awesome text with some links, here there are: <a href="http://0.0.0.0/test.html">the awesome</a></p>', 7).'</div>', 'http://0.0.0.0');
+        $readability = $this->getReadability('<div>'.str_repeat('<p>This is an awesome text with some links, here there are: <a href="http://0.0.0.0/test.html">the awesome</a></p>', 7).'</div>', 'http://0.0.0.0');
         $readability->debug = true;
         $readability->convertLinksToFootnotes = true;
         $res = $readability->init();
@@ -140,7 +138,7 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
 
     public function testStandardClean()
     {
-        $readability = new ReadabilityTested('<div><h2>Title</h2>'.str_repeat('<p>This is an awesome text with some links, here there are: <a href="http://0.0.0.0/test.html">the awesome</a></p>', 7).'<a href="#nofollow" rel="nofollow">will NOT be removed</a></div>', 'http://0.0.0.0');
+        $readability = $this->getReadability('<div><h2>Title</h2>'.str_repeat('<p>This is an awesome text with some links, here there are: <a href="http://0.0.0.0/test.html">the awesome</a></p>', 7).'<a href="#nofollow" rel="nofollow">will NOT be removed</a></div>', 'http://0.0.0.0');
         $readability->debug = true;
         $readability->lightClean = false;
         $res = $readability->init();
@@ -157,7 +155,7 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
 
     public function testWithIframe()
     {
-        $readability = new ReadabilityTested('<div><h2>Title</h2>'.str_repeat('<p>This is an awesome text with some links, here there are: <a href="http://0.0.0.0/test.html">the awesome</a></p>', 7).'<p>This is an awesome text with some links, here there are <iframe src="http://youtube.com/test" href="#nofollow" rel="nofollow"></iframe><iframe>http://soundcloud.com/test</iframe></p></div>', 'http://0.0.0.0');
+        $readability = $this->getReadability('<div><h2>Title</h2>'.str_repeat('<p>This is an awesome text with some links, here there are: <a href="http://0.0.0.0/test.html">the awesome</a></p>', 7).'<p>This is an awesome text with some links, here there are <iframe src="http://youtube.com/test" href="#nofollow" rel="nofollow"></iframe><iframe>http://soundcloud.com/test</iframe></p></div>', 'http://0.0.0.0');
         $readability->debug = true;
         $res = $readability->init();
 
@@ -172,7 +170,7 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
 
     public function testWithArticle()
     {
-        $readability = new ReadabilityTested('<article><p>'.str_repeat('This is an awesome text with some links, here there are: the awesome', 20).'</p><p>This is an awesome text with some links, here there are <iframe src="http://youtube.com/test" href="#nofollow" rel="nofollow"></iframe></p></article>', 'http://0.0.0.0');
+        $readability = $this->getReadability('<article><p>'.str_repeat('This is an awesome text with some links, here there are: the awesome', 20).'</p><p>This is an awesome text with some links, here there are <iframe src="http://youtube.com/test" href="#nofollow" rel="nofollow"></iframe></p></article>', 'http://0.0.0.0');
         $readability->debug = true;
         $res = $readability->init();
 
@@ -187,7 +185,22 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
 
     public function testWithAside()
     {
-        $readability = new ReadabilityTested('<article>'.str_repeat('<p>This is an awesome text with some links, here there are: <a href="http://0.0.0.0/test.html">the awesome</a></p>', 7).'<footer><aside>'.str_repeat('<p>This is an awesome text with some links, here there are</p>', 8).'</aside></footer></article>', 'http://0.0.0.0');
+        $readability = $this->getReadability('<article>'.str_repeat('<p>This is an awesome text with some links, here there are: <a href="http://0.0.0.0/test.html">the awesome</a></p>', 7).'<footer><aside>'.str_repeat('<p>This is an awesome text with some links, here there are</p>', 8).'</aside></footer></article>', 'http://0.0.0.0');
+        $readability->debug = true;
+        $res = $readability->init();
+
+        $this->assertTrue($res);
+        $this->assertInstanceOf('Readability\JSLikeHTMLElement', $readability->getContent());
+        $this->assertInstanceOf('Readability\JSLikeHTMLElement', $readability->getTitle());
+        $this->assertEmpty($readability->getTitle()->innerHTML);
+        $this->assertContains('This is an awesome text with some links, here there are', $readability->getContent()->innerHTML);
+        $this->assertNotContains('<aside>', $readability->getContent()->innerHTML);
+        $this->assertContains('<footer readability="4"/>', $readability->getContent()->innerHTML);
+    }
+
+    public function testWithClasses()
+    {
+        $readability = $this->getReadability('<article>'.str_repeat('<p>This is an awesome text with some links, here there are: <a href="http://0.0.0.0/test.html">the awesome</a></p>', 7).'<div style="display:none">'.str_repeat('<p class="clock">This text should be removed</p>', 10).'</div></article>', 'http://0.0.0.0');
         $readability->debug = true;
         $res = $readability->init();
 
@@ -197,14 +210,14 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
         $this->assertContains('alt="article"', $readability->getContent()->innerHTML);
         $this->assertEmpty($readability->getTitle()->innerHTML);
         $this->assertContains('This is an awesome text with some links, here there are', $readability->getContent()->innerHTML);
-        $this->assertNotContains('<aside>', $readability->getContent()->innerHTML);
+        $this->assertNotContains('This text should be removed', $readability->getContent()->innerHTML);
-        $this->assertContains('<footer/>', $readability->getContent()->innerHTML);
     }
 
-    public function testWithClasses()
+    public function testWithClassesWithoutLightClean()
     {
-        $readability = new ReadabilityTested('<article>'.str_repeat('<p>This is an awesome text with some links, here there are: <a href="http://0.0.0.0/test.html">the awesome</a></p>', 7).'<div style="display:none">'.str_repeat('<p class="clock">This text should be removed</p>', 10).'</div></article>', 'http://0.0.0.0');
+        $readability = $this->getReadability('<article>'.str_repeat('<p>This is an awesome text with some links, here there are: <a href="http://0.0.0.0/test.html">the awesome</a></p>', 7).'<div style="display:none">'.str_repeat('<p class="clock">This text should be removed</p>', 10).'</div></article>', 'http://0.0.0.0');
         $readability->debug = true;
+        $readability->lightClean = false;
         $res = $readability->init();
 
         $this->assertTrue($res);
@@ -218,28 +231,26 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
 
     public function testWithTd()
     {
-        $readability = new ReadabilityTested('<table><tr>'.str_repeat('<td><p>This is an awesome text with some links, here there are the awesome</td>', 7).'</tr></table>', 'http://0.0.0.0');
+        $readability = $this->getReadability('<table><tr>'.str_repeat('<td><p>This is an awesome text with some links, here there are the awesome</td>', 7).'</tr></table>', 'http://0.0.0.0');
         $readability->debug = true;
         $res = $readability->init();
 
         $this->assertTrue($res);
         $this->assertInstanceOf('Readability\JSLikeHTMLElement', $readability->getContent());
         $this->assertInstanceOf('Readability\JSLikeHTMLElement', $readability->getTitle());
-        $this->assertContains('alt="tr"', $readability->getContent()->innerHTML);
         $this->assertEmpty($readability->getTitle()->innerHTML);
         $this->assertContains('This is an awesome text with some links, here there are', $readability->getContent()->innerHTML);
     }
 
     public function testWithSameClasses()
     {
-        $readability = new ReadabilityTested('<article class="awesomecontent">'.str_repeat('<p>This is an awesome text with some links, here there are the awesome</p>', 7).'<div class="awesomecontent">This text is also an awesome text and you should know that !</div></article>', 'http://0.0.0.0');
+        $readability = $this->getReadability('<article class="awesomecontent">'.str_repeat('<p>This is an awesome text with some links, here there are the awesome</p>', 7).'<div class="awesomecontent">This text is also an awesome text and you should know that !</div></article>', 'http://0.0.0.0');
         $readability->debug = true;
         $res = $readability->init();
 
         $this->assertTrue($res);
         $this->assertInstanceOf('Readability\JSLikeHTMLElement', $readability->getContent());
         $this->assertInstanceOf('Readability\JSLikeHTMLElement', $readability->getTitle());
-        $this->assertContains('alt="article"', $readability->getContent()->innerHTML);
         $this->assertEmpty($readability->getTitle()->innerHTML);
         $this->assertContains('This is an awesome text with some links, here there are', $readability->getContent()->innerHTML);
         $this->assertContains('This text is also an awesome text and you should know that', $readability->getContent()->innerHTML);
@@ -247,14 +258,13 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
 
     public function testWithScript()
     {
-        $readability = new ReadabilityTested('<article class="awesomecontent">'.str_repeat('<p>This is an awesome text with some links, here there are the awesome</p>', 7).'<p><script>This text is also an awesome text and you should know that !</script></p></article>', 'http://0.0.0.0');
+        $readability = $this->getReadability('<article class="awesomecontent">'.str_repeat('<p>This is an awesome text with some links, here there are the awesome</p>', 7).'<p><script>This text is also an awesome text and you should know that !</script></p></article>', 'http://0.0.0.0');
         $readability->debug = true;
         $res = $readability->init();
 
         $this->assertTrue($res);
         $this->assertInstanceOf('Readability\JSLikeHTMLElement', $readability->getContent());
         $this->assertInstanceOf('Readability\JSLikeHTMLElement', $readability->getTitle());
-        $this->assertContains('alt="article"', $readability->getContent()->innerHTML);
         $this->assertEmpty($readability->getTitle()->innerHTML);
         $this->assertContains('This is an awesome text with some links, here there are', $readability->getContent()->innerHTML);
         $this->assertNotContains('This text is also an awesome text and you should know that', $readability->getContent()->innerHTML);
@@ -262,14 +272,13 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
 
     public function testTitle()
     {
-        $readability = new ReadabilityTested('<title>this is my title</title><article class="awesomecontent">'.str_repeat('<p>This is an awesome text with some links, here there are the awesome</p>', 7).'<p></p></article>', 'http://0.0.0.0');
+        $readability = $this->getReadability('<title>this is my title</title><article class="awesomecontent">'.str_repeat('<p>This is an awesome text with some links, here there are the awesome</p>', 7).'<p></p></article>', 'http://0.0.0.0');
         $readability->debug = true;
         $res = $readability->init();
 
         $this->assertTrue($res);
         $this->assertInstanceOf('Readability\JSLikeHTMLElement', $readability->getContent());
         $this->assertInstanceOf('Readability\JSLikeHTMLElement', $readability->getTitle());
-        $this->assertContains('alt="article"', $readability->getContent()->innerHTML);
         $this->assertEquals('this is my title', $readability->getTitle()->innerHTML);
         $this->assertContains('This is an awesome text with some links, here there are', $readability->getContent()->innerHTML);
         $this->assertNotContains('This text is also an awesome text and you should know that', $readability->getContent()->innerHTML);
@@ -277,14 +286,13 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
 
     public function testTitleWithDash()
     {
-        $readability = new ReadabilityTested('<title> title2 - title3 </title><article class="awesomecontent">'.str_repeat('<p>This is an awesome text with some links, here there are the awesome</p>', 7).'<p></p></article>', 'http://0.0.0.0');
+        $readability = $this->getReadability('<title> title2 - title3 </title><article class="awesomecontent">'.str_repeat('<p>This is an awesome text with some links, here there are the awesome</p>', 7).'<p></p></article>', 'http://0.0.0.0');
         $readability->debug = true;
         $res = $readability->init();
 
         $this->assertTrue($res);
         $this->assertInstanceOf('Readability\JSLikeHTMLElement', $readability->getContent());
         $this->assertInstanceOf('Readability\JSLikeHTMLElement', $readability->getTitle());
-        $this->assertContains('alt="article"', $readability->getContent()->innerHTML);
         $this->assertEquals('title2 - title3', $readability->getTitle()->innerHTML);
         $this->assertContains('This is an awesome text with some links, here there are', $readability->getContent()->innerHTML);
         $this->assertNotContains('This text is also an awesome text and you should know that', $readability->getContent()->innerHTML);
@@ -292,14 +300,13 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
 
     public function testTitleWithDoubleDot()
     {
-        $readability = new ReadabilityTested('<title> title2 : title3 </title><article class="awesomecontent">'.str_repeat('<p>This is an awesome text with some links, here there are the awesome</p>', 7).'<p></p></article>', 'http://0.0.0.0');
+        $readability = $this->getReadability('<title> title2 : title3 </title><article class="awesomecontent">'.str_repeat('<p>This is an awesome text with some links, here there are the awesome</p>', 7).'<p></p></article>', 'http://0.0.0.0');
         $readability->debug = true;
         $res = $readability->init();
 
         $this->assertTrue($res);
         $this->assertInstanceOf('Readability\JSLikeHTMLElement', $readability->getContent());
         $this->assertInstanceOf('Readability\JSLikeHTMLElement', $readability->getTitle());
-        $this->assertContains('alt="article"', $readability->getContent()->innerHTML);
         $this->assertEquals('title2 : title3', $readability->getTitle()->innerHTML);
         $this->assertContains('This is an awesome text with some links, here there are', $readability->getContent()->innerHTML);
         $this->assertNotContains('This text is also an awesome text and you should know that', $readability->getContent()->innerHTML);
@@ -307,14 +314,13 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
 
     public function testTitleTooShortUseH1()
     {
-        $readability = new ReadabilityTested('<title>too short</title><h1>this is my h1 title !</h1><article class="awesomecontent">'.str_repeat('<p>This is an awesome text with some links, here there are the awesome</p>', 7).'<p></p></article>', 'http://0.0.0.0');
+        $readability = $this->getReadability('<title>too short</title><h1>this is my h1 title !</h1><article class="awesomecontent">'.str_repeat('<p>This is an awesome text with some links, here there are the awesome</p>', 7).'<p></p></article>', 'http://0.0.0.0');
         $readability->debug = true;
         $res = $readability->init();
 
         $this->assertTrue($res);
         $this->assertInstanceOf('Readability\JSLikeHTMLElement', $readability->getContent());
         $this->assertInstanceOf('Readability\JSLikeHTMLElement', $readability->getTitle());
-        $this->assertContains('alt="article"', $readability->getContent()->innerHTML);
         $this->assertEquals('this is my h1 title !', $readability->getTitle()->innerHTML);
         $this->assertContains('This is an awesome text with some links, here there are', $readability->getContent()->innerHTML);
         $this->assertNotContains('This text is also an awesome text and you should know that', $readability->getContent()->innerHTML);
@@ -322,13 +328,9 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
 
     // public function testConstructParser()
     // {
-    //     $readability = new ReadabilityTested('<html/>', 'http://0.0.0.0', 'html5lib');
+    //     $readability = $this->getReadability('<html/>', 'http://0.0.0.0', 'html5lib');
 
     //     $this->assertEquals('http://0.0.0.0', $readability->url);
-    //     $this->assertContains('Parsing URL', $readability->getDebugText());
-    //     $this->assertContains('Tidying document', $readability->getDebugText());
-    //     $this->assertEquals('/0\.0\.0\.0/', $readability->getDomainRegexp());
-    //     $this->assertInstanceOf('DomDocument', $readability->dom);
     // }
 
     // dummy function to be used to the next test
@@ -368,7 +370,7 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
             </body>
             </html>';
 
-        $readability = new ReadabilityTested($data, 'http://iosgames.ru/?p=22030');
+        $readability = $this->getReadability($data, 'http://iosgames.ru/?p=22030');
         $readability->debug = true;
 
         $res = $readability->init();
@@ -379,4 +381,87 @@ class ReadabilityTest extends \PHPUnit_Framework_TestCase
         $this->assertContains('<iframe src="https://www.youtube.com/embed/PUep6xNeKjA" width="560" height="315" frameborder="0" allowfullscreen="allowfullscreen"> </iframe>', $readability->getContent()->innerHTML);
         $this->assertContains('3D Touch', $readability->getTitle()->innerHTML);
     }
+
+    /**
+     * This should generate an Exception "DOMElement::setAttribute(): ID post-60 already defined"
+     */
+    public function testAppendIdAlreadyHere()
+    {
+        $data = '<!DOCTYPE html>
+            <html lang="fr">
+            <head>
+                <meta charset="utf-8">
+                <meta name="viewport" content="width=device-width, user-scalable=yes, initial-scale=1.0">
+            </head>
+            <body>
+                <div class="container">
+                    <header class="header sml-text-center med-text-left" role="banner">
+                        <h1 class="no-margin"><a class="maintitle" href="https://0.0.0.0" title="Bloc-notes">Bloc-notes</a></h1>
+                        <h2 class="h5 no-margin"></h2>
+                    </header>
+
+                    <nav class="nav" role="navigation">
+                        <div class="responsive-menu">
+                            <label for="menu">Menu</label>
+                            <input type="checkbox" id="menu">
+                        </div>
+                    </nav>
+
+                    <article class="article" role="article" id="post-60">
+                        <section>
+                            This is an awesome text with some links, here there are<br/>
+                            This is an awesome text with some links, here there are<br/>
+                            This is an awesome text with some links, here there are<br/>
+                            This is an awesome text with some links, here there are<br/>
+                            This is an awesome text with some links, here there are<br/>
+                            This is an awesome text with some links, here there are<br/>
+                            This is an awesome text with some links, here there are<br/>
+                            This is an awesome text with some links, here there are<br/>
+                            This is an awesome text with some links, here there are<br/>
+                            This is an awesome text with some links, here there are<br/>
+                            This is an awesome text with some links, here there are
+                        </section>
+                        <footer>
+                            <small>
+                                Classé dans : <a class="noactive" title="Services réseaux">Services réseaux</a>
+                            </small>
+                        </footer>
+                    </article>
+                </div>
+            </body>
+            </html>';
+
+        $readability = $this->getReadability($data, 'http://0.0.0.0');
+        $readability->debug = true;
+
+        $res = $readability->init();
+
+        $this->assertTrue($res);
+        $this->assertInstanceOf('Readability\JSLikeHTMLElement', $readability->getContent());
+        $this->assertInstanceOf('Readability\JSLikeHTMLElement', $readability->getTitle());
+    }
+
+    public function testPostFilters()
+    {
+        $readability = $this->getReadability('<div>'.str_repeat('<p>This <b>is</b> the awesome content :)</p>', 7).'</div>', 'http://0.0.0.0');
+        $readability->addPostFilter('!<strong[^>]*>(.*?)</strong>!is', '');
+
+        $res = $readability->init();
+
+        $this->assertTrue($res);
+        $this->assertContains('This  the awesome content :)', $readability->getContent()->innerHTML);
+    }
+
+    public function testPreFilters()
+    {
+        $this->markTestSkipped('Won\'t work until loadHtml() is moved in init() instead of __construct()');
+
+        $readability = $this->getReadability('<div>'.str_repeat('<p>This <b>is</b> the awesome and WONDERFUL content :)</p>', 7).'</div>', 'http://0.0.0.0');
+        $readability->addPreFilter('!<b[^>]*>(.*?)</b>!is', '');
+
+        $res = $readability->init();
+
+        $this->assertTrue($res);
+        $this->assertContains('This the awesome and WONDERFUL content :)', $readability->getContent()->innerHTML);
+    }
 }