From 3d99ce9dadca82765832ab669dabc22c554fd78e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Nicolas=20L=C5=93uillet?= ".$data." Para 1 Para 2 Para 1 Para 2 The story begins...".$this->settings["title"]."
";
-
- list($text, $entries) = $this->generateText();
-
- if($this->settings["toc"]) {
- $toc = $this->generateTOC($entries); //Generate TOC to get the right length
- $toc = $this->generateTOC($entries, strlen($prefix)+strlen($toc)+strlen($title)); //Generate the real TOC
- }
-
- $suffix = "";
-
- return $prefix.$toc.$title.$text.$suffix;
- }
-
- /**
- * Generate the body's text and the chapter entries
- * @return array($string, $entries) $string is the html data, $entries
- * contains the level, the title and the position of the titles.
- */
- public function generateText(){
- $str = "";
- $entries = array();
-
- for($i = 0; $i < sizeof($this->parts); $i++){
- list($type, $data) = $this->parts[$i];
- $id = "title_".$i;
- switch($type){
- case self::PARAGRAPH:
- $str .= "".$data."
";
- break;
- case self::H3:
- $entries[] = array("level" => 3, "position" => strlen($str), "title" => $data, "id" => $id);
- $str .= "".$data."
";
- break;
- case self::IMAGE:
- $str .= "";
- break;
- }
- }
- return array($str, $entries);
- }
-
- /**
- * Generate a TOC
- * @param $entries The entries array generated by generateText
- * @param $base The zero position
- */
- public function generateTOC($entries, $base = 0){
- $toc = "Contents
";
- $toc .= " ";
- }
- return $toc."".$entry["title"]." Contents
";
- $toc .= "
";
-
- return $toc;
- }
- /**
- * Get the images (an array containing the jpeg data). Array entry 0 will
- * correspond to image record 0.
- * @return array
- */
- public function getImages(){
- return $this->images;
- }
-
- /**
- * Get the metadata in the form of a hashtable (for example, title or author).
- * @return array
- */
- public function getMetaData(){
- return $this->metadata;
- }
-
-}
-?>
diff --git a/inc/3rdparty/libraries/MOBIClass/OnlineArticle.php b/inc/3rdparty/libraries/MOBIClass/OnlineArticle.php
deleted file mode 100644
index ec3182fe..00000000
--- a/inc/3rdparty/libraries/MOBIClass/OnlineArticle.php
+++ /dev/null
@@ -1,116 +0,0 @@
-init();
- if(!isset($this->metadata["title"])){
- $this->metadata["title"] = CharacterEntities::convert(strip_tags($r->getTitle()->innerHTML));
- }
- if(!isset($this->metadata["author"])){
- $parts = parse_url($url);
- $this->metadata["author"] = $parts["host"];
- }
-
- $article = $r->getContent()->innerHTML;
- if(substr($article, 0, 5) == "".$article."";
- }else{
- $article = "".$article."";
- }
- $doc = new DOMDocument();
- @$doc->loadHTML($article) or die($article);
- $doc->normalizeDocument();
-
- $this->images = $this->handleImages($doc, $url);
- $this->text = $doc->saveHTML();
- }
-
- /**
- * Get the text data to be integrated in the MOBI file
- * @return string
- */
- public function getTextData(){
- return $this->text;
- }
- /**
- * Get the images (an array containing the jpeg data). Array entry 0 will
- * correspond to image record 0.
- * @return array
- */
- public function getImages(){
- return $this->images;
- }
- /**
- * Get the metadata in the form of a hashtable (for example, title or author).
- * @return array
- */
- public function getMetaData(){
- return $this->metadata;
- }
- /**
- *
- * @param DOMElement $dom
- * @return array
- */
- private function handleImages($dom, $url){
- $images = array();
-
- $parts = parse_url($url);
-
- $savedImages = array();
-
- $imgElements = $dom->getElementsByTagName('img');
- foreach($imgElements as $img) {
- $src = $img->getAttribute("src");
-
- $is_root = false;
- if(substr($src, 0, 1) == "/"){
- $is_root = true;
- }
-
- $parsed = parse_url($src);
-
- if(!isset($parsed["host"])){
- if($is_root){
- $src = http_build_url($url, $parsed, HTTP_URL_REPLACE);
- }else{
- $src = http_build_url($url, $parsed, HTTP_URL_JOIN_PATH);
- }
- }
- $img->setAttribute("src", "");
- if(isset($savedImages[$src])){
- $img->setAttribute("recindex", $savedImages[$src]);
- }else{
- $image = ImageHandler::DownloadImage($src);
-
- if($image !== false){
- $images[$this->imgCounter] = new FileRecord(new Record($image));
-
- $img->setAttribute("recindex", $this->imgCounter);
- $savedImages[$src] = $this->imgCounter;
- $this->imgCounter++;
- }
- }
- }
-
- return $images;
- }
-}
-?>
diff --git a/inc/3rdparty/libraries/MOBIClass/PalmRecord.php b/inc/3rdparty/libraries/MOBIClass/PalmRecord.php
deleted file mode 100644
index d0de8dfe..00000000
--- a/inc/3rdparty/libraries/MOBIClass/PalmRecord.php
+++ /dev/null
@@ -1,136 +0,0 @@
-elements = new FileElement(array(
- "compression"=>new FileShort(),
- "unused"=>new FileShort(),
- "textLength"=>new FileInt(),
- "recordCount"=>new FileShort(),
- "recordSize"=>new FileShort(),
- "encryptionType"=>new FileShort(),
- "unused2"=>new FileShort(),
- //MOBI Header
- "mobiIdentifier"=>new FileString("MOBI", 4),
- "mobiHeaderLength"=>new FileInt(),
- "mobiType"=>new FileInt(),
- "textEncoding"=>new FileInt(),
- "uniqueID"=>new FileInt(),
- "fileVersion"=>new FileInt(),
- "reserved"=>new FileString(40),
- "firstNonBookIndex"=>new FileInt(),
- "fullNameOffset"=>new FileInt(),
- "fullNameLength"=>new FileInt(),
- "locale"=>new FileInt(),
- "inputLanguage"=>new FileInt(),
- "outputLanguage"=>new FileInt(),
- "minimumVersion"=>new FileInt(),
- "firstImageIndex"=>new FileInt(),
- "huffmanRecordOffset"=>new FileInt(),
- "huffmanRecordCount"=>new FileInt(),
- "unused3"=>new FileString(8),
- "exthFlags"=>new FileInt(0x40),
- "unknown"=>new FileString(32),
- "drmOffset"=>new FileInt(0xFFFFFFFF),
- "drmCount"=>new FileShort(0xFFFFFFFF),
- "drmSize"=>new FileShort(),
- "drmFlags"=>new FileInt(),
- "mobiFiller"=>new FileString(72),
- //EXTH Header
- "exthIdentifier"=>new FileString("EXTH", 4),
- "exthHeaderLength"=>new FileInt(),
- "exthRecordCount"=>new FileInt(),
- "exthRecords"=>new FileElement(),
- "exthPadding"=>new FileString(),
- //"fullNamePadding"=>new FileString(100),
- "fullName"=>new FileString()
- ));
-
- //Set values from the info block
- foreach($settings->values as $name => $val){
- //echo $name.", ";
- if($this->elements->exists($name)){
- $this->elements->get($name)->set($settings->get($name));
- }
- }
-
- $els = $settings->values;
-
- $exthElems = new FileElement();
- $i = 0;
- $l = 0;
- foreach($els as $name=>$val){
- $type = EXTHHelper::textToType($name);
- if($type !== false){
- $type = new FileInt($type);
- $length = new FileInt(8+strlen($val));
- $data = new FileString($val);
- $l += 8+strlen($val);
- $exthElems->add("type".$i, $type);
- $exthElems->add("length".$i, $length);
- $exthElems->add("data".$i, $data);
- $i++;
- }
- }
-
- if($images > 0){
- $this->elements->get("firstImageIndex")->set($textRecords+1);
- }
- $this->elements->get("firstNonBookIndex")->set($textRecords+2+$images);
- $this->elements->get("reserved")->set(str_pad("", 40, chr(255), STR_PAD_RIGHT));
- $this->elements->get("exthRecordCount")->set($i);
- $this->elements->set("exthRecords", $exthElems);
- $pad = $l%4;
- $pad = (4-$pad)%4;
- $this->elements->get("exthPadding")->set(str_pad("", $pad, "\0", STR_PAD_RIGHT));
- $this->elements->get("exthHeaderLength")->set(12+$l+$pad);
-
-
- $this->elements->get("recordCount")->set($textRecords);
-
- $this->elements->get("fullNameOffset")->set($this->elements->offsetToEntry("fullName"));
- $this->elements->get("fullNameLength")->set(strlen($settings->get("title")));
- $this->elements->get("fullName")->set($settings->get("title"));
- $this->elements->get("textLength")->set($textLength);
- }
-
- public function getByteLength(){
- return $this->getLength();
- }
-
- public function getLength(){
- return $this->elements->getByteLength();
- }
-
- public function get(){
- return $this;
- }
-
- public function set($elements){
- throw new Exception("Unallowed set");
- }
-
- public function serialize() {
- return $this->elements->serialize();
- }
-
- public function unserialize($data) {
- $this->elements->unserialize($data);
- }
-
- public function __toString(){
- $output = "PalmDoc Record (".$this->getByteLength()." bytes):\n";
- $output .= $this->elements;
- return $output;
- }
-}
-?>
diff --git a/inc/3rdparty/libraries/MOBIClass/Prc.php b/inc/3rdparty/libraries/MOBIClass/Prc.php
deleted file mode 100644
index c40b5805..00000000
--- a/inc/3rdparty/libraries/MOBIClass/Prc.php
+++ /dev/null
@@ -1,97 +0,0 @@
-new FileString(32),
- "attributes"=>new FileShort(),
- "version"=>new FileShort(),
- "creationTime"=>new FileDate(),
- "modificationTime"=>new FileDate(),
- "backupTime"=>new FileDate(),
- "modificationNumber"=>new FileInt(),
- "appInfoID"=>new FileInt(),
- "sortInfoID"=>new FileInt(),
- "prcType"=>new FileString(4),
- "creator"=>new FileString(4),
- "uniqueIDSeed"=>new FileInt(),
- "nextRecordListID"=>new FileInt(),
- "numberRecords"=>new FileShort(),
- "recordList"=>new FileElement(),
- "filler"=>new FileShort(),
- "records"=>new FileElement()
- ));
-
- //Set values from the info block
- foreach($this->elements as $name => $val){
- if($settings->exists($name)){
- $this->get($name)->set($settings->get($name));
- }
- }
-
- $this->get("numberRecords")->set(sizeof($records));
-
- $i = 0;
- foreach($records as $record){
- $offset = new FileInt();
- $attr = new FileByte();
- $uniqueID = new FileTri($i);
-
- $this->elements["recordList"]->add("Rec".$i, new FileElement(array(
- "offset"=>$offset,
- "attribute"=>$attr,
- "uniqueID"=>$uniqueID
- )));
-
- $this->elements["records"]->add("Rec".$i, $record);
- $i++;
- }
-
- $this->updateOffsets($records);
- }
-
- public function getByteLength(){
- throw new Exception("Test");
- }
-
- public function updateOffsets($records){
- $base = $this->offsetToEntry("records");
-
- $i = 0;
-
- foreach($records as $record){
- $el = $this->elements["recordList"]->get("Rec".$i);
-
- $local = $this->elements["records"]->offsetToEntry("Rec".$i);
-
- $el->get("offset")->set($base+$local);
-
- $i++;
- }
- }
-
- public function save($file){
- $handle = fopen($file, "w");
- fwrite($handle, $this->serialize());
- fclose($handle);
- }
-
- public function output(){
- echo $this->serialize();
- }
-
- public function __toString(){
- $output = "Prc (".$this->getByteLength()." bytes): {\n";
- foreach($this->elements as $key=>$value){
- $output .= "\t".$key.": ".$value."\n";
- }
- $output .= "}";
- return $output;
- }
-}
-?>
diff --git a/inc/3rdparty/libraries/MOBIClass/PreprocessedArticle.php b/inc/3rdparty/libraries/MOBIClass/PreprocessedArticle.php
deleted file mode 100644
index 2e992404..00000000
--- a/inc/3rdparty/libraries/MOBIClass/PreprocessedArticle.php
+++ /dev/null
@@ -1,89 +0,0 @@
-text = $textData;
- $this->metadata = $metadata;
-
- $this->images = $this->downloadImages($imageLinks);
- }
-
- /**
- * Create a Preprocessed article from a json string
- * @param string $json JSON data. Should be of the following format:
- * {"text": "TEXT", "images: ["imageURL1", "imageURL2"], "metadata": {"key": "value"}}
- *
- * Note: Any image tags should have the recindex attribute set to the appropriate index (the
- * same index as the image in the array)
- * @return PreprocessedArticle The generated preprocessed array
- */
- static public function CreateFromJson($json){
- $data = json_decode($json);
- return new PreprocessedArticle($data["text"], $data["images"], $data["metadata"]);
- }
-
- /**
- * Get the text data to be integrated in the MOBI file
- * @return string
- */
- public function getTextData(){
- return $this->text;
- }
- /**
- * Get the images (an array containing the jpeg data). Array entry 0 will
- * correspond to image record 0.
- * @return array
- */
- public function getImages(){
- return $this->images;
- }
- /**
- * Get the metadata in the form of a hashtable (for example, title or author).
- * @return array
- */
- public function getMetaData(){
- return $this->metadata;
- }
- /**
- *
- * @param DOMElement $dom
- * @return array
- */
- private function downloadImages($links){
- $images = array();
- foreach($links as $link) {
- $imgFile = @imagecreatefromstring(Http::Request($link));
-
- if($imgFile === false){
- $imgFile = @imagecreate(1, 1);
- $black = @imagecolorallocate($imgFile, 255, 255, 255);
- }
- if($imgFile !== false){
- @imagefilter($imgFile, IMG_FILTER_GRAYSCALE);
-
- ob_start();
- @imagejpeg($imgFile);
- $image = ob_get_contents();
- ob_end_clean();
-
- $images[$this->imgCounter] = new FileRecord(new Record($image));
- imagedestroy($imgFile);
-
- $this->imgCounter++;
- }
- }
-
- return $images;
- }
-}
-?>
diff --git a/inc/3rdparty/libraries/MOBIClass/RecognizeURL.php b/inc/3rdparty/libraries/MOBIClass/RecognizeURL.php
deleted file mode 100644
index 6319ed57..00000000
--- a/inc/3rdparty/libraries/MOBIClass/RecognizeURL.php
+++ /dev/null
@@ -1,16 +0,0 @@
-
diff --git a/inc/3rdparty/libraries/MOBIClass/Record.php b/inc/3rdparty/libraries/MOBIClass/Record.php
deleted file mode 100644
index 3cb39582..00000000
--- a/inc/3rdparty/libraries/MOBIClass/Record.php
+++ /dev/null
@@ -1,96 +0,0 @@
-data = $data;
- if($length >= 0){
- $this->length = $length;
- }else{
- $this->length = strlen($data);
- }
- }
-
- public function compress($compression_method){
- switch($compression_method){
- case NO_COMPRESSION:
- //Finished!
- break;
- case PALMDOC_COMPRESSION:
- throw new Exception("Not implemented yet");
- break;
- case HUFF:
- throw new Exception("Not implemented yet");
- break;
- default:
- throw new Exception("Invalid argument");
- }
- }
-
- public function getByteLength(){
- return $this->getLength();
- }
-
- /**
- * Get the length of the record
- * @return int Length of the data
- */
- public function getLength(){
- return $this->length;
- }
-
- /**
- * Get the data contained in the record
- * @return string Data contained in the record
- */
- public function get(){
- return $this->data;
- }
-
- /**
- * Set the data contained in the record
- * @param string $value Data contained in the record
- */
- public function set($value){
- $this->data = $value;
- }
-
- public function serialize(){
- return $this->data;
- }
- public function unserialize($data){
- __construct($data);
- }
-
- public function __toString() {
- $toShow = $this->data;
- if(strlen($this->data) > 103){
- $toShow = substr($this->data, 0, 100)."...";
- }
- $out = "Record: {\n";
- $out .= "\t".htmlspecialchars($toShow)."\n";
- $out .= "}";
- return $out;
- }
-}
-?>
diff --git a/inc/3rdparty/libraries/MOBIClass/RecordFactory.php b/inc/3rdparty/libraries/MOBIClass/RecordFactory.php
deleted file mode 100644
index 12806fe3..00000000
--- a/inc/3rdparty/libraries/MOBIClass/RecordFactory.php
+++ /dev/null
@@ -1,115 +0,0 @@
-settings = $settings;
- }
-
- /**
- * Create records from a data string
- * @param string $data
- * @return array(Record)
- */
- public function createRecords($data){
- $records = array();
- $size = $this->settings->get("recordSize");
- $compression = $this->settings->get("compression");
-
- $dataEntries = mb_str_split($data, $size);
-
- for($i = 0, $len = sizeof($dataEntries); $i < $len; $i++){
- $records[$i] = new Record($dataEntries[$i]);
- $records[$i]->compress($compression);
- }
-
- return $records;
- }
-
- public function createEOFRecord(){
- return new Record(0xe98e0d0a);
- }
-
- public function createFCISRecord($textLength){
- $r = "FCIS";
- $r .= $this->asString(20, 4);
- $r .= $this->asString(16, 4);
- $r .= $this->asString(1, 4);
- $r .= $this->asString(0, 4);
- $r .= $this->asString($textLength, 4);
- $r .= $this->asString(0, 4);
- $r .= $this->asString(32, 4);
- $r .= $this->asString(8, 4);
- $r .= $this->asString(1, 2);
- $r .= $this->asString(1, 2);
- $r .= $this->asString(0, 4);
- return new Record($r);
- }
-
- public function createFLISRecord(){
- $r = "FLIS";
- $r .= $this->asString(8, 4);
- $r .= $this->asString(65, 2);
- $r .= $this->asString(0, 2);
- $r .= $this->asString(0, 4);
- $r .= $this->asString(-1, 4);
- $r .= $this->asString(1, 2);
- $r .= $this->asString(3, 2);
- $r .= $this->asString(3, 4);
- $r .= $this->asString(1, 4);
- $r .= $this->asString(-1, 4);
- return new Record($r);
- }
-
- private function asString($int, $size){
- $out = "";
- for($i = 0; $i < $size; $i++){
- if($i > 0) $out = " ".$out;
- $byte = dechex($int & 0xFF);
- if(strlen($byte) == 1) $byte = "0".$byte;
- $out = $byte.$out;
- $int = $int >> 8;
- }
- return $out;
- }
-
- public function __toString() {
- $out = "Record Factory: {\n";
- $out .= "\tRecord Size: ".$this->settings->get("recordSize")."\n";
- $out .= "\tCompression: ".$this->settings->get("compression")."\n";
- $out .= "}";
- return $out;
- }
-}
-function mb_str_split($string, $split_length = 1){
- mb_internal_encoding('UTF-8');
- mb_regex_encoding('UTF-8');
-
- $split_length = ($split_length <= 0) ? 1 : $split_length;
-
- $mb_strlen = mb_strlen($string, 'utf-8');
-
- $array = array();
-
- for($i = 0; $i < $mb_strlen; $i += $split_length){
- $array[] = mb_substr($string, $i, $split_length);
- }
-
- return $array;
-}
-?>
diff --git a/inc/3rdparty/libraries/MOBIClass/Settings.php b/inc/3rdparty/libraries/MOBIClass/Settings.php
deleted file mode 100644
index ddcf2054..00000000
--- a/inc/3rdparty/libraries/MOBIClass/Settings.php
+++ /dev/null
@@ -1,97 +0,0 @@
-values = array(
- "attributes"=>0,
- "version"=>0,
- "creationTime"=>time()+94694400,
- "modificationTime"=>time()+94694400,
- "backupTime"=>0,
- "modificationNumber"=>0,
- "appInfoID"=>0,
- "sortInfoID"=>0,
- "prcType"=>"BOOK",
- "creator"=>"MOBI",
- "uniqueIDSeed"=>rand(),
- "nextRecordListID"=>0,
- "recordAttributes"=>0,
- "compression"=>NO_COMPRESSION,
- "recordSize"=>RECORD_SIZE,
- "encryptionType"=>NO_ENCRYPTION,
- "mobiIdentifier"=>"MOBI",
- "mobiHeaderLength"=>0xe8,
- "mobiType"=>MOBIPOCKET_BOOK,
- "textEncoding"=>UTF8,
- "uniqueID"=>rand(),
- "fileVersion"=>6,
- "locale"=>0x09,
- "inputLanguage"=>0,
- "outputLanguage"=>0,
- "minimumVersion"=>6,
- "huffmanRecordOffset"=>0,
- "huffmanRecordCount"=>0,
- "exthFlags"=>0x40,
- "drmOffset"=>0xFFFFFFFF,
- "drmCount"=>0,
- "drmSize"=>0,
- "drmFlags"=>0,
- "extraDataFlags"=>0,
- "exthIdentifier"=>"EXTH",
- // These can be changed without any risk
- "title"=>"Unknown title",
- "author"=>"Unknown author",
- "subject"=>"Unknown subject"
- );
-
- foreach($additionalSettings as $key=>$value){
- $this->values[$key] = $value;
- }
- }
-
- /**
- * Get a value from the settings
- * @param string $key Key of the setting
- * @return mixed The value of the setting
- */
- public function get($key){
- return $this->values[$key];
- }
-
- /**
- * Checks if a value is set
- * @param string $key Key of the setting
- * @return bool True if the value exists
- */
- public function exists($key){
- return isset($this->values[$key]);
- }
-
- public function __toString() {
- $out = "Settings: {\n";
- foreach($this->values as $key=>$value){
- $out .= "\t".$key.": ".$value."\n";
- }
- $out .= "}";
- return $out;
- }
-}
-?>
diff --git a/inc/3rdparty/libraries/MOBIClass/constants.php b/inc/3rdparty/libraries/MOBIClass/constants.php
deleted file mode 100644
index bd363118..00000000
--- a/inc/3rdparty/libraries/MOBIClass/constants.php
+++ /dev/null
@@ -1,11 +0,0 @@
-id = intval(substr($ending, 0, strpos($ending, "/")));
-
- for($i = 1; $i <= max(1, $this->chapterCount); $i++){
- $this->addChapter($i);
- }
- }
-
- private function addChapter($n){
- $doc = new DOMDocument();
- $file = Http::Request(self::$prefix.$this->id."/".$n."/");
- @$doc->loadHTML($file) or die($file);
-
- if(!$this->downloadedMetadata){
- $this->loadMetadata($doc);
- $this->downloadedMetadata = true;
- }
- if($this->chapterCount < 0){
- $this->chapterCount = $this->getNumberChapters($doc);
-
- if($this->chapterCount > 4){
- die("Too many files to download, don't use php for this!");
- }
- }
-
- $textEl = $doc->getElementById("storytext");
- if($textEl == null) die("Error: ".$doc->saveHTML());
- $horizontalRulebars = $doc->getElementsByTagName('hr');
- /**
- * @var DOMNode
- */
- $hr;
- foreach($horizontalRulebars as $hr) {
- $hr->setAttribute("size", null);
- $hr->setAttribute("noshade", null);
- }
- $text = $this->innerHtml($textEl);
-
- $title = "";
- $selects = $doc->getElementsByTagName('select');
- foreach($selects as $select) {
- if($select->hasAttribute("name") && $select->getAttribute("name") == "chapter"){
- $options = $select->getElementsByTagName("option");
-
- $test = $n.". ";
- foreach($options as $option){
- $val = $option->nodeValue;
- if(substr($val, 0, strlen($test)) == $test){
- $title = substr($val, strlen($test));
- break;
- }
- }
- break;
- }
- }
- $this->addPage($text, $title);
- }
-
- private function getNumberChapters($doc){
- $selects = $doc->getElementsByTagName('select');
- foreach($selects as $select) {
- if($select->hasAttribute("name") && $select->getAttribute("name") == "chapter"){
- $options = $select->getElementsByTagName("option");
-
- $count = $options->length;
- return $count;
- }
- }
- }
-
- private function loadMetadata($doc){
- //Author
- $links = $doc->getElementsByTagName('a');
- foreach($links as $link) {
- if($link == null){
- var_dump($link);
- }
- if($link->hasAttribute("href") && substr($link->getAttribute("href"), 0, 3) == "/u/"){
- $this->setMetadata("author", $link->nodeValue);
- }
- }
- //Title
- /*
- $links = $doc->getElementsByTagName('link');
- foreach($links as $link) {
- if($link->hasAttribute("rel") && $link->getAttribute("rel") == "canonical"){
- $url = $link->getAttribute("href");
- $title = str_replace("_", " ", substr($url, strrpos($url, "/")+1));
- $this->setMetadata("title", $title);
- }
- }*/
-
- //TODO: Find a more reliable way to extract the title
- $title = $doc->getElementsByTagName("b")->item(0)->nodeValue;
- $this->setMetadata("title", $title);
- }
-
- private function innerHtml($node){
- $doc = new DOMDocument();
- foreach ($node->childNodes as $child)
- $doc->appendChild($doc->importNode($child, true));
-
- return $doc->saveHTML();
- }
-
- public static function Matches($url){
- //TODO: Implement with regex
- return strpos($url, self::$prefix) !== false;
- }
-}
-?>
diff --git a/inc/3rdparty/libraries/MOBIClass/http_build_url.php b/inc/3rdparty/libraries/MOBIClass/http_build_url.php
deleted file mode 100644
index b475edb0..00000000
--- a/inc/3rdparty/libraries/MOBIClass/http_build_url.php
+++ /dev/null
@@ -1,94 +0,0 @@
-registerNodeClass('DOMElement', 'JSLikeHTMLElement');
-* $doc->loadHTML(' ";
- }
- $toc .= "".($i+1).". ".$entry["title"]." Chapter 2
]*>[ \n\r\t]*){2,}/i',
- 'replaceFonts' => '/<(\/?)font[^>]*>/i',
- // 'trimRe' => '/^\s+|\s+$/g', // PHP has trim()
- 'normalize' => '/\s{2,}/',
- 'killBreaks' => '/(
(\s| ?)*){1,}/',
- 'video' => '!//(player\.|www\.)?(youtube|vimeo|viddler)\.com!i',
- 'skipFootnoteLink' => '/^\s*(\[?[a-z0-9]{1,2}\]?|^|edit|citation needed)\s*$/i'
- );
-
- /* constants */
- const FLAG_STRIP_UNLIKELYS = 1;
- const FLAG_WEIGHT_CLASSES = 2;
- const FLAG_CLEAN_CONDITIONALLY = 4;
-
- /**
- * Create instance of Readability
- * @param string UTF-8 encoded string
- * @param string (optional) URL associated with HTML (used for footnotes)
- * @param string which parser to use for turning raw HTML into a DOMDocument (either 'libxml' or 'html5lib')
- */
- function __construct($html, $url=null, $parser='libxml')
- {
- $this->url = $url;
- /* Turn all double br's into p's */
- $html = preg_replace($this->regexps['replaceBrs'], '
', $html); - $html = preg_replace($this->regexps['replaceFonts'], '<$1span>', $html); - $html = mb_convert_encoding($html, 'HTML-ENTITIES', "UTF-8"); - if (trim($html) == '') $html = ''; - if ($parser=='html5lib' && ($this->dom = HTML5_Parser::parse($html))) { - // all good - } else { - $this->dom = new DOMDocument(); - $this->dom->preserveWhiteSpace = false; - @$this->dom->loadHTML($html); - } - $this->dom->registerNodeClass('DOMElement', 'JSLikeHTMLElement'); - } - - /** - * Get article title element - * @return DOMElement - */ - public function getTitle() { - return $this->articleTitle; - } - - /** - * Get article content element - * @return DOMElement - */ - public function getContent() { - return $this->articleContent; - } - - /** - * Runs readability. - * - * Workflow: - * 1. Prep the document by removing script tags, css, etc. - * 2. Build readability's DOM tree. - * 3. Grab the article content from the current dom tree. - * 4. Replace the current DOM tree with the new one. - * 5. Read peacefully. - * - * @return boolean true if we found content, false otherwise - **/ - public function init() - { - if (!isset($this->dom->documentElement)) return false; - $this->removeScripts($this->dom); - //die($this->getInnerHTML($this->dom->documentElement)); - - // Assume successful outcome - $this->success = true; - - $bodyElems = $this->dom->getElementsByTagName('body'); - if ($bodyElems->length > 0) { - if ($this->bodyCache == null) { - $this->bodyCache = $bodyElems->item(0)->innerHTML; - } - if ($this->body == null) { - $this->body = $bodyElems->item(0); - } - } - - $this->prepDocument(); - - //die($this->dom->documentElement->parentNode->nodeType); - //$this->setInnerHTML($this->dom->documentElement, $this->getInnerHTML($this->dom->documentElement)); - //die($this->getInnerHTML($this->dom->documentElement)); - - /* Build readability's DOM tree */ - $overlay = $this->dom->createElement('div'); - $innerDiv = $this->dom->createElement('div'); - $articleTitle = $this->getArticleTitle(); - $articleContent = $this->grabArticle(); - - if (!$articleContent) { - $this->success = false; - $articleContent = $this->dom->createElement('div'); - $articleContent->setAttribute('id', 'readability-content'); - $articleContent->innerHTML = '
Sorry, Readability was unable to parse this page for content.
'; - } - - $overlay->setAttribute('id', 'readOverlay'); - $innerDiv->setAttribute('id', 'readInner'); - - /* Glue the structure of our document together. */ - $innerDiv->appendChild($articleTitle); - $innerDiv->appendChild($articleContent); - $overlay->appendChild($innerDiv); - - /* Clear the old HTML, insert the new content. */ - $this->body->innerHTML = ''; - $this->body->appendChild($overlay); - //document.body.insertBefore(overlay, document.body.firstChild); - $this->body->removeAttribute('style'); - - $this->postProcessContent($articleContent); - - // Set title and content instance variables - $this->articleTitle = $articleTitle; - $this->articleContent = $articleContent; - - return $this->success; - } - - /** - * Debug - */ - protected function dbg($msg) { - if ($this->debug) echo '* ',$msg, "\n"; - } - - /** - * Run any post-process modifications to article content as necessary. - * - * @param DOMElement - * @return void - */ - public function postProcessContent($articleContent) { - if ($this->convertLinksToFootnotes && !preg_match('/wikipedia\.org/', @$this->url)) { - $this->addFootnotes($articleContent); - } - } - - /** - * Get the article title as an H1. - * - * @return DOMElement - */ - protected function getArticleTitle() { - $curTitle = ''; - $origTitle = ''; - - try { - $curTitle = $origTitle = $this->getInnerText($this->dom->getElementsByTagName('title')->item(0)); - } catch(Exception $e) {} - - if (preg_match('/ [\|\-] /', $curTitle)) - { - $curTitle = preg_replace('/(.*)[\|\-] .*/i', '$1', $origTitle); - - if (count(explode(' ', $curTitle)) < 3) { - $curTitle = preg_replace('/[^\|\-]*[\|\-](.*)/i', '$1', $origTitle); - } - } - else if (strpos($curTitle, ': ') !== false) - { - $curTitle = preg_replace('/.*:(.*)/i', '$1', $origTitle); - - if (count(explode(' ', $curTitle)) < 3) { - $curTitle = preg_replace('/[^:]*[:](.*)/i','$1', $origTitle); - } - } - else if(strlen($curTitle) > 150 || strlen($curTitle) < 15) - { - $hOnes = $this->dom->getElementsByTagName('h1'); - if($hOnes->length == 1) - { - $curTitle = $this->getInnerText($hOnes->item(0)); - } - } - - $curTitle = trim($curTitle); - - if (count(explode(' ', $curTitle)) <= 4) { - $curTitle = $origTitle; - } - - $articleTitle = $this->dom->createElement('h1'); - $articleTitle->innerHTML = $curTitle; - - return $articleTitle; - } - - /** - * Prepare the HTML document for readability to scrape it. - * This includes things like stripping javascript, CSS, and handling terrible markup. - * - * @return void - **/ - protected function prepDocument() { - /** - * In some cases a body element can't be found (if the HTML is totally hosed for example) - * so we create a new body node and append it to the document. - */ - if ($this->body == null) - { - $this->body = $this->dom->createElement('body'); - $this->dom->documentElement->appendChild($this->body); - } - $this->body->setAttribute('id', 'readabilityBody'); - - /* Remove all style tags in head */ - $styleTags = $this->dom->getElementsByTagName('style'); - for ($i = $styleTags->length-1; $i >= 0; $i--) - { - $styleTags->item($i)->parentNode->removeChild($styleTags->item($i)); - } - - /* Turn all double br's into p's */ - /* Note, this is pretty costly as far as processing goes. Maybe optimize later. */ - //document.body.innerHTML = document.body.innerHTML.replace(readability.regexps.replaceBrs, '').replace(readability.regexps.replaceFonts, '<$1span>'); - // We do this in the constructor for PHP as that's when we have raw HTML - before parsing it into a DOM tree. - // Manipulating innerHTML as it's done in JS is not possible in PHP. - } - - /** - * For easier reading, convert this document to have footnotes at the bottom rather than inline links. - * @see http://www.roughtype.com/archives/2010/05/experiments_in.php - * - * @return void - **/ - public function addFootnotes($articleContent) { - $footnotesWrapper = $this->dom->createElement('div'); - $footnotesWrapper->setAttribute('id', 'readability-footnotes'); - $footnotesWrapper->innerHTML = '
tags, etc.
- *
- * @param DOMElement
- * @return void
- */
- function prepArticle($articleContent) {
- $this->cleanStyles($articleContent);
- $this->killBreaks($articleContent);
- if ($this->revertForcedParagraphElements) {
- $this->revertReadabilityStyledElements($articleContent);
- }
-
- /* Clean out junk from the article content */
- $this->cleanConditionally($articleContent, 'form');
- $this->clean($articleContent, 'object');
- $this->clean($articleContent, 'h1');
-
- /**
- * If there is only one h2, they are probably using it
- * as a header and not a subheader, so remove it since we already have a header.
- ***/
- if (!$this->lightClean && ($articleContent->getElementsByTagName('h2')->length == 1)) {
- $this->clean($articleContent, 'h2');
- }
- $this->clean($articleContent, 'iframe');
-
- $this->cleanHeaders($articleContent);
-
- /* Do these last as the previous stuff may have removed junk that will affect these */
- $this->cleanConditionally($articleContent, 'table');
- $this->cleanConditionally($articleContent, 'ul');
- $this->cleanConditionally($articleContent, 'div');
-
- /* Remove extra paragraphs */
- $articleParagraphs = $articleContent->getElementsByTagName('p');
- for ($i = $articleParagraphs->length-1; $i >= 0; $i--)
- {
- $imgCount = $articleParagraphs->item($i)->getElementsByTagName('img')->length;
- $embedCount = $articleParagraphs->item($i)->getElementsByTagName('embed')->length;
- $objectCount = $articleParagraphs->item($i)->getElementsByTagName('object')->length;
- $iframeCount = $articleParagraphs->item($i)->getElementsByTagName('iframe')->length;
-
- if ($imgCount === 0 && $embedCount === 0 && $objectCount === 0 && $iframeCount === 0 && $this->getInnerText($articleParagraphs->item($i), false) == '')
- {
- $articleParagraphs->item($i)->parentNode->removeChild($articleParagraphs->item($i));
- }
- }
-
- try {
- $articleContent->innerHTML = preg_replace('/
]*>\s*
innerHTML);
- //articleContent.innerHTML = articleContent.innerHTML.replace(/
]*>\s*
dbg("Cleaning innerHTML of breaks failed. This is an IE strict-block-elements bug. Ignoring.: " . $e);
- }
- }
-
- /**
- * Initialize a node with the readability object. Also checks the
- * className/id for special names to add to its score.
- *
- * @param Element
- * @return void
- **/
- protected function initializeNode($node) {
- $readability = $this->dom->createAttribute('readability');
- $readability->value = 0; // this is our contentScore
- $node->setAttributeNode($readability);
-
- switch (strtoupper($node->tagName)) { // unsure if strtoupper is needed, but using it just in case
- case 'DIV':
- $readability->value += 5;
- break;
-
- case 'PRE':
- case 'TD':
- case 'BLOCKQUOTE':
- $readability->value += 3;
- break;
-
- case 'ADDRESS':
- case 'OL':
- case 'UL':
- case 'DL':
- case 'DD':
- case 'DT':
- case 'LI':
- case 'FORM':
- $readability->value -= 3;
- break;
-
- case 'H1':
- case 'H2':
- case 'H3':
- case 'H4':
- case 'H5':
- case 'H6':
- case 'TH':
- $readability->value -= 5;
- break;
- }
- $readability->value += $this->getClassWeight($node);
- }
-
- /***
- * grabArticle - 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
- **/
- protected function grabArticle($page=null) {
- $stripUnlikelyCandidates = $this->flagIsActive(self::FLAG_STRIP_UNLIKELYS);
- if (!$page) $page = $this->dom;
- $allElements = $page->getElementsByTagName('*');
- /**
- * First, node prepping. Trash nodes that look cruddy (like ones with the class name "comment", etc), and turn divs
- * into P tags where they have been used inappropriately (as in, where they contain no other block level elements.)
- *
- * Note: Assignment from index for performance. See http://www.peachpit.com/articles/article.aspx?p=31567&seqNum=5
- * TODO: Shouldn't this be a reverse traversal?
- **/
- $node = null;
- $nodesToScore = array();
- for ($nodeIndex = 0; ($node = $allElements->item($nodeIndex)); $nodeIndex++) {
- //for ($nodeIndex=$targetList->length-1; $nodeIndex >= 0; $nodeIndex--) {
- //$node = $targetList->item($nodeIndex);
- $tagName = strtoupper($node->tagName);
- /* Remove unlikely candidates */
- if ($stripUnlikelyCandidates) {
- $unlikelyMatchString = $node->getAttribute('class') . $node->getAttribute('id');
- if (
- preg_match($this->regexps['unlikelyCandidates'], $unlikelyMatchString) &&
- !preg_match($this->regexps['okMaybeItsACandidate'], $unlikelyMatchString) &&
- $tagName != 'BODY'
- )
- {
- $this->dbg('Removing unlikely candidate - ' . $unlikelyMatchString);
- //$nodesToRemove[] = $node;
- $node->parentNode->removeChild($node);
- $nodeIndex--;
- continue;
- }
- }
-
- if ($tagName == 'P' || $tagName == 'TD' || $tagName == 'PRE') {
- $nodesToScore[] = $node;
- }
-
- /* Turn all divs that don't have children block level elements into p's */
- if ($tagName == 'DIV') {
- if (!preg_match($this->regexps['divToPElements'], $node->innerHTML)) {
- //$this->dbg('Altering div to p');
- $newNode = $this->dom->createElement('p');
- try {
- $newNode->innerHTML = $node->innerHTML;
- //$nodesToReplace[] = array('new'=>$newNode, 'old'=>$node);
- $node->parentNode->replaceChild($newNode, $node);
- $nodeIndex--;
- $nodesToScore[] = $node; // or $newNode?
- }
- catch(Exception $e) {
- $this->dbg('Could not alter div to p, reverting back to div.: ' . $e);
- }
- }
- else
- {
- /* EXPERIMENTAL */
- // TODO: change these p elements back to text nodes after processing
- for ($i = 0, $il = $node->childNodes->length; $i < $il; $i++) {
- $childNode = $node->childNodes->item($i);
- if ($childNode->nodeType == 3) { // XML_TEXT_NODE
- //$this->dbg('replacing text node with a p tag with the same content.');
- $p = $this->dom->createElement('p');
- $p->innerHTML = $childNode->nodeValue;
- $p->setAttribute('style', 'display: inline;');
- $p->setAttribute('class', 'readability-styled');
- $childNode->parentNode->replaceChild($p, $childNode);
- }
- }
- }
- }
- }
-
- /**
- * Loop through all paragraphs, and assign a score to them based on how content-y they look.
- * Then add their score to their parent node.
- *
- * A score is determined by things like number of commas, class names, etc. Maybe eventually link density.
- **/
- $candidates = array();
- for ($pt=0; $pt < count($nodesToScore); $pt++) {
- $parentNode = $nodesToScore[$pt]->parentNode;
- // $grandParentNode = $parentNode ? $parentNode->parentNode : null;
- $grandParentNode = !$parentNode ? null : (($parentNode->parentNode instanceof DOMElement) ? $parentNode->parentNode : null);
- $innerText = $this->getInnerText($nodesToScore[$pt]);
-
- if (!$parentNode || !isset($parentNode->tagName)) {
- continue;
- }
-
- /* If this paragraph is less than 25 characters, don't even count it. */
- if(strlen($innerText) < 25) {
- continue;
- }
-
- /* Initialize readability data for the parent. */
- if (!$parentNode->hasAttribute('readability'))
- {
- $this->initializeNode($parentNode);
- $candidates[] = $parentNode;
- }
-
- /* Initialize readability data for the grandparent. */
- if ($grandParentNode && !$grandParentNode->hasAttribute('readability') && isset($grandParentNode->tagName))
- {
- $this->initializeNode($grandParentNode);
- $candidates[] = $grandParentNode;
- }
-
- $contentScore = 0;
-
- /* Add a point for the paragraph itself as a base. */
- $contentScore++;
-
- /* Add points for any commas within this paragraph */
- $contentScore += count(explode(',', $innerText));
-
- /* For every 100 characters in this paragraph, add another point. Up to 3 points. */
- $contentScore += min(floor(strlen($innerText) / 100), 3);
-
- /* Add the score to the parent. The grandparent gets half. */
- $parentNode->getAttributeNode('readability')->value += $contentScore;
-
- if ($grandParentNode) {
- $grandParentNode->getAttributeNode('readability')->value += $contentScore/2;
- }
- }
-
- /**
- * After we've calculated scores, loop through all of the possible candidate nodes we found
- * and find the one with the highest score.
- **/
- $topCandidate = null;
- for ($c=0, $cl=count($candidates); $c < $cl; $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.
- **/
- $readability = $candidates[$c]->getAttributeNode('readability');
- $readability->value = $readability->value * (1-$this->getLinkDensity($candidates[$c]));
-
- $this->dbg('Candidate: ' . $candidates[$c]->tagName . ' (' . $candidates[$c]->getAttribute('class') . ':' . $candidates[$c]->getAttribute('id') . ') with score ' . $readability->value);
-
- if (!$topCandidate || $readability->value > (int)$topCandidate->getAttribute('readability')) {
- $topCandidate = $candidates[$c];
- }
- }
-
- /**
- * If we still have no top candidate, just use the body as a last resort.
- * We also have to copy the body node so it is something we can modify.
- **/
- if ($topCandidate === null || strtoupper($topCandidate->tagName) == 'BODY')
- {
- $topCandidate = $this->dom->createElement('div');
- if ($page instanceof DOMDocument) {
- if (!isset($page->documentElement)) {
- // we don't have a body either? what a mess! :)
- } else {
- $topCandidate->innerHTML = $page->documentElement->innerHTML;
- $page->documentElement->innerHTML = '';
- $this->reinitBody();
- $page->documentElement->appendChild($topCandidate);
- }
- } else {
- $topCandidate->innerHTML = $page->innerHTML;
- $page->innerHTML = '';
- $page->appendChild($topCandidate);
- }
- $this->initializeNode($topCandidate);
- }
-
- /**
- * Now that we have the top candidate, look through its siblings for content that might also be related.
- * Things like preambles, content split by ads that we removed, etc.
- **/
- $articleContent = $this->dom->createElement('div');
- $articleContent->setAttribute('id', 'readability-content');
- $siblingScoreThreshold = max(10, ((int)$topCandidate->getAttribute('readability')) * 0.2);
- $siblingNodes = @$topCandidate->parentNode->childNodes;
- if (!isset($siblingNodes)) {
- $siblingNodes = new stdClass;
- $siblingNodes->length = 0;
- }
-
- for ($s=0, $sl=$siblingNodes->length; $s < $sl; $s++)
- {
- $siblingNode = $siblingNodes->item($s);
- $append = false;
-
- $this->dbg('Looking at sibling node: ' . $siblingNode->nodeName . (($siblingNode->nodeType === XML_ELEMENT_NODE && $siblingNode->hasAttribute('readability')) ? (' with score ' . $siblingNode->getAttribute('readability')) : ''));
-
- //dbg('Sibling has score ' . ($siblingNode->readability ? siblingNode.readability.contentScore : 'Unknown'));
-
- if ($siblingNode === $topCandidate)
- // or if ($siblingNode->isSameNode($topCandidate))
- {
- $append = true;
- }
-
- $contentBonus = 0;
- /* Give a bonus if sibling nodes and top candidates have the example same classname */
- if ($siblingNode->nodeType === XML_ELEMENT_NODE && $siblingNode->getAttribute('class') == $topCandidate->getAttribute('class') && $topCandidate->getAttribute('class') != '') {
- $contentBonus += ((int)$topCandidate->getAttribute('readability')) * 0.2;
- }
-
- if ($siblingNode->nodeType === XML_ELEMENT_NODE && $siblingNode->hasAttribute('readability') && (((int)$siblingNode->getAttribute('readability')) + $contentBonus) >= $siblingScoreThreshold)
- {
- $append = true;
- }
-
- if (strtoupper($siblingNode->nodeName) == 'P') {
- $linkDensity = $this->getLinkDensity($siblingNode);
- $nodeContent = $this->getInnerText($siblingNode);
- $nodeLength = strlen($nodeContent);
-
- if ($nodeLength > 80 && $linkDensity < 0.25)
- {
- $append = true;
- }
- else if ($nodeLength < 80 && $linkDensity === 0 && preg_match('/\.( |$)/', $nodeContent))
- {
- $append = true;
- }
- }
-
- if ($append)
- {
- $this->dbg('Appending node: ' . $siblingNode->nodeName);
-
- $nodeToAppend = null;
- $sibNodeName = strtoupper($siblingNode->nodeName);
- if ($sibNodeName != 'DIV' && $sibNodeName != 'P') {
- /* 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 of ' . $sibNodeName . ' to div.');
- $nodeToAppend = $this->dom->createElement('div');
- try {
- $nodeToAppend->setAttribute('id', $siblingNode->getAttribute('id'));
- $nodeToAppend->innerHTML = $siblingNode->innerHTML;
- }
- catch(Exception $e)
- {
- $this->dbg('Could not alter siblingNode to div, reverting back to original.');
- $nodeToAppend = $siblingNode;
- $s--;
- $sl--;
- }
- } else {
- $nodeToAppend = $siblingNode;
- $s--;
- $sl--;
- }
-
- /* To ensure a node does not interfere with readability styles, remove its classnames */
- $nodeToAppend->removeAttribute('class');
-
- /* Append sibling and subtract from our list because it removes the node when you append to another node */
- $articleContent->appendChild($nodeToAppend);
- }
- }
-
- /**
- * So we have all of the content that we need. Now we clean it up for presentation.
- **/
- $this->prepArticle($articleContent);
-
- /**
- * Now that we've gone through the full algorithm, check to see if we got any meaningful content.
- * If we didn't, we may need to re-run grabArticle with different flags set. This gives us a higher
- * likelihood of finding the content, and the sieve approach gives us a higher likelihood of
- * finding the -right- content.
- **/
- if (strlen($this->getInnerText($articleContent, false)) < 250)
- {
- // 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
- // in the meantime, we check and create an empty element if it's not there.
- $this->reinitBody();
-
- if ($this->flagIsActive(self::FLAG_STRIP_UNLIKELYS)) {
- $this->removeFlag(self::FLAG_STRIP_UNLIKELYS);
- return $this->grabArticle($this->body);
- }
- else if ($this->flagIsActive(self::FLAG_WEIGHT_CLASSES)) {
- $this->removeFlag(self::FLAG_WEIGHT_CLASSES);
- return $this->grabArticle($this->body);
- }
- else if ($this->flagIsActive(self::FLAG_CLEAN_CONDITIONALLY)) {
- $this->removeFlag(self::FLAG_CLEAN_CONDITIONALLY);
- return $this->grabArticle($this->body);
- }
- else {
- return false;
- }
- }
- return $articleContent;
- }
-
- /**
- * Remove script tags from document
- *
- * @param DOMElement
- * @return void
- */
- public function removeScripts($doc) {
- $scripts = $doc->getElementsByTagName('script');
- for($i = $scripts->length-1; $i >= 0; $i--)
- {
- $scripts->item($i)->parentNode->removeChild($scripts->item($i));
- }
- }
-
- /**
- * Get the inner text of a node.
- * This also strips out any excess whitespace to be found.
- *
- * @param DOMElement $
- * @param boolean $normalizeSpaces (default: true)
- * @return string
- **/
- public function getInnerText($e, $normalizeSpaces=true) {
- $textContent = '';
-
- if (!isset($e->textContent) || $e->textContent == '') {
- return '';
- }
-
- $textContent = trim($e->textContent);
-
- if ($normalizeSpaces) {
- return preg_replace($this->regexps['normalize'], ' ', $textContent);
- } else {
- return $textContent;
- }
- }
-
- /**
- * Get the number of times a string $s appears in the node $e.
- *
- * @param DOMElement $e
- * @param string - what to count. Default is ","
- * @return number (integer)
- **/
- public function getCharCount($e, $s=',') {
- return substr_count($this->getInnerText($e), $s);
- }
-
- /**
- * Remove the style attribute on every $e and under.
- *
- * @param DOMElement $e
- * @return void
- */
- public function cleanStyles($e) {
- if (!is_object($e)) return;
- $elems = $e->getElementsByTagName('*');
- foreach ($elems as $elem) {
- $elem->removeAttribute('style');
- }
- }
-
- /**
- * Get the density of links as a percentage of the content
- * This is the amount of text that is inside a link divided by the total text in the node.
- *
- * @param DOMElement $e
- * @return number (float)
- */
- public function getLinkDensity($e) {
- $links = $e->getElementsByTagName('a');
- $textLength = strlen($this->getInnerText($e));
- $linkLength = 0;
- for ($i=0, $il=$links->length; $i < $il; $i++)
- {
- $linkLength += strlen($this->getInnerText($links->item($i)));
- }
- if ($textLength > 0) {
- return $linkLength / $textLength;
- } else {
- return 0;
- }
- }
-
- /**
- * Get an elements class/id weight. Uses regular expressions to tell if this
- * element looks good or bad.
- *
- * @param DOMElement $e
- * @return number (Integer)
- */
- public function getClassWeight($e) {
- if(!$this->flagIsActive(self::FLAG_WEIGHT_CLASSES)) {
- return 0;
- }
-
- $weight = 0;
-
- /* Look for a special classname */
- if ($e->hasAttribute('class') && $e->getAttribute('class') != '')
- {
- if (preg_match($this->regexps['negative'], $e->getAttribute('class'))) {
- $weight -= 25;
- }
- if (preg_match($this->regexps['positive'], $e->getAttribute('class'))) {
- $weight += 25;
- }
- }
-
- /* Look for a special ID */
- if ($e->hasAttribute('id') && $e->getAttribute('id') != '')
- {
- if (preg_match($this->regexps['negative'], $e->getAttribute('id'))) {
- $weight -= 25;
- }
- if (preg_match($this->regexps['positive'], $e->getAttribute('id'))) {
- $weight += 25;
- }
- }
- return $weight;
- }
-
- /**
- * Remove extraneous break tags from a node.
- *
- * @param DOMElement $node
- * @return void
- */
- public function killBreaks($node) {
- $html = $node->innerHTML;
- $html = preg_replace($this->regexps['killBreaks'], '
', $html);
- $node->innerHTML = $html;
- }
-
- /**
- * Clean a node of all elements of type "tag".
- * (Unless it's a youtube/vimeo video. People love movies.)
- *
- * Updated 2012-09-18 to preserve youtube/vimeo iframes
- *
- * @param DOMElement $e
- * @param string $tag
- * @return void
- */
- public function clean($e, $tag) {
- $targetList = $e->getElementsByTagName($tag);
- $isEmbed = ($tag == 'iframe' || $tag == 'object' || $tag == 'embed');
-
- for ($y=$targetList->length-1; $y >= 0; $y--) {
- /* Allow youtube and vimeo videos through as people usually want to see those. */
- if ($isEmbed) {
- $attributeValues = '';
- for ($i=0, $il=$targetList->item($y)->attributes->length; $i < $il; $i++) {
- $attributeValues .= $targetList->item($y)->attributes->item($i)->value . '|'; // DOMAttr? (TODO: test)
- }
-
- /* First, check the elements attributes to see if any of them contain youtube or vimeo */
- if (preg_match($this->regexps['video'], $attributeValues)) {
- continue;
- }
-
- /* Then check the elements inside this element for the same. */
- if (preg_match($this->regexps['video'], $targetList->item($y)->innerHTML)) {
- continue;
- }
- }
- $targetList->item($y)->parentNode->removeChild($targetList->item($y));
- }
- }
-
- /**
- * Clean an element of all tags of type "tag" if they look fishy.
- * "Fishy" is an algorithm based on content length, classnames,
- * link density, number of images & embeds, etc.
- *
- * @param DOMElement $e
- * @param string $tag
- * @return void
- */
- public function cleanConditionally($e, $tag) {
- if (!$this->flagIsActive(self::FLAG_CLEAN_CONDITIONALLY)) {
- return;
- }
-
- $tagsList = $e->getElementsByTagName($tag);
- $curTagsLength = $tagsList->length;
-
- /**
- * Gather counts for other typical elements embedded within.
- * Traverse backwards so we can remove nodes at the same time without effecting the traversal.
- *
- * TODO: Consider taking into account original contentScore here.
- */
- for ($i=$curTagsLength-1; $i >= 0; $i--) {
- $weight = $this->getClassWeight($tagsList->item($i));
- $contentScore = ($tagsList->item($i)->hasAttribute('readability')) ? (int)$tagsList->item($i)->getAttribute('readability') : 0;
-
- $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')) : ''));
-
- if ($weight + $contentScore < 0) {
- $tagsList->item($i)->parentNode->removeChild($tagsList->item($i));
- }
- else if ( $this->getCharCount($tagsList->item($i), ',') < 10) {
- /**
- * If there are not very many commas, and the number of
- * non-paragraph elements is more than paragraphs or other ominous signs, remove the element.
- **/
- $p = $tagsList->item($i)->getElementsByTagName('p')->length;
- $img = $tagsList->item($i)->getElementsByTagName('img')->length;
- $li = $tagsList->item($i)->getElementsByTagName('li')->length-100;
- $input = $tagsList->item($i)->getElementsByTagName('input')->length;
- $a = $tagsList->item($i)->getElementsByTagName('a')->length;
-
- $embedCount = 0;
- $embeds = $tagsList->item($i)->getElementsByTagName('embed');
- for ($ei=0, $il=$embeds->length; $ei < $il; $ei++) {
- if (preg_match($this->regexps['video'], $embeds->item($ei)->getAttribute('src'))) {
- $embedCount++;
- }
- }
- $embeds = $tagsList->item($i)->getElementsByTagName('iframe');
- for ($ei=0, $il=$embeds->length; $ei < $il; $ei++) {
- if (preg_match($this->regexps['video'], $embeds->item($ei)->getAttribute('src'))) {
- $embedCount++;
- }
- }
-
- $linkDensity = $this->getLinkDensity($tagsList->item($i));
- $contentLength = strlen($this->getInnerText($tagsList->item($i)));
- $toRemove = false;
-
- if ($this->lightClean) {
- $this->dbg('Light clean...');
- if ( ($img > $p) && ($img > 4) ) {
- $this->dbg(' more than 4 images and more image elements than paragraph elements');
- $toRemove = true;
- } else if ($li > $p && $tag != 'ul' && $tag != 'ol') {
- $this->dbg(' too many