php提取网页正文内容的例子

<?php

class Readability {
    // 保存判定结果的标记位名称
    const ATTR_CONTENT_SCORE = "contentScore";

    // DOM 解析类目前只支持 UTF-8 编码
    const DOM_DEFAULT_CHARSET = "utf-8";

    // 当判定失败时显示的内容
    const MESSAGE_CAN_NOT_GET = "Readability was unable to parse this page for content.";

    // DOM 解析类（PHP5 已内置）
    protected $DOM = null;

    // 需要解析的源代码
    protected $source = "";

    // 章节的父元素列表
    private $parentNodes = array();

    // 需要删除的标签
    // Note: added extra tags from http://www.111cn.net

    private $junkTags = Array("style", "form", "iframe", "script", "button", "input", "textarea",
                                "noscript", "select", "option", "object", "applet", "basefont",
                                "bgsound", "blink", "canvas", "command", "menu", "nav", "datalist",
                                "embed", "frame", "frameset", "keygen", "label", "marquee", "link");

    // 需要删除的属性
    private $junkAttrs = Array("style", "class", "onclick", "onmouseover", "align", "border", "margin");

    /**
     * 构造函数
     *      @param $input_char 字符串的编码。默认 utf-8，可以省略
     */
    function __construct($source, $input_char = "utf-8") {
        $this->source = $source;

        // DOM 解析类只能处理 UTF-8 格式的字符
        $source = mb_convert_encoding($source, 'HTML-ENTITIES', $input_char);

        // 预处理 HTML 标签，剔除冗余的标签等
        $source = $this->preparSource($source);

        // 生成 DOM 解析类
        $this->DOM = new DOMDocument('1.0', $input_char);
        try {
            //libxml_use_internal_errors(true);
            // 会有些错误信息，不过不要紧 :^)
            if (!@$this->DOM->loadHTML('<?xml encoding="'.Readability::DOM_DEFAULT_CHARSET.'">'.$source)) {
                throw new Exception("Parse HTML Error!");
            }

            foreach ($this->DOM->childNodes as $item) {
                if ($item->nodeType == XML_PI_NODE) {
                    $this->DOM->removeChild($item); // remove hack
                }
            }

            // insert proper
            $this->DOM->encoding = Readability::DOM_DEFAULT_CHARSET;
        } catch (Exception $e) {
            // ...
        }
    }

    /**
     * 预处理 HTML 标签，使其能够准确被 DOM 解析类处理
     *
     * @return String
     */
    private function preparSource($string) {
        // 剔除多余的 HTML 编码标记，避免解析出错
        preg_match("/charset=([＼w|＼-]+);?/", $string, $match);
        if (isset($match[1])) {
            $string = preg_replace("/charset=([＼w|＼-]+);?/", "", $string, 1);
        }

        // Replace all doubled-up <BR> tags with <P> tags, and remove fonts.
        $string = preg_replace("/<br＼/?>[ ＼r＼n＼s]*<br＼/?>/i", "</p><p>", $string);
        $string = preg_replace("/<＼/?font[^>]*>/i", "", $string);

        // @see https://github.com/feelinglucky/php-readability/issues/7
        //   - from http://stackoverflow.com/questions/7130867/remove-script-tag-from-html-content
        $string = preg_replace("#<script(.*?)>(.*?)</script>#is", "", $string);

        return trim($string);
    }

    /**
     * 删除 DOM 元素中所有的 $TagName 标签
     *
     * @return DOMDocument
     */
    private function removeJunkTag($RootNode, $TagName) {
       
        $Tags = $RootNode->getElementsByTagName($TagName);
       
        //Note: always index 0, because removing a tag removes it from the results as well.
        while($Tag = $Tags->item(0)){
            $parentNode = $Tag->parentNode;
            $parentNode->removeChild($Tag);
        }
       
        return $RootNode;
       
    }

    /**
     * 删除元素中所有不需要的属性
     */
    private function removeJunkAttr($RootNode, $Attr) {
        $Tags = $RootNode->getElementsByTagName("*");

        $i = 0;
        while($Tag = $Tags->item($i++)) {
            $Tag->removeAttribute($Attr);
        }

        return $RootNode;
    }

    /**
     * 根据评分获取页面主要内容的盒模型
     *      判定算法来自：http://code.google.com/p/arc90labs-readability/  
     *      这里由郑晓博客转发
     * @return DOMNode
     */
    private function getTopBox() {
        // 获得页面所有的章节
        $allParagraphs = $this->DOM->getElementsByTagName("p");

        // Study all the paragraphs and find the chunk that has the best score.
        // A score is determined by things like: Number of <p>'s, commas, special classes, etc.
        $i = 0;
        while($paragraph = $allParagraphs->item($i++)) {
            $parentNode   = $paragraph->parentNode;
            $contentScore = intval($parentNode->getAttribute(Readability::ATTR_CONTENT_SCORE));
            $className    = $parentNode->getAttribute("class");
            $id           = $parentNode->getAttribute("id");

            // Look for a special classname
            if (preg_match("/(comment|meta|footer|footnote)/i", $className)) {
                $contentScore -= 50;
            } else if(preg_match(
                "/((^|＼＼s)(post|hentry|entry[-]?(content|text|body)?|article[-]?(content|text|body)?)(＼＼s|$))/i",
                $className)) {
                $contentScore += 25;
            }

            // Look for a special ID
            if (preg_match("/(comment|meta|footer|footnote)/i", $id)) {
                $contentScore -= 50;
            } else if (preg_match(
                "/^(post|hentry|entry[-]?(content|text|body)?|article[-]?(content|text|body)?)$/i",
                $id)) {
                $contentScore += 25;
            }

            // Add a point for the paragraph found
            // Add points for any commas within this paragraph
            if (strlen($paragraph->nodeValue) > 10) {
                $contentScore += strlen($paragraph->nodeValue);
            }

            // 保存父元素的判定得分
            $parentNode->setAttribute(Readability::ATTR_CONTENT_SCORE, $contentScore);

            // 保存章节的父元素，以便下次快速获取
            array_push($this->parentNodes, $parentNode);
        }

        $topBox = null;
       
        // Assignment from index for performance.
        //     See http://www.peachpit.com/articles/article.aspx?p=31567&seqNum=5
        for ($i = 0, $len = sizeof($this->parentNodes); $i < $len; $i++) {
            $parentNode      = $this->parentNodes[$i];
            $contentScore    = intval($parentNode->getAttribute(Readability::ATTR_CONTENT_SCORE));
            $orgContentScore = intval($topBox ? $topBox->getAttribute(Readability::ATTR_CONTENT_SCORE) : 0);

            if ($contentScore && $contentScore > $orgContentScore) {
                $topBox = $parentNode;
            }
        }
       
        // 此时，$topBox 应为已经判定后的页面内容主元素
        return $topBox;
    }

    /**
     * 获取 HTML 页面标题
     *
     * @return String
     */
    public function getTitle() {
        $split_point = ' - ';
        $titleNodes = $this->DOM->getElementsByTagName("title");

        if ($titleNodes->length
            && $titleNode = $titleNodes->item(0)) {
            // @see http://stackoverflow.com/questions/717328/how-to-explode-string-right-to-left
            $title  = trim($titleNode->nodeValue);
            $result = array_map('strrev', explode($split_point, strrev($title)));
            return sizeof($result) > 1 ? array_pop($result) : $title;
        }

        return null;
    }

    /**
     * Get Leading Image Url
     *
     * @return String
     */
    public function getLeadImageUrl($node) {
        $images = $node->getElementsByTagName("img");

        if ($images->length && $leadImage = $images->item(0)) {
            return $leadImage->getAttribute("src");
        }

        return null;
    }

    /**
     * 获取页面的主要内容（Readability 以后的内容）
     *
     * @return Array
     */
    public function getContent() {
        if (!$this->DOM) return false;

        // 获取页面标题
        $ContentTitle = $this->getTitle();

        // 获取页面主内容
        $ContentBox = $this->getTopBox();
       
        //Check if we found a suitable top-box.
        if($ContentBox === null)
            throw new RuntimeException(Readability::MESSAGE_CAN_NOT_GET);
       
        // 复制内容到新的 DOMDocument
        $Target = new DOMDocument;
        $Target->appendChild($Target->importNode($ContentBox, true));

        // 删除不需要的标签
        foreach ($this->junkTags as $tag) {
            $Target = $this->removeJunkTag($Target, $tag);
        }

        // 删除不需要的属性
        foreach ($this->junkAttrs as $attr) {
            $Target = $this->removeJunkAttr($Target, $attr);
        }

        $content = mb_convert_encoding($Target->saveHTML(), Readability::DOM_DEFAULT_CHARSET, "HTML-ENTITIES");

        // 多个数据，以数组的形式返回
        return Array(
            'lead_image_url' => $this->getLeadImageUrl($Target),
            'word_count' => mb_strlen(strip_tags($content), Readability::DOM_DEFAULT_CHARSET),
            'title' => $ContentTitle ? $ContentTitle : null,
            'content' => $content
        );
    }

    function __destruct() { }
}

Array
(
    [1] => 第十五课.ass
    [2] => 第二十一课.ass
    [3] => 第二十课.ass
    [4] => 第九课.ass
    [5] => 第二十六课.ass
    [6] => 第四课.ass
    [7] => 第十七课.ass
    [8] => 第二十二课.ass
    [9] => 第七课.ass
    [10] => 第三课.ass
    [11] => 第十八课.ass
)

function getCombinationToString($arr,$m){
    $result = array();
    if ($m ==1){
       return $arr;
    }
   
    if ($m == count($arr)){
        $result[] = implode(',' , $arr);
        return $result;
    }
       
    $temp_firstelement = $arr[0];
    unset($arr[0]);
    $arr = array_values($arr);
    $temp_list1 = getCombinationToString($arr, ($m-1));
   
    foreach ($temp_list1 as $s){
        $s = $temp_firstelement.','.$s;
        $result[] = $s;
    }
    unset($temp_list1);

    $temp_list2 = getCombinationToString($arr, $m);
    foreach ($temp_list2 as $s){
        $result[] = $s;
    }   
    unset($temp_list2);
   
    return $result;
}
$arr = array(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18);
$t = getCombinationToString($arr, 6);

var_dump($t);

function getCombinAryByNum( $arr, $num,$t=array()) {
  if ($num == 0) {
    return array($t);
  }
  $r = array();
  for ($i=0,$l=count($arr); $i <= $l-$num; $i++) {
    $tmp = getCombinAryByNum( array_slice($arr, $i+1, $l, false), $num-1,array_merge($t, array($arr[$i])));
    $r = array_merge($r, $tmp);
  }
  return $r;
}

$arr = array(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18);
$numum = 6;
$ss = getCombinAryByNum($arr,$numum);

var_dump($ss);

// 获取评论次数,参数是文章ID
function getCommCount($postid)
{
 $jsondata = file_get_contents("http://api.duoshuo.com/threads/counts.json?short_name=i94web&threads=$postid");
 // 设置true返回数组,不设置或者是false则返回对象
 $resjson= json_decode($jsondata,true);
 return $resjson['response'][$postid]['comments'];
}
 

function get_url ($url,$cookie=false)
{
$url = parse_url($url);
$query = $url[path].'?'.$url[query];
echo 'Query:'.$query;
$fp = fsockopen( $url[host], $url[port]?$url[port]:80 , $errno, $errstr, 30);
if (!$fp) {
return false;
} else {
$request = 'GET $query HTTP/1.1rn';
$request .= 'Host: $url[host]rn';
$request .= 'Connection: Closern';
if($cookie) $request.='Cookie:  $cookien';
$request.='rn';
fwrite($fp,$request);
while(!@feof($fp)) {
$result .= @fgets($fp, 1024);
}
fclose($fp);
return $result;
}
}
//获取url的html部分，去掉header
function GetUrlHTML($url,$cookie=false)
{
$rowdata = get_url($url,$cookie);
if($rowdata)
{
$body= stristr($rowdata,'rnrn');
$body=substr($body,4,strlen($body));
return $body;
}
 
return false;
}
  

$ch = curl_init();
$timeout = 5;
curl_setopt ($ch, CURLOPT_URL, ‘http://www.111cn.net/');
curl_setopt ($ch, CURLOPT_RETURNTRANSFER, 1);
curl_setopt ($ch, CURLOPT_CONNECTTIMEOUT, $timeout);
$file_contents = curl_exec($ch);
curl_close($ch);
 
echo $file_contents;