分享

基于HttpClient实现网络爬虫~以百度新闻为例

问题导读
1.如何通过浏览器获得http请求?
2.http的请求的代码实现是什么?
3.百度新闻是如何实现这一爬虫的?













1 序言

在以前的工作中,实现过简单的网络爬虫,没有系统的介绍过,这篇博客就系统的介绍以下如何使用java的HttpClient实现网络爬虫。
关于网络爬虫的一些理论知识、实现思想以及策略问题,可以参考百度百科“网络爬虫”,那里已经介绍的十分详细,这里也不再啰嗦,下面就主要介绍如何去实现。


2 Http请求的实现

http请求:
      代码开始之前,还是首先介绍以下如何通过浏览器获取http请求信息,这一步是分析网站资源的第一步。在浏览器界面右键有“审查元素”这一功能(如果没找到,F12一样可以的),谷歌浏览器效果如下:

20141107133656984.png

    点击“审查元素”之后会出现如下界面:

20141107133610920.png


    其中的Network栏目是做爬虫应该重点关注的,打开会看到当前网页所有的http请求信息,如下图:

20141107133705921.png

    单击每个信息,可以看到http请求的详细信息,如下图所示:

20141107133619407.png

    通过程序伪装成浏览器请求的时候,就多需要关注Request Headers里面的信息,还有一些需要登录的网站也是需要关注这些的。Response里面的信息就是服务器返回的内容,这里只做对文本信息的处理,对图片、音频、视频等信息不做介绍。
    Response里面就包含这我们爬虫想获取的信息内容。如果里面的格式不好看的话,可以在浏览器中输入该http请求的url地址,然后右键-->查看网页源代码的形式查看相关信息。通过分析网页源代码中的字符串,总结出统一的规则,提取相应的文本信息。

代码实现:

    1 CrawlBase类,模拟http请求的基类

[mw_shl_code=java,true]
    /**
    *@Description: 获取网页信息基类
    */
    package com.lulei.crawl;
    import java.io.BufferedReader;
    import java.io.ByteArrayInputStream;
    import java.io.IOException;
    import java.io.InputStream;
    import java.io.InputStreamReader;
    import java.util.HashMap;
    import java.util.Iterator;
    import java.util.Map;
    import java.util.Map.Entry;
    import org.apache.commons.httpclient.Header;
    import org.apache.commons.httpclient.HttpClient;
    import org.apache.commons.httpclient.HttpException;
    import org.apache.commons.httpclient.HttpMethod;
    import org.apache.commons.httpclient.HttpStatus;
    import org.apache.commons.httpclient.methods.GetMethod;
    import org.apache.commons.httpclient.methods.PostMethod;
    import org.apache.log4j.Logger;
    import com.lulei.util.CharsetUtil;
    public abstract class CrawlBase {
        private static Logger log = Logger.getLogger(CrawlBase.class);
        //链接源代码
        private String pageSourceCode = "";
        //返回头信息
        private Header[] responseHeaders = null;
        //连接超时时间
        private static int connectTimeout = 3500;
        //连接读取时间
        private static int readTimeout = 3500;
        //默认最大访问次数
        private static int maxConnectTimes = 3;
        //网页默认编码方式
        private static String charsetName = "iso-8859-1";
        private static HttpClient httpClient = new HttpClient();
        static {
            httpClient.getHttpConnectionManager().getParams().setConnectionTimeout(connectTimeout);
            httpClient.getHttpConnectionManager().getParams().setSoTimeout(readTimeout);
        }
        /**
         * @param urlStr
         * @param charsetName
         * @param method
         * @param params
         * @return
         * @throws HttpException
         * @throws IOException
         * @Author: lulei
         * @Description: method方式访问页面
         */
        public boolean readPage(String urlStr, String charsetName, String method, HashMap<String, String> params) throws HttpException, IOException {
            if ("post".equals(method) || "POST".equals(method)) {
                return readPageByPost(urlStr, charsetName, params);
            } else {
                return readPageByGet(urlStr, charsetName, params);
            }
        }
        /**
         * @param urlStr
         * @param charsetName
         * @param params
         * @return 访问是否成功
         * @throws HttpException
         * @throws IOException
         * @Author: lulei
         * @Description: Get方式访问页面
         */
        public boolean readPageByGet(String urlStr, String charsetName, HashMap<String, String> params) throws HttpException, IOException {
            GetMethod getMethod = createGetMethod(urlStr, params);
            return readPage(getMethod, charsetName, urlStr);
        }
        /**
         * @param urlStr
         * @param charsetName
         * @param params
         * @return 访问是否成功
         * @throws HttpException
         * @throws IOException
         * @Author: lulei
         * @Description: Post方式访问页面
         */
        public boolean readPageByPost(String urlStr, String charsetName, HashMap<String, String> params) throws HttpException, IOException{
            PostMethod postMethod = createPostMethod(urlStr, params);
            return readPage(postMethod, charsetName, urlStr);
        }
        /**
         * @param method
         * @param defaultCharset
         * @param urlStr
         * @return 访问是否成功
         * @throws HttpException
         * @throws IOException
         * @Author: lulei
         * @Description: 读取页面信息和头信息
         */
        private boolean readPage(HttpMethod method, String defaultCharset, String urlStr) throws HttpException, IOException{
            int n = maxConnectTimes;
            while (n > 0) {
                try {
                    if (httpClient.executeMethod(method) != HttpStatus.SC_OK){
                        log.error("can not connect " + urlStr + "\t" + (maxConnectTimes - n + 1) + "\t" + httpClient.executeMethod(method));
                        n--;
                    } else {
                        //获取头信息
                        responseHeaders = method.getResponseHeaders();
                        //获取页面源代码
                        InputStream inputStream = method.getResponseBodyAsStream();
                        BufferedReader bufferedReader = new BufferedReader(new InputStreamReader(inputStream, charsetName));
                        StringBuffer stringBuffer = new StringBuffer();
                        String lineString = null;
                        while ((lineString = bufferedReader.readLine()) != null){
                            stringBuffer.append(lineString);
                            stringBuffer.append("\n");
                        }
                        pageSourceCode = stringBuffer.toString();
                        InputStream in =new  ByteArrayInputStream(pageSourceCode.getBytes(charsetName));
                        String charset = CharsetUtil.getStreamCharset(in, defaultCharset);
                        //下面这个判断是为了IP归属地查询特意加上去的
                        if ("Big5".equals(charset)) {
                            charset = "gbk";
                        }
                        if (!charsetName.toLowerCase().equals(charset.toLowerCase())) {
                            pageSourceCode = new String(pageSourceCode.getBytes(charsetName), charset);
                        }
                        return true;
                    }
                } catch (Exception e) {
                    e.printStackTrace();
                    System.out.println(urlStr + " -- can't connect  " + (maxConnectTimes - n + 1));
                    n--;
                }
            }
            return false;
        }
        /**
         * @param urlStr
         * @param params
         * @return GetMethod
         * @Author: lulei
         * @Description: 设置get请求参数
         */
        @SuppressWarnings("rawtypes")
        private GetMethod createGetMethod(String urlStr, HashMap<String, String> params){
            GetMethod getMethod = new GetMethod(urlStr);
            if (params == null){
                return getMethod;
            }
            Iterator iter = params.entrySet().iterator();
            while (iter.hasNext()) {
                Map.Entry entry = (Map.Entry) iter.next();
                String key = (String) entry.getKey();
                String val = (String) entry.getValue();
                getMethod.setRequestHeader(key, val);
            }
            return getMethod;
        }
        /**
         * @param urlStr
         * @param params
         * @return PostMethod
         * @Author: lulei
         * @Description: 设置post请求参数
         */
        private PostMethod createPostMethod(String urlStr, HashMap<String, String> params){
            PostMethod postMethod = new PostMethod(urlStr);
            if (params == null){
                return postMethod;
            }
            Iterator<Entry<String, String>> iter = params.entrySet().iterator();
            while (iter.hasNext()) {
                Map.Entry<String, String> entry =  iter.next();
                String key = (String) entry.getKey();
                String val = (String) entry.getValue();
                postMethod.setParameter(key, val);
            }
            return postMethod;
        }
        /**
         * @param urlStr
         * @param charsetName
         * @return 访问是否成功
         * @throws IOException
         * @Author: lulei
         * @Description: 不设置任何头信息直接访问网页
         */
        public boolean readPageByGet(String urlStr, String charsetName) throws IOException{
            return this.readPageByGet(urlStr, charsetName, null);
        }
        /**
         * @return String
         * @Author: lulei
         * @Description: 获取网页源代码
         */
        public String getPageSourceCode(){
            return pageSourceCode;
        }
        /**
         * @return Header[]
         * @Author: lulei
         * @Description: 获取网页返回头信息
         */
        public Header[] getHeader(){
            return responseHeaders;
        }
        /**
         * @param timeout
         * @Author: lulei
         * @Description: 设置连接超时时间
         */
        public void setConnectTimeout(int timeout){
            httpClient.getHttpConnectionManager().getParams().setConnectionTimeout(timeout);
        }
        /**
         * @param timeout
         * @Author: lulei
         * @Description: 设置读取超时时间
         */
        public void setReadTimeout(int timeout){
            httpClient.getHttpConnectionManager().getParams().setSoTimeout(timeout);
        }
        /**
         * @param maxConnectTimes
         * @Author: lulei
         * @Description: 设置最大访问次数,链接失败的情况下使用
         */
        public static void setMaxConnectTimes(int maxConnectTimes) {
            CrawlBase.maxConnectTimes = maxConnectTimes;
        }
        /**
         * @param connectTimeout
         * @param readTimeout
         * @Author: lulei
         * @Description: 设置连接超时时间和读取超时时间
         */
        public void setTimeout(int connectTimeout, int readTimeout){
            setConnectTimeout(connectTimeout);
            setReadTimeout(readTimeout);
        }
        /**
         * @param charsetName
         * @Author: lulei
         * @Description: 设置默认编码方式
         */
        public static void setCharsetName(String charsetName) {
            CrawlBase.charsetName = charsetName;
        }
    }
[/mw_shl_code]

    2 CrawlListPageBase类是CrawlBase的子类,实现了从页面中获取链接的URL信息基类

[mw_shl_code=java,true]
/**
  *@Description: 获取页面链接地址信息基类
*/
package com.lulei.crawl;
import java.io.IOException;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import com.lulei.util.DoRegex;
public abstract class CrawlListPageBase extends CrawlBase {
    private String pageurl;
    /**
    * @param urlStr
    * @param charsetName
    * @throws IOException
     */
    public CrawlListPageBase(String urlStr, String charsetName) throws IOException{
        readPageByGet(urlStr, charsetName);
        pageurl = urlStr;
    }
    /**
    * @param urlStr
    * @param charsetName
    * @param method
    * @param params
    * @throws IOException
     */
    public CrawlListPageBase(String urlStr, String charsetName, String method, HashMap<String, String> params) throws IOException{
        readPage(urlStr, charsetName, method, params);
        pageurl = urlStr;
    }
    /**
     * @return List<String>
     * @Author: lulei
     * @Description: 返回页面上需求的链接地址
     */
    public List<String> getPageUrls(){
        List<String> pageUrls = new ArrayList<String>();
        pageUrls = DoRegex.getArrayList(getPageSourceCode(), getUrlRegexString(), pageurl, getUrlRegexStringNum());
        return pageUrls;
    }
    /**
     * @return String
     * @Author: lulei
     * @Description: 返回页面上需求的网址连接的正则表达式
     */
    public abstract String getUrlRegexString();
    /**
     * @return int
     * @Author: lulei
     * @Description: 正则表达式中要去的字段位置
     */
    public abstract int getUrlRegexStringNum();
}
[/mw_shl_code]

    3 DoRegex类,封装的一些基于正则表达式字符串匹配查找类

[mw_shl_code=java,true]
    /**
    * @Description: 正则处理工具
    */
    package com.lulei.util;
    import java.io.UnsupportedEncodingException;
    import java.net.URLEncoder;
    import java.util.ArrayList;
    import java.util.List;
    import java.util.regex.Matcher;
    import java.util.regex.Pattern;
    public class DoRegex {
        private static String rootUrlRegex = "(http://.*?/)";
        private static String currentUrlRegex = "(http://.*/)";
        private static String ChRegex = "([\u4e00-\u9fa5]+)";
        /**
         * @param dealStr
         * @param regexStr
         * @param splitStr
         * @param n
         * @return String
         * @Author: lulei
         * @Description: 正则匹配结果,每条记录用splitStr分割
         */
        public static String getString(String dealStr, String regexStr, String splitStr, int n){
            String reStr = "";
            if (dealStr == null || regexStr == null || n < 1 || dealStr.isEmpty()){
                return reStr;
            }
            splitStr = (splitStr == null) ? "" : splitStr;
            Pattern pattern = Pattern.compile(regexStr, Pattern.CASE_INSENSITIVE | Pattern.DOTALL);
            Matcher matcher = pattern.matcher(dealStr);
            StringBuffer stringBuffer = new StringBuffer();
            while (matcher.find()) {
                stringBuffer.append(matcher.group(n).trim());
                stringBuffer.append(splitStr);
            }
            reStr = stringBuffer.toString();
            if (splitStr != "" && reStr.endsWith(splitStr)){
                reStr = reStr.substring(0, reStr.length() - splitStr.length());
            }
            return reStr;
        }
        /**
         * @param dealStr
         * @param regexStr
         * @param n
         * @return String
         * @Author: lulei
         * @Description: 正则匹配结果,将所有匹配记录组装成字符串
         */
        public static String getString(String dealStr, String regexStr, int n){
            return getString(dealStr, regexStr, null, n);
        }
        /**
         * @param dealStr
         * @param regexStr
         * @param n
         * @return String
         * @Author: lulei
         * @Description: 正则匹配第一条结果
         */
        public static String getFirstString(String dealStr, String regexStr, int n){
            if (dealStr == null || regexStr == null || n < 1 || dealStr.isEmpty()){
                return "";
            }
            Pattern pattern = Pattern.compile(regexStr, Pattern.CASE_INSENSITIVE | Pattern.DOTALL);
            Matcher matcher = pattern.matcher(dealStr);
            while (matcher.find()) {
                return matcher.group(n).trim();
            }
            return "";
        }
        /**
         * @param dealStr
         * @param regexStr
         * @param n
         * @return ArrayList<String>
         * @Author: lulei
         * @Description: 正则匹配结果,将匹配结果组装成数组
         */
        public static List<String> getList(String dealStr, String regexStr, int n){
            List<String> reArrayList = new ArrayList<String>();
            if (dealStr == null || regexStr == null || n < 1 || dealStr.isEmpty()){
                return reArrayList;
            }
            Pattern pattern = Pattern.compile(regexStr, Pattern.CASE_INSENSITIVE | Pattern.DOTALL);
            Matcher matcher = pattern.matcher(dealStr);
            while (matcher.find()) {
                reArrayList.add(matcher.group(n).trim());
            }
            return reArrayList;
        }
        /**
         * @param url
         * @param currentUrl
         * @return String
         * @Author: lulei
         * @Description: 组装网址,网页的url
         */
        private static String getHttpUrl(String url, String currentUrl){
            try {
                url = encodeUrlCh(url);
            } catch (UnsupportedEncodingException e) {
                // TODO Auto-generated catch block
                e.printStackTrace();
            }
            if (url.indexOf("http") == 0){
                return url;
            }
            if  (url.indexOf("/") == 0){
                return getFirstString(currentUrl, rootUrlRegex, 1) + url.substring(1);
            }
            return getFirstString(currentUrl, currentUrlRegex, 1) + url;
        }
        /**
         * @param dealStr
         * @param regexStr
         * @param currentUrl
         * @param n
         * @return ArrayList<String>
         * @Author: lulei
         * @Description: 获取和正则匹配的绝对链接地址
         */
        public static List<String> getArrayList(String dealStr, String regexStr, String currentUrl, int n){
            List<String> reArrayList = new ArrayList<String>();
            if (dealStr == null || regexStr == null || n < 1 || dealStr.isEmpty()){
                return reArrayList;
            }
            Pattern pattern = Pattern.compile(regexStr, Pattern.CASE_INSENSITIVE | Pattern.DOTALL);
            Matcher matcher = pattern.matcher(dealStr);
            while (matcher.find()) {
                reArrayList.add(getHttpUrl(matcher.group(n).trim(), currentUrl));
            }
            return reArrayList;
        }
        /**
         * @param url
         * @return
         * @throws UnsupportedEncodingException
         * @Author: lulei
         * @Description: 将连接地址中的中文进行编码处理
         */
        public static String encodeUrlCh (String url) throws UnsupportedEncodingException {
            while (true) {
                String s = getFirstString(url, ChRegex, 1);
                if ("".equals(s)){
                    return url;
                }
                url = url.replaceAll(s, URLEncoder.encode(s, "utf-8"));
            }
        }
        /**
         * @param dealStr
         * @param regexStr
         * @param array 正则位置数组
         * @return
         * @Author:lulei
         * @Description: 获取全部
         */
        public static List<String[]> getListArray(String dealStr, String regexStr, int[] array) {
            List<String[]> reArrayList = new ArrayList<String[]>();
            if (dealStr == null || regexStr == null || array == null) {
                return reArrayList;
            }
            for (int i = 0; i < array.length; i++) {
                if (array < 1) {
                    return reArrayList;
                }
            }
            Pattern pattern = Pattern.compile(regexStr, Pattern.CASE_INSENSITIVE | Pattern.DOTALL);
            Matcher matcher = pattern.matcher(dealStr);
            while (matcher.find()) {
                String[] ss = new String[array.length];
                for (int i = 0; i < array.length; i++) {
                    ss = matcher.group(array).trim();
                }
                reArrayList.add(ss);
            }
            return reArrayList;
        }
        /**
         * @param dealStr
         * @param regexStr
         * @param array
         * @return
         * @Author:lulei
         * @Description: 获取全部
         */
        public static List<String> getStringArray(String dealStr, String regexStr, int[] array) {
            List<String> reStringList = new ArrayList<String>();
            if (dealStr == null || regexStr == null || array == null) {
                return reStringList;
            }
            for (int i = 0; i < array.length; i++) {
                if (array < 1) {
                    return reStringList;
                }
            }
            Pattern pattern = Pattern.compile(regexStr, Pattern.CASE_INSENSITIVE | Pattern.DOTALL);
            Matcher matcher = pattern.matcher(dealStr);
            while (matcher.find()) {
                StringBuffer sb = new StringBuffer();
                for (int i = 0; i < array.length; i++) {
                    sb.append(matcher.group(array).trim());
                }
                reStringList.add(sb.toString());
            }
            return reStringList;
        }
        /**
         * @param dealStr
         * @param regexStr
         * @param array 正则位置数组
         * @return
         * @Author:lulei
         * @Description: 获取第一个
         */
        public static String[] getFirstArray(String dealStr, String regexStr, int[] array) {
            if (dealStr == null || regexStr == null || array == null) {
                return null;
            }
            for (int i = 0; i < array.length; i++) {
                if (array < 1) {
                    return null;
                }
            }
            Pattern pattern = Pattern.compile(regexStr, Pattern.CASE_INSENSITIVE | Pattern.DOTALL);
            Matcher matcher = pattern.matcher(dealStr);
            while (matcher.find()) {
                String[] ss = new String[array.length];
                for (int i = 0; i < array.length; i++) {
                    ss = matcher.group(array).trim();
                }
                return ss;
            }
            return null;
        }
    }
[/mw_shl_code]

    4 CharsetUtil类,编码方式检测类

[mw_shl_code=java,true]
    /**
    *@Description:  编码方式检测类
    */
    package com.lulei.util;
    import java.io.IOException;
    import java.io.InputStream;
    import java.net.URL;
    import java.nio.charset.Charset;
    import info.monitorenter.cpdetector.io.ASCIIDetector;
    import info.monitorenter.cpdetector.io.CodepageDetectorProxy;
    import info.monitorenter.cpdetector.io.JChardetFacade;
    import info.monitorenter.cpdetector.io.ParsingDetector;
    import info.monitorenter.cpdetector.io.UnicodeDetector;
    public class CharsetUtil {
        private static final CodepageDetectorProxy detector;
        static {//初始化探测器
            detector = CodepageDetectorProxy.getInstance();
            detector.add(new ParsingDetector(false));
            detector.add(ASCIIDetector.getInstance());
            detector.add(UnicodeDetector.getInstance());
            detector.add(JChardetFacade.getInstance());
        }
        /**
         * @param url
         * @param defaultCharset
         * @Author:lulei
         * @return 获取文件的编码方式
         */
        public static String getStreamCharset (URL url, String defaultCharset) {
            if (url == null) {
                return defaultCharset;
            }
            try {
                //使用第三方jar包检测文件的编码
                Charset charset = detector.detectCodepage(url);
                if (charset != null) {
                    return charset.name();
                }
            } catch (Exception e1) {
                // TODO Auto-generated catch block
                e1.printStackTrace();
            }
            return defaultCharset;
        }
        /**
         * @param inputStream
         * @param defaultCharset
         * @return
         * @Author:lulei
         * @Description: 获取文件流的编码方式
         */
        public static String getStreamCharset (InputStream inputStream, String defaultCharset) {
            if (inputStream == null) {
                return defaultCharset;
            }
            int count = 200;
            try {
                count = inputStream.available();
            } catch (IOException e) {
                // TODO Auto-generated catch block
                e.printStackTrace();
            }
            try {
                //使用第三方jar包检测文件的编码
                Charset charset = detector.detectCodepage(inputStream, count);
                if (charset != null) {
                    return charset.name();
                }
            } catch (Exception e1) {
                // TODO Auto-generated catch block
                e1.printStackTrace();
            }
            return defaultCharset;
        }
    }
[/mw_shl_code]

    上面四个类,就实现了网络文本资源信息抓取的基本架构,下面就通过一个实际的例子介绍如何使用上述类实现网络文本资源信息的获取。


3 百度新闻案例

    1)找到百度新闻更新列表页,如http://news.baidu.com/n?cmd=4&class=civilnews&pn=1&from=tab 界面如下图所示:

20141107133715031.png

   
文章URL链接地址如下图所示:

20141107133628252.png

   
通过对源文件的分析,编写BaiduNewList类,实现百度新闻列表页信息的抓取,代码如下:

[mw_shl_code=java,true]
/**
*@Description:   百度新闻滚动列表页,可以获取当前页面上的链接
*/
package com.lulei.crawl.news;
import java.io.IOException;
import java.util.HashMap;
import com.lulei.crawl.CrawlListPageBase;
public class BaiduNewList extends CrawlListPageBase{
    private static HashMap<String, String> params;
    /**
     * 添加相关头信息,对请求进行伪装
     */
    static {
        params = new HashMap<String, String>();
        params.put("Referer", "http://www.baidu.com");
        params.put("User-Agent", "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/36.0.1985.125 Safari/537.36");
    }
    public BaiduNewList(String urlStr) throws IOException {
        super(urlStr, "utf-8", "get", params);
    }
    @Override
    public String getUrlRegexString() {
        // TODO Auto-generated method stub
        //新闻列表页中文章链接地址的正则表达式
        return "&#8226; <a href=\"(.*?)\"";
    }
    @Override
    public int getUrlRegexStringNum() {
        // TODO Auto-generated method stub
        //链接地址在正则表达式中的位置
        return 1;
    }
    /**
     * @param args
     * @throws IOException
     * @Author:lulei
     * @Description:  测试用例
     */
    public static void main(String[] args) throws IOException {
        // TODO Auto-generated method stub
        BaiduNewList baidu = new BaiduNewList("http://news.baidu.com/n?cmd=4&class=sportnews&pn=1&from=tab");
        for (String s : baidu.getPageUrls()) {
            System.out.println(s);
        }
    }
}
[/mw_shl_code]

    2)通过第一步获取的URL,得到新闻所在的内容页面URL,由于百度新闻列表页面上的新闻来自不同的站,所以很难找到一个通用的结构,大多数的新闻类网站,内容都是放在p标签内,所以就采用了如下的方式获取新闻的内容,如下图:

20141107133632480.png


    News类具体实现如下所示:

[mw_shl_code=java,true]/**
*@Description:   新闻类网站新闻内容
*/
package com.lulei.crawl.news;
import java.io.IOException;
import java.util.HashMap;
import org.apache.commons.httpclient.HttpException;
import com.lulei.crawl.CrawlBase;
import com.lulei.util.DoRegex;
public class News extends CrawlBase{
    private String url;
    private String content;
    private String title;
    private String type;
    private static String contentRegex = "<p.*?>(.*?)</p>";
    private static String titleRegex = "<title>(.*?)</title>";
    private static int maxLength = 300;
    private static HashMap<String, String> params;
    /**
     * 添加相关头信息,对请求进行伪装
     */
    static {
        params = new HashMap<String, String>();
        params.put("Referer", "http://www.baidu.com");
        params.put("User-Agent", "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/36.0.1985.125 Safari/537.36");
    }
    /**
     * @Author:lulei
     * @Description: 默认p标签内的内容为正文,如果正文长度查过设置的最大长度,则截取前半部分
     */
    private void setContent() {
        String content = DoRegex.getString(getPageSourceCode(), contentRegex, 1);
        content = content.replaceAll("\n", "")
                                      .replaceAll("<script.*?/script>", "")
                                      .replaceAll("<style.*?/style>", "")
                                      .replaceAll("<.*?>", "");
        this.content = content.length() > maxLength ? content.substring(0, maxLength) : content;
    }
    /**
     * @Author:lulei
     * @Description: 默认title标签内的内容为标题
     */
    private void setTitle() {
        this.title = DoRegex.getString(getPageSourceCode(), titleRegex, 1);;
    }
    public News(String url) throws HttpException, IOException {
        this.url = url;
        readPageByGet(url, "utf-8", params);
        setContent();
        setTitle();
    }
    public String getUrl() {
        return url;
    }
    public void setUrl(String url) {
        this.url = url;
    }
    public String getContent() {
        return content;
    }
    public String getTitle() {
        return title;
    }
    public String getType() {
        return type;
    }
    public void setType(String type) {
        this.type = type;
    }
    public static void setMaxLength(int maxLength) {
        News.maxLength = maxLength;
    }
    /**
     * @param args
     * @throws HttpException
     * @throws IOException
     * @Author:lulei
     * @Description: 测试用例
     */
    public static void main(String[] args) throws HttpException, IOException {
        // TODO Auto-generated method stub
        News news = new News("http://we.sportscn.com/viewnews-1634777.html");
        System.out.println(news.getContent());
        System.out.println(news.getTitle());
    }
}
[/mw_shl_code]

    3)编写抓取的入口,这里为了简单,只做了两层的分析,所以新闻更新列表也的URL就直接写在程序中

如下图所示:

20141107133637050.png

    执行一次采集任务如下图所示:

20141107133651059.png


    在main函数里面只需要一次性或周期性的去执行run函数即可,具体代码如下:

[mw_shl_code=java,true]
    /**
    *@Description:
    */
    package com.lulei.knn.data;
    import java.io.IOException;
    import java.util.ArrayList;
    import java.util.HashMap;
    import java.util.List;
    import com.lulei.crawl.news.BaiduNewList;
    import com.lulei.crawl.news.News;
    import com.lulei.knn.index.KnnIndex;
    import com.lulei.knn.index.KnnSearch;
    import com.lulei.util.ParseMD5;
    public class CrawlNews {
        private static List<Info> infos;
        private static KnnIndex knnIndex = new KnnIndex();
        private static KnnSearch knnSearch = new KnnSearch();
        private static HashMap<String, Integer> result;
        static {
            infos = new ArrayList<Info>();
            infos.add(new Info("http://news.baidu.com/n?cmd=4&class=sportnews&pn=1&from=tab", "体育类"));
            infos.add(new Info("http://news.baidu.com/n?cmd=4&class=sportnews&pn=2&from=tab", "体育类"));
            infos.add(new Info("http://news.baidu.com/n?cmd=4&class=sportnews&pn=3&from=tab", "体育类"));
            infos.add(new Info("http://news.baidu.com/n?cmd=4&class=mil&pn=1&sub=0", "军事类"));
            infos.add(new Info("http://news.baidu.com/n?cmd=4&class=mil&pn=2&sub=0", "军事类"));
            infos.add(new Info("http://news.baidu.com/n?cmd=4&class=mil&pn=3&sub=0", "军事类"));
            infos.add(new Info("http://news.baidu.com/n?cmd=4&class=finannews&pn=1&sub=0", "财经类"));
            infos.add(new Info("http://news.baidu.com/n?cmd=4&class=finannews&pn=2&sub=0", "财经类"));
            infos.add(new Info("http://news.baidu.com/n?cmd=4&class=finannews&pn=3&sub=0", "财经类"));
            infos.add(new Info("http://news.baidu.com/n?cmd=4&class=internet&pn=1&from=tab", "互联网"));
            infos.add(new Info("http://news.baidu.com/n?cmd=4&class=housenews&pn=1&sub=0", "房产类"));
            infos.add(new Info("http://news.baidu.com/n?cmd=4&class=housenews&pn=2&sub=0", "房产类"));
            infos.add(new Info("http://news.baidu.com/n?cmd=4&class=housenews&pn=3&sub=0", "房产类"));
            infos.add(new Info("http://news.baidu.com/n?cmd=4&class=gamenews&pn=1&sub=0", "游戏类"));
            infos.add(new Info("http://news.baidu.com/n?cmd=4&class=gamenews&pn=2&sub=0", "游戏类"));
            infos.add(new Info("http://news.baidu.com/n?cmd=4&class=gamenews&pn=3&sub=0", "游戏类"));
        }
        /**
         *@Description:  抓取网址信息
         *@Author:lulei
         */
        static class Info{
            String url;
            String type;
            Info(String url, String type) {
                this.url = url;
                this.type = type;
            }
        }
        /**
         * @param info
         * @Author:lulei
         * @Description: 抓取一个列表页面下的新闻信息
         */
        private void crawl(Info info) {
            if (info == null) {
                return;
            }
            try {
                BaiduNewList baiduNewList = new BaiduNewList(info.url);
                List<String> urls = baiduNewList.getPageUrls();
                for (String url : urls) {
                    News news = new News(url);
                    NewsBean newBean = new NewsBean();
                    newBean.setId(ParseMD5.parseStrToMd5L32(url));
                    newBean.setType(info.type);
                    newBean.setUrl(url);
                    newBean.setTitle(news.getTitle());
                    newBean.setContent(news.getContent());
                    //保存到索引文件中
                    knnIndex.add(newBean);
                    //knn验证
                    if (news.getContent() == null || "".equals(news.getContent())) {
                        result.put("E", 1+result.get("E"));
                        continue;
                    }
                    if (info.type.equals(knnSearch.getType(news.getContent()))) {
                        result.put("R", 1+result.get("R"));
                    } else {
                        result.put("W", 1+result.get("W"));
                    }
                }
            } catch (Exception e) {
                e.printStackTrace();
            }
        }
        /**
         * @Author:lulei
         * @Description: 启动入口
         */
        public void run() {
            result = new HashMap<String, Integer>();
            result.put("R", 0);
            result.put("W", 0);
            result.put("E", 0);
            for (Info info : infos) {
                System.out.println(info.url + "------start");
                crawl(info);
                System.out.println(info.url + "------end");
            }
            try {
                knnIndex.commit();
                System.out.println("R = " + result.get("R"));
                System.out.println("W = " + result.get("W"));
                System.out.println("E = " + result.get("E"));
                System.out.println("精确度:" + (result.get("R") * 1.0 / (result.get("R") + result.get("W"))));
                System.out.println("-------------finished---------------");
            } catch (IOException e) {
                e.printStackTrace();
            }
        }
        public static void main(String[] args) {
            new CrawlNews().run();
        }
    }
[/mw_shl_code]

到此为止,一个完整的采集程序就完成了。

来源:www.llwjy.com小鸡慢慢

已有(3)人评论

跳转到指定楼层
strivecheng 发表于 2015-6-12 08:53:05
之前稍微弄过了一下httpclient
回复

使用道具 举报

hb1984 发表于 2015-6-12 17:15:34
连接都不关闭......
回复

使用道具 举报

tang 发表于 2015-6-20 10:45:25
回复

使用道具 举报

您需要登录后才可以回帖 登录 | 立即注册

本版积分规则

关闭

推荐上一条 /2 下一条