Java精确抽取网页发布时间

对网页中各种不同格式的发布时间进行抽取，将发布时间以规整的“yyyy-mm-dd hh:mm:ss”格式表示出来，只能尽量追求精确，但是因为网络发布时间的格式十分灵活，所以做不到百分百地正确抽取

package whu.extract.pubtime.core;
 
import java.util.arraylist;
import java.util.calendar;
import java.util.collections;
import java.util.list;
import java.util.regex.matcher;
import java.util.regex.pattern;
 
import whu.utils.timeutil;
 
/**
 * created on 2014年3月13日 下午2:49:05
 * @description 获取网页的发布时间 
 */
public class fetchpubtime {
  /** 表示url中连续的8位日期，例如http://www.baidu.com/20140311/2356.html */
  private static string url_reg_whole= "([-|/|_]{1}20\\d{6})";
  /** 表示 用-或者/隔开的日期,有年月日的，例如 http://www.baidu.com/2014-3-11/2356.html */
  private static string url_reg_sep_ymd = "([-|/|_]{1}20\\d{2}[-|/|_]{1}\\d{1,2}[-|/|_]{1}\\d{1,2})";
  /** 表示 用-或者/隔开的日期,只有年和月份的，例如 http://www.baidu.com/2014-3/2356.html */
  private static string url_reg_sep_ym = "([-|/|_]{1}20\\d{2}[-|/|_]{1}\\d{1,2})";
  private static calendar current = calendar.getinstance();
  /** 格式正确的时间正则表达式*/
  private static string righttimereg = "^((\\d{2}(([02468][048])|([13579][26]))[\\-\\/\\s]?((((0?[13578])|(1[02]))[\\-\\/\\s]?((0?[1-9])|([1-2][0-9])|(3[01])))|(((0?[469])|(11))[\\-\\/\\s]?((0?[1-9])|([1-2][0-9])|(30)))|(0?2[\\-\\/\\s]?((0?[1-9])|([1-2][0-9])))))|(\\d{2}(([02468][1235679])|([13579][01345789]))[\\-\\/\\s]?((((0?[13578])|(1[02]))[\\-\\/\\s]?((0?[1-9])|([1-2][0-9])|(3[01])))|(((0?[469])|(11))[\\-\\/\\s]?((0?[1-9])|([1-2][0-9])|(30)))|(0?2[\\-\\/\\s]?((0?[1-9])|(1[0-9])|(2[0-8]))))))(\\s(((0?[0-9])|([1-2][0-3]))\\:([0-5]?[0-9])((\\s)|(\\:([0-5]?[0-9])))))?$";
   
  /**
   * @param url
   * @param urlcontent
   * @return
   */
  public static string getpubtimevarious(string url,string urlcontent) {
     
    string pubtime = getpubtimefromurl(url);
     
   //链接里面没有，匹配文本中的
    if(pubtime == null)
    {
      if(urlcontent!=null&&!urlcontent.trim().equals(""))
        return extractpagedate(urlcontent);
    }
     
    return pubtime;
  }
   
  /**从url里面抽取出发布时间，返回yyyy-mm-dd hh:mm:ss格式的字符串
   * @param url
   * @return
   */
  public static string getpubtimefromurl(string url)
  {
    pattern p_whole = pattern.compile(url_reg_whole);
    matcher m_whole = p_whole.matcher(url);
    if(m_whole.find(0)&&m_whole.groupcount()>0)
    {
      string time = m_whole.group(0);
      time = time.substring(1,time.length());
      //每一步都不能够超出当前时间     
    if(current.compareto(timeutil.strtocalendar(time, "yyyymmdd"))>=0)
    {
 
      return time.substring(0,4)+"-"+time.substring(4,6)+"-"+
         time.substring(6,8)+" "+"00:00:00";
    }
    }
    
    p_whole = null;
    m_whole = null;
    pattern p_sep = pattern.compile(url_reg_sep_ymd);
    matcher m_sep = p_sep.matcher(url);
    if(m_sep.find(0)&&m_sep.groupcount()>0)
    {
       string time = m_sep.group(0);
       time = time.substring(1,time.length());
       string[] seg = time.split("[-|/|_]{1}");
       calendar thetime = calendar.getinstance();
       thetime.set(calendar.year,integer.parseint(seg[0]));
       thetime.set(calendar.month, integer.parseint(seg[1]));
       thetime.set(calendar.day_of_month, integer.parseint(seg[2]));
       if(current.compareto(thetime)>=0)
        {
       
      return seg[0]+"-"+seg[1]+"-"+seg[2]+" "+"00:00:00";
        }
    }
    p_sep = null;
    m_sep = null;
    pattern p_sep_ym = pattern.compile(url_reg_sep_ym);
    matcher m_sep_ym = p_sep_ym.matcher(url);
    if(m_sep_ym.find(0)&&m_sep_ym.groupcount()>0)
    {
       string time = m_sep_ym.group(0);
       time = time.substring(1,time.length());
       calendar thetime = calendar.getinstance();
       string[] seg = time.split("[-|/|_]{1}");
       thetime.set(calendar.year,integer.parseint(seg[0]));
       thetime.set(calendar.month, integer.parseint(seg[1]));
       thetime.set(calendar.day_of_month, 1);
       if(current.compareto(thetime)>=0)
      {
       
      return seg[0]+"-"+seg[1]+"-"+"01"+" "+"00:00:00";
      }
    }
     
    return null;
  }
   
 
  /** 从网页源码中取出发布时间
   * java中正则表达式提取字符串中日期实现代码
   * 2013年12月19日15:58:42
   * 读取出2013-12-19 15:48:33或者2013-12-19或者2012/3/05形式的时间
   * @param text 待提取的字符串
   * @return 返回日期
   * @author: oschina
   * @createtime: jan 21, 2013
   */
  public static string extractpagedate(string text) { 
    boolean containshms =false;
    string datestr = text.replaceall("r?n", " ");
    try { 
      list matches = null; 
      pattern p_detail = pattern.compile("(20\\d{2}[-/]\\d{1,2}[-/]\\d{1,2} \\d{1,2}:\\d{1,2}:\\d{1,2})|(20\\d{2}年\\d{1,2}月\\d{1,2}日)", pattern.case_insensitive|pattern.multiline); 
      //如果是仅仅抽取年月日，则按照上面的，如果是抽取年月日-时分秒，则按照下面的
      pattern p = pattern.compile("(20\\d{2}[-/]\\d{1,2}[-/]\\d{1,2})|(20\\d{2}年\\d{1,2}月\\d{1,2}日)", pattern.case_insensitive|pattern.multiline);
      //matcher matcher = p.matcher(datestr);
      matcher matcher_detail = p_detail.matcher(datestr);
       
      if(!(matcher_detail.find(0) && matcher_detail.groupcount() >= 1)) 
      {
        matcher_detail = p.matcher(datestr);
        containshms = true;
      }else
        matcher_detail = p_detail.matcher(datestr);
      if (matcher_detail.find() && matcher_detail.groupcount() >= 1) { 
        matches = new arraylist(); 
        for (int i = 1; i <= matcher_detail.groupcount(); i++) { 
          string temp = matcher_detail.group(i); 
          matches.add(temp); 
        } 
      } else { 
        matches = collections.empty_list; 
      }      
 
      if (matches.size() > 0) { 
        for(int i=0;i<matches.size();i++)
        {
          string pubtime = matches.get(i).tostring().trim();
          //取出第一个值
          pubtime = pubtime.replace("/", "-").replace("年", "-").replace("月", "-").replace("日", "-");
          if(current.compareto(timeutil.strtocalendar(pubtime, "yyyy-mm-dd"))>=0)
          {
            if(containshms)
              pubtime+=" "+"00:00:00";
            if(pubtime.matches(righttimereg))
            {
              return pubtime; 
            }
          }
        }
      } else { 
        return null; 
      } 
       
    } catch (exception e) { 
      return null; 
    } 
    return null;
  }
}

以上就是本文的全部内容，希望对大家学习java程序设计有所帮助。