Android Open Source - SimpleReader Html Filter

From Project

Back to project page SimpleReader.

License

The source code is released under:

Apache License

If you think the Android project SimpleReader listed in this page is inappropriate, such as containing malicious code/tools or violating the copyright, please email info at java2s dot com, thanks.

Java Source Code

package com.dreamteam.app.commons;
//from w  w  w  . j ava2  s .c om
import java.io.BufferedReader;
import java.io.IOException;
import java.io.InputStream;
import java.io.InputStreamReader;
import java.util.ArrayList;
import java.util.regex.Matcher;
import java.util.regex.Pattern;

/**
 * @description ??html
 * @author zcloud
 * @date 2013/11/14
 */
public class HtmlFilter
{
  //??????
  public static final String regexpForHtml = "<([^>]*)>";
  //??img??
  public static final String regexpForImg = "(<|;)\\s*(IMG|img)\\s+([^;^>]*)\\s*(;|>)";
  //????img???url
  public static final String regexpForImgUrl = "http://([^\"]+)\"";
  //??<>??style
  public static final String regexpForStyle = "\\s*style=\"([^\"]*)\"";
  //????encoding
  public static final String regexpForEncoding = "\\s*encoding=\"([^\"]*)\"";
  
  
  /**
   * @description
   * @param input
   * @return String
   */
  public static String filterHtml(String input)
  {
    Pattern pattern = Pattern.compile(regexpForHtml);
    Matcher matcher = pattern.matcher(input);
    StringBuffer sb = new StringBuffer();
    while(matcher.find())
    {
      matcher.appendReplacement(sb, "");
    }
    matcher.appendTail(sb);
    return sb.toString();
  }
  
  public static ArrayList<String> getImgTags(String input)
  {
    Pattern pattern = Pattern.compile(regexpForImg);
    Matcher matcher = pattern.matcher(input);
    ArrayList<String> imgTags = new ArrayList<String>();
    
    while(matcher.find())
    {
      imgTags.add(matcher.group());
    }
    return imgTags;
  }
  
  public static ArrayList<String> getImageSrcs(String input)
  {
    ArrayList<String> srcs = new ArrayList<String>();
    
    Pattern tagPattern = Pattern.compile(regexpForImg);
    Matcher tagMatcher = tagPattern.matcher(input);
    Pattern srcPattern = Pattern.compile(regexpForImgUrl);
    
    while(tagMatcher.find())
    {
      Matcher srcMatcher = srcPattern.matcher(tagMatcher.group());
      while(srcMatcher.find())
      {
        String src = srcMatcher.group().replace("\"", "");
        srcs.add(src);
      }
    }
    return srcs;
  }
  
  public static String getEncoding(InputStream is)
  {
    String encoding = null;
    InputStreamReader isr = new InputStreamReader(is);
    BufferedReader br = new BufferedReader(isr);
    String context;
    try
    {
      context = br.readLine();
      Pattern p = Pattern.compile(regexpForEncoding);
      Matcher m = p.matcher(context);
      if(m.find())
      {
        encoding = m.group();
      }
    } catch (IOException e)
    {
      e.printStackTrace();
    }
    return encoding.replace("encoding=", "").replace("\"", "");
  }
}

Java Source Code List