将HTML转化为TEXT的Java类

　　为了支持全文检索，有必要将HTML格式的文章转化为纯文本格式，因此我设计了一个基本的WebFormatter类，提供一个简单的public static String html2text(String html)，将HTML格式转化为Text：

/*
* File: WebFormatter.java
* Created on 2005-6-24
* Author: Liao Xuefeng, asklxf@163.com
* Copyright (C) 2005, Liao Xuefeng.
*/
package com.mboker.blog.web.util;

import java.util.*;
import java.text.SimpleDateFormat;

/**
* Do some format on web display.
*
* @author Xuefeng
*/
public class WebFormatter {

public static String html2text(String html) {
StringBuffer sb = new StringBuffer(html.length());
char[] data = html.toCharArray();
int start = 0;
boolean previousIsPre = false;
Token token = null;
for(;;) {
token = parse(data, start, previousIsPre);
if(token==null)
break;
previousIsPre = token.isPreTag();
sb = sb.append(token.getText());
start = token.getLength();
}
return sb.toString();
}

private static Token parse(char[] data, int start, boolean previousIsPre) {
if(start>=data.length)
return null;
// try to read next char:
char c = data[start];
if(c=='<') {
// this is a tag or comment or script:
int end_index = indexOf(data, start 1, '>');
if(end_index==(-1)) {
// the left is all text!
return new Token(Token.TOKEN_TEXT, data, start, data.length, previousIsPre);
}
String s = new String(data, start, end_index-start 1);
// now we got s="<...>":
if(s.startsWith("<!--")) { // this is a comment!
int end_comment_index = indexOf(data, start 1, "-->");
if(end_comment_index==(-1)) {
// illegal end, but treat as comment:
return new Token(Token.TOKEN_COMMENT, data, start, data.length, previousIsPre);
}
else
return new Token(Token.TOKEN_COMMENT, data, start, end_comment_index 3, previousIsPre);
}
String s_lowerCase = s.toLowerCase();
if(s_lowerCase.startsWith("<script")) { // this is a script:
int end_script_index = indexOf(data, start 1, "</script>");
if(end_script_index==(-1))
// illegal end, but treat as script:
return new Token(Token.TOKEN_SCRIPT, data, start, data.length, previousIsPre);
else
return new Token(Token.TOKEN_SCRIPT, data, start, end_script_index 9, previousIsPre);
}
else { // this is a tag:
return new Token(Token.TOKEN_TAG, data, start, start s.length(), previousIsPre);
}
}
// this is a text:
int next_tag_index = indexOf(data, start 1, '<');
if(next_tag_index==(-1))
return new Token(Token.TOKEN_TEXT, data, start, data.length, previousIsPre);
return new Token(Token.TOKEN_TEXT, data, start, next_tag_index, previousIsPre);
}

private static int indexOf(char[] data, int start, String s) {
char[] ss = s.toCharArray();
// TODO: performance can improve!
for(int i=start; i<(data.length-ss.length); i ) {
// compare from data[i] with ss[0]:
boolean match = true;
for(int j=0; j<ss.length; j ) {
if(data[i j]!=ss[j]) {
match = false;
break;
}
}
if(match)
return i;
}
return (-1);
}

private static int indexOf(char[] data, int start, char c) {
for(int i=start; i<data.length; i ) {
if(data[i]==c)
return i;
}
return (-1);
}

}

class Token {

public static final int TOKEN_TEXT = 0; // html text.
public static final int TOKEN_COMMENT = 1; // comment like <!-- comments... -->
public static final int TOKEN_TAG = 2; // tag like <pre>, <font>, etc.
public static final int TOKEN_SCRIPT = 3;

private static final char[] TAG_BR = "<br".toCharArray();
private static final char[] TAG_P = "<p".toCharArray();
private static final char[] TAG_LI = "<li".toCharArray();
private static final char[] TAG_PRE = "<pre".toCharArray();
private static final char[] TAG_HR = "<hr".toCharArray();

private static final char[] END_TAG_TD = "</td>".toCharArray();
private static final char[] END_TAG_TR = "</tr>".toCharArray();

标签：

版权申明：本站文章部分自网络，如有侵权，请联系：west999com@outlook.com
特别注意：本站所有转载文章言论不代表本站观点，本站所提供的摄影照片，插画，设计作品，如需使用，请与原作者联系，版权归原作者所有