package com.avic.common.utils;
import java.util.regex.Pattern;
public class HtmlUtil {
//从html中提取纯文本
public static String Html2Text(String inputString) {
String htmlStr = inputString; // 含html标签的字符串
String textStr = "";
Pattern p_script;
java.util.regex.Matcher m_script;
Pattern p_style;
java.util.regex.Matcher m_style;
Pattern p_html;
java.util.regex.Matcher m_html;
try {
String regEx_script = "<[\\s]*?script[^>]*?>[\\s\\S]*?<[\\s]*?\\/[\\s]*?script[\\s]*?>"; // 定义script的正则表达式{或<script[^>]*?>[\\s\\S]*?<\\/script>
String regEx_style = "<[\\s]*?style[^>]*?>[\\s\\S]*?<[\\s]*?\\/[\\s]*?style[\\s]*?>"; // 定义style的正则表达式{或<style[^>]*?>[\\s\\S]*?<\\/style>
String regEx_html = "<[^>]+>"; // 定义HTML标签的正则表达式
p_script = Pattern.compile(regEx_script, Pattern.CASE_INSENSITIVE);
m_script = p_script.matcher(htmlStr);
htmlStr = m_script.replaceAll(""); // 过滤script标签
p_style = Pattern.compile(regEx_style, Pattern.CASE_INSENSITIVE);
m_style = p_style.matcher(htmlStr);
htmlStr = m_style.replaceAll(""); // 过滤style标签
p_html = Pattern.compile(regEx_html, Pattern.CASE_INSENSITIVE);
m_html = p_html.matcher(htmlStr);
htmlStr = m_html.replaceAll(""); // 过滤html标签
textStr = htmlStr;
} catch (Exception e) {
System.err.println("Html2Text: " + e.getMessage());
}
//剔除空格行
textStr = textStr.replaceAll("[ ]+", " ");
textStr = textStr.replaceAll("(?m)^\\s*$(\\n|\\r\\n)", "");
return textStr;// 返回文本字符串
}
/**
* 在固定位置插入字符串
*
* @param orginStr
* @param fixedPos
* @param segStr
* @return
*/
public static String fixedInsert(String orginStr, Integer fixedPos, String segStr) {
StringBuilder newStrBuilder = new StringBuilder(orginStr);
newStrBuilder.insert(fixedPos, segStr);
return newStrBuilder.toString();
}
/**
* 处理url segment部分
*
* @param oldUr http://localhost:8010/biaozhun/gjbz/d-2?attrpro=52__asdf&attrpro=start60__2018-11-21&attrpro=end60__2018-11-23
* @param flag
* @return
*/
public static String dealTimeSeg(String oldUr, String flag) {
if (countStr(oldUr, "attrpro") == 1) {
if (flag.equals("60")) {
oldUr = oldUr.replaceAll("[\\?]*attrpro=start60__[0-9]*-[0-9]*-[0-9]*", "");
oldUr = oldUr.replaceAll("[\\?]*attrpro=end60__[0-9]*-[0-9]*-[0-9]*", "");
} else if (flag.equals("61")) {
oldUr = oldUr.replaceAll("[\\?]*attrpro=start61__[0-9]*-[0-9]*-[0-9]*", "");
oldUr = oldUr.replaceAll("[\\?]*attrpro=end61__[0-9]*-[0-9]*-[0-9]*", "");
}
} else {
if (countStr(oldUr, "start60") == 1) {
oldUr = oldUr.replaceAll("[\\?|&]*attrpro=start60__[0-9]*-[0-9]*-[0-9]*", "");
oldUr = oldUr.replaceAll("[&]*attrpro=end60__[0-9]*-[0-9]*-[0-9]*", "");
oldUr = oldUr.replaceAll("[&]*attrpro=start61__[0-9]*-[0-9]*-[0-9]*", "");
oldUr = oldUr.replaceAll("[&]*attrpro=end61__[0-9]*-[0-9]*-[0-9]*", "");
} else if (countStr(oldUr, "end60") == 1) {
oldUr = oldUr.replaceAll("[&]*attrpro=start60__[0-9]*-[0-9]*-[0-9]*", "");
oldUr = oldUr.replaceAll("[\\?|&]*attrpro=end60__[0-9]*-[0-9]*-[0-9]*", "");
oldUr = oldUr.replaceAll("[&]*attrpro=start61__[0-9]*-[0-9]*-[0-9]*", "");
oldUr = oldUr.replaceAll("[&]*attrpro=end61__[0-9]*-[0-9]*-[0-9]*", "");
} else if (countStr(oldUr, "start61") == 1) {
oldUr = oldUr.replaceAll("[&]*attrpro=start60__[0-9]*-[0-9]*-[0-9]*", "");
oldUr = oldUr.replaceAll("[&]*attrpro=end60__[0-9]*-[0-9]*-[0-9]*", "");
oldUr = oldUr.replaceAll("[\\?|&]*attrpro=start61__[0-9]*-[0-9]*-[0-9]*", "");
oldUr = oldUr.replaceAll("[&]*attrpro=end61__[0-9]*-[0-9]*-[0-9]*", "");
} else if (countStr(oldUr, "end61") == 1) {
oldUr = oldUr.replaceAll("[&]*attrpro=start60__[0-9]*-[0-9]*-[0-9]*", "");
oldUr = oldUr.replaceAll("[&]*attrpro=end60__[0-9]*-[0-9]*-[0-9]*", "");
oldUr = oldUr.replaceAll("[&]*attrpro=start61__[0-9]*-[0-9]*-[0-9]*", "");
oldUr = oldUr.replaceAll("[\\?|&]*attrpro=end61__[0-9]*-[0-9]*-[0-9]*", "");
}
}
return oldUr;
}
public static String dealBookTimeSeg(String oldUr) {
if (countStr(oldUr, "attrpro") == 1) {
oldUr = oldUr.replaceAll("[\\?]*attrpro=start93__[0-9]*-[0-9]*-[0-9]*", "");
oldUr = oldUr.replaceAll("[\\?]*attrpro=end93__[0-9]*-[0-9]*-[0-9]*", "");
}else{
oldUr = oldUr.replaceAll("[\\?|&]*attrpro=start93__[0-9]*-[0-9]*-[0-9]*", "");
oldUr = oldUr.replaceAll("[\\?|&]*attrpro=end93__[0-9]*-[0-9]*-[0-9]*", "");
}
return oldUr;
}
private static int countStr(String orginStr, String sToFind) {
int num = 0;
while (orginStr.contains(sToFind)) {
orginStr = orginStr.substring(orginStr.indexOf(sToFind) + sToFind.length());
num++;
}
return num;
}
}