java中文转繁体汉字
程序员文章站
2022-05-18 18:06:34
基本常用的中文汉字,和一些生僻字都包含在内。import java.io.BufferedReader;import java.io.FileNotFoundException;import java.io.IOException;import java.io.InputStream;import java.io.InputStreamReader;import java.util.HashMap;import java.util.HashSet;import java.util.Ite...
基本常用的中文汉字,和一些生僻字都包含在内。
import java.io.BufferedReader;
import java.io.FileNotFoundException;
import java.io.IOException;
import java.io.InputStream;
import java.io.InputStreamReader;
import java.util.HashMap;
import java.util.HashSet;
import java.util.Iterator;
import java.util.Map;
import java.util.Properties;
import java.util.Set;
public class ZHConverter {
private Properties charMap = new Properties();
private Set conflictingSets = new HashSet();
public static final int TRADITIONAL = 0;
public static final int SIMPLIFIED = 1;
private static final int NUM_OF_CONVERTERS = 2;
private static final ZHConverter[] converters = new ZHConverter[NUM_OF_CONVERTERS];
private static final String[] propertyFiles = new String[2];
static {
propertyFiles[TRADITIONAL] = "zh2Hant.properties";
propertyFiles[SIMPLIFIED] = "zh2Hans.properties";
}
/**
*
* @param converterType 0 for traditional and 1 for simplified
* @return
*/
public static ZHConverter getInstance(int converterType) {
if (converterType >= 0 && converterType < NUM_OF_CONVERTERS) {
if (converters[converterType] == null) {
synchronized(ZHConverter.class) {
if (converters[converterType] == null) {
converters[converterType] = new ZHConverter(propertyFiles[converterType]);
}
}
}
return converters[converterType];
} else {
return null;
}
}
public static String convert(String text, int converterType) {
ZHConverter instance = getInstance(converterType);
return instance.convert(text);
}
private ZHConverter(String propertyFile) {
InputStream is = null;
is = getClass().getResourceAsStream(propertyFile);
//File propertyFile = new File("C:/Temp/testMDB/TestTranslator/abc.txt");
if (is != null) {
BufferedReader reader = null;
try {
reader = new BufferedReader(new InputStreamReader(is));
charMap.load(reader);
} catch (FileNotFoundException e) {
} catch (IOException e) {
// TODO Auto-generated catch block
e.printStackTrace();
} finally {
try {
if (reader != null)
reader.close();
if (is != null)
is.close();
} catch (IOException e) {
}
}
}
initializeHelper();
}
private void initializeHelper() {
Map stringPossibilities = new HashMap();
Iterator iter = charMap.keySet().iterator();
while (iter.hasNext()) {
String key = (String) iter.next();
if (key.length() >= 1) {
for (int i = 0; i < (key.length()); i++) {
String keySubstring = key.substring(0, i + 1);
if (stringPossibilities.containsKey(keySubstring)) {
Integer integer = (Integer)(stringPossibilities.get(keySubstring));
stringPossibilities.put(keySubstring, new Integer(
integer.intValue() + 1));
} else {
stringPossibilities.put(keySubstring, new Integer(1));
}
}
}
}
iter = stringPossibilities.keySet().iterator();
while (iter.hasNext()) {
String key = (String) iter.next();
if (((Integer)(stringPossibilities.get(key))).intValue() > 1) {
conflictingSets.add(key);
}
}
}
public String convert(String in) {
StringBuilder outString = new StringBuilder();
StringBuilder stackString = new StringBuilder();
for (int i = 0; i < in.length(); i++) {
char c = in.charAt(i);
String key = "" + c;
stackString.append(key);
if (conflictingSets.contains(stackString.toString())) {
} else if (charMap.containsKey(stackString.toString())) {
outString.append(charMap.get(stackString.toString()));
stackString.setLength(0);
} else {
CharSequence sequence = stackString.subSequence(0, stackString.length()-1);
stackString.delete(0, stackString.length()-1);
flushStack(outString, new StringBuilder(sequence));
}
}
flushStack(outString, stackString);
return outString.toString();
}
private void flushStack(StringBuilder outString, StringBuilder stackString) {
while (stackString.length() > 0){
if (charMap.containsKey(stackString.toString())) {
outString.append(charMap.get(stackString.toString()));
stackString.setLength(0);
} else {
outString.append("" + stackString.charAt(0));
stackString.delete(0, 1);
}
}
}
String parseOneChar(String c) {
if (charMap.containsKey(c)) {
return (String) charMap.get(c);
}
return c;
}
public static void main(String[] args) throws IOException {
// 繁体转简体
ZHConverter converter = ZHConverter.getInstance(ZHConverter.SIMPLIFIED);
String simplifiedStr = converter.convert("繁體字轉化");
System.out.println(simplifiedStr);
// 简体转繁体
ZHConverter converter2 = ZHConverter.getInstance(ZHConverter.TRADITIONAL);
String traditionalStr = converter2.convert("繁体字转化");
System.out.println(traditionalStr);
}
}
本文需要使用文字库共两个 百度网盘链接:提取码: ve57
本文地址:https://blog.csdn.net/qq_32909681/article/details/107663345
上一篇: 广脸达笔试复盘7.29
下一篇: numpy中函数shape的用法