在python里,直接decode('utf-8')即可
>>> "\xE5\x85\x84\xE5\xBC\x9F\xE9\x9A\xBE\xE5\xBD\x93 \xE6\x9D\x9C\xE6\xAD\x8C".decode('utf-8')
u'\u5144\u5f1f\u96be\u5f53 \u675c\u6b4c'
>>> print "\xE5\x85\x84\xE5\xBC\x9F\xE9\x9A\xBE\xE5\xBD\x93 \xE6\x9D\x9C\xE6\xAD\x8C".decode('utf-8')
兄弟難當 杜歌
>>>
在java里未發現直接解碼的函數,不過只要理解了數據是如何編碼的,就可以很快的進行解碼,推薦閱讀http://www.ruanyifeng.com/blog/2007/10/ascii_unicode_and_utf-8.html
UTF-8是unicode編碼的一種落地方案:
Unicode符號范圍 | UTF-8編碼方式
(十六進制) | (二進制)
--------------------+---------------------------------------------
0000 0000-0000 007F | 0xxxxxxx
0000 0080-0000 07FF | 110xxxxx 10xxxxxx
0000 0800-0000 FFFF | 1110xxxx 10xxxxxx 10xxxxxx
0001 0000-0010 FFFF | 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
\x對應的是UTF-8編碼的數據,通過轉化規則可以轉換為Unicode編碼,就能得到對應的漢字,轉換規則很簡單,先將\x去掉,轉換為數字,然后進行對應的位移操作即可,需要注意的是先要判斷utf-8的位數:
val pattern = """(\d+\.\d+\.\d+\.\d+) \- (\S+) (\S+) \[([^\]]+)\] \"(\w+) (\S+) \S+\" (\S+) (\S+) \"([^\"]+)\" \"([^\"]+)\" \"([^\"]+)\" \"([^\"]+)""".r
val decodeDataPattern = """(\\x([0-9A-Z]){2})+""".r
def decodeUtf8(utf8Str:String):String={
var data = decodeDataPattern.replaceAllIn(utf8Str, m=>{
var item = decodeXdata(m.toString())
item
})
return data
}
def decodeXdata(utf8Str:String):String={
var arr = utf8Str.split("\\\\x")
var result = new StringBuilder()
var isMatchEnd = true
var matchIndex = 0
var currentWordLength = 0
var current = 0
var e0=0xe0;
for(item <-arr){
var str = item.trim
if(str.length()>0){
var currentCode = Integer.parseInt(str, 16);
if(isMatchEnd){
isMatchEnd = false
var and = currentCode & e0;
if(and == 0xe0){
matchIndex = 1;
currentWordLength = 3;
current = (currentCode & 0x1f) <<12 // 3位編碼的
}else if(and==96){
matchIndex = 1;
currentWordLength = 2;
current = (currentCode & 0x1f) <<6 // 2位編碼的
}else{
current = currentCode // 1位編碼的
}
}else{
matchIndex = matchIndex+1;
if(matchIndex == 2)
{
current+=(currentCode & 0x3f) <<6
}else{
current+=(currentCode & 0x3f)
}
}
if(matchIndex==currentWordLength){
var hex = Integer.toHexString(current)
hex = if(hex.length()<4) "\\u00"+hex else "\\u"+hex //補0
result.append(new String(StringEscapeUtils.unescapeJava(hex).getBytes,"utf-8"))
current = 0
matchIndex=0
isMatchEnd = true
}
}
}
return result.toString()
}
