今天在解析json數據的時候得到了一堆這樣的數據：{"errNum":0,"errMsg":"success","retData":[{"title":"\u6536\u5e9f\u54c1\u5927\u53d4\u521a\u4e0a\u53f0\uff0c\u5c31\u60e8\u906d\u8bc4\u59d4\u706d\u706f\uff0c\u4f46\u63a5\u4e0b\u6765\u5168\u573a\u90fd\u9707\u60ca\u4e86\uff01","url":"http:\/\/toutiao.com\/group\/6263036756505920002\/","abstract":"\u8ba2\u9605\u6211\u83b7\u53d6\u66f4\u591a\u7cbe\u5f69\u5185\u5bb9\uff01","image_url":"http:\/\/p1.pstatp.com\/list\/2f90009a31a7ee8bb15"}]}

這是因為，為了更好的傳輸中文，json進行了Unicode編碼。

這樣一來，我們在解析json之前，就得要先將json數據中的Unicode編碼轉換為我們使用的中文；

一：http請求數據返回json中string字段包含unicode的轉碼

1. public static String decodeUnicode(String theString) {
2. char aChar;
3. int len = theString.length();
4. StringBuffer outBuffer = new StringBuffer(len);
5. for (int x = 0; x < len;) {
6. aChar = theString.charAt(x++);
7. if (aChar == ‘\\‘) {
8. aChar = theString.charAt(x++);
9. if (aChar == ‘u‘) {
10. // Read the xxxx
11. int value = 0;
12. for (int i = 0; i < 4; i++) {
13. aChar = theString.charAt(x++);
14. switch (aChar) {
15. case ‘0‘:
16. case ‘1‘:
17. case ‘2‘:
18. case ‘3‘:
19. case ‘4‘:
20. case ‘5‘:
21. case ‘6‘:
22. case ‘7‘:
23. case ‘8‘:
24. case ‘9‘:
25. value = (value << 4) + aChar - ‘0‘;
26. break;
27. case ‘a‘:
28. case ‘b‘:
29. case ‘c‘:
30. case ‘d‘:
31. case ‘e‘:
32. case ‘f‘:
33. value = (value << 4) + 10 + aChar - ‘a‘;
34. break;
35. case ‘A‘:
36. case ‘B‘:
37. case ‘C‘:
38. case ‘D‘:
39. case ‘E‘:
40. case ‘F‘:
41. value = (value << 4) + 10 + aChar - ‘A‘;
42. break;
43. default:
44. throw new IllegalArgumentException(
45. "Malformed \\uxxxx encoding.");
46. }
48. }
49. outBuffer.append((char) value);
50. } else {
51. if (aChar == ‘t‘)
52. aChar = ‘\t‘;
53. else if (aChar == ‘r‘)
54. aChar = ‘\r‘;
55. else if (aChar == ‘n‘)
56. aChar = ‘\n‘;
57. else if (aChar == ‘f‘)
58. aChar = ‘\f‘;
59. outBuffer.append(aChar);
60. }
61. } else
62. outBuffer.append(aChar);
63. }
64. return outBuffer.toString();
65. }

二、普通string含有unicode轉碼方法

1. public static String reEncoding(String text, String newEncoding) {
2. String str = null;
3. try {
4. str = new String(text.getBytes(), newEncoding);
5. } catch (UnsupportedEncodingException e) {
6. log.error("不支持的字符編碼" + newEncoding);
7. throw new RuntimeException(e);
8. }
9. return str;
10. }

三、說一下比較奇怪的方案，測試中無意發現的，暫時沒弄明白原理（有明白原理的大神，請告知一聲，謝謝）

我用HttpClent的post方式獲取的json數據，得到的是帶Unicode碼的數據，需要轉換成中文才行，但是轉換的時間感覺有點長，就用HttpURLConnection的get方式又試了一下，在不轉碼的情況下，經過gson解析後，竟然神奇的自動轉換成了中文：

簡直是太神奇了，而且需要的時間相對於HttpClient的post請求方式的請求和處理時間更短，所以，果斷換用HttpURLConnection的get方式了

①現在先貼一下HttpURLConnection的get的方式：

1. @Test
2. public void test() {
3. try {
4. long start = System.currentTimeMillis();
5. URL url = new URL("http://apis.baidu.com/songshuxiansheng/news/news");
7. HttpURLConnection connection = (HttpURLConnection) url.openConnection();
8. connection.addRequestProperty("apikey","0fc807e45a37ce264f45d169646f4a9e" );
9. String dataString = new String(GsonTools.IsToByte(connection.getInputStream()),"utf-8");
10. HeadlineJson newsJson = GsonTools.getObjectData(dataString, HeadlineJson.class);
11. List<Headline>list = newsJson.getRetData();
12. System.out.println(list.toString());
13. long end = System.currentTimeMillis();
15. System.out.println("timeGap:"+(end-start));
16. } catch (Exception e) {
17. // TODO Auto-generated catch block
18. e.printStackTrace();
19. }
20. }

調用的GsonTools的方法：（之前的博文中有寫到過）

1. public static <T> T getObjectData(String jsonString, Class<T> type) {
3. T t = null;
4. try {
5. Gson gson = new Gson();
6. t = gson.fromJson(jsonString, type);
7. } catch (JsonSyntaxException e) {
8. // TODO Auto-generated catch block
9. e.printStackTrace();
10. }
11. return t;
12. }

②然後貼一下HttpClient的post方式：

1. @Test
2. public void TestHeadLine() {
3. long start = System.currentTimeMillis();
4. List<NameValuePair> params = new ArrayList<NameValuePair>();
5. String url = "http://apis.baidu.com/songshuxiansheng/news/news";
6. String jsonString = HttpUtils.getBaiDuString2(url, params);
7. HeadlineJson lineJson = GsonTools.getObjectData(jsonString, HeadlineJson.class);
8. System.out.println(lineJson.toString());
9. long end = System.currentTimeMillis();
11. System.out.println("timeGap:"+(end-start));
12. }

調用的HttpUtils的方法：

1. public static String getBaiDuString(String url,List<NameValuePair> params) {
2. String serverDataString = null;
3. HttpPost post = new HttpPost(url);
4. try {
5. post.setEntity(new UrlEncodedFormEntity(params, HTTP.UTF_8));
6. post.addHeader("apikey", UrlUtils.BAIDU_API_KEY);
7. HttpClient client = new DefaultHttpClient();
8. HttpResponse response = client.execute(post);
9. int code = response.getStatusLine().getStatusCode();
10. System.out.println("StatusCode:" + code);
12. if (code == 200) {
13. serverDataString = decodeUnicode(EntityUtils.toString(response.getEntity()));
14. // serverDataString = EntityUtils.toString(response.getEntity());
16. System.out.println("接收字符串數據成功\nServerData:"+serverDataString);
17. }
18. } catch (Exception e) {
19. // TODO Auto-generated catch block
20. e.printStackTrace();
21. }
23. return serverDataString;
24. }

③調用的HttpClient的get方式

1. public static String getBaiDuString2(String url,List<NameValuePair> params) {
2. String serverDataString = null;
3. HttpGet get = new HttpGet(url);
4. try {get.addHeader("apikey", UrlUtils.BAIDU_API_KEY);
5. HttpClient client = new DefaultHttpClient();
6. HttpResponse response = client.execute(get);
7. int code = response.getStatusLine().getStatusCode();
8. System.out.println("StatusCode:" + code);
10. if (code == 200) {
11. // serverDataString = decodeUnicode(EntityUtils.toString(response.getEntity()));
12. serverDataString = EntityUtils.toString(response.getEntity());
14. System.out.println("接收字符串數據成功\nServerData:"+serverDataString);
15. }
16. } catch (Exception e) {
17. // TODO Auto-generated catch block
18. e.printStackTrace();
19. }
21. return serverDataString;
22. }

谷歌提供的HttpClient的通信和HttpURLConnection網絡通信的時間間隔我也做了比較，明顯，HttpURLConnection的請求時間更短，所以果斷使用HttpURLConnection的方式

四、java中本身就提供了對Unicode 的url進行解碼的方法了：

1. System.out.println(URLDecoder.decode("\u82f9\u679c", "utf-8"));
   

通過http.client解析url返回的數據時為什麽中文變成了unicode碼