有時在解析json資料中的欄位key是動態可變的時候，由於Gson是使用靜態註解的方式來設定實體物件的，因此我們很難直接對返回的型別來判斷。但Gson在解析過程中如果不知道解析的欄位，就會將所有變數儲存在一個Map中，我們只要例項化這個map就能動態地取出key和value了。

先給出一段jsondata，這是天氣預報的資料，其中day_20151002這種key是隨日期而變化的，在實體類中就不能當做靜態變數來處理，我們就通過map來取出其對映物件。

{
    "resultcode": "200",
    "reason": "successed!",
    "result": {
        "sk"
 
: {
            "temp": "24",
            "wind_direction": "東北風",
            "wind_strength": "2級",
            "humidity": "28%",
            "time": "17:38"
        },
        "today": {
            "temperature": "15℃~26℃",
            "weather": "多雲轉晴",
            "wind": "東北風微風",
            "week"
 
: "星期日",
            "city": "桂林",
            "date_y": "2015年10月11日",
            "dressing_index": "舒適",
            "dressing_advice": "建議著長袖T恤、襯衫加單褲等服裝。年老體弱者宜著針織長袖襯衫、馬甲和長褲。",
            "uv_index": "弱",
            "comfort_index": "",
            "wash_index": "較適宜",
            "travel_index": "較適宜"
 
,
            "exercise_index": "較適宜",
            "drying_index": ""
        },
        "future": {
            "day_20151011": {
                "temperature": "15℃~26℃",
                "weather": "多雲轉晴",
                "wind": "東北風微風",
                "week": "星期日",
                "date": "20151011"
            },
            "day_20151012": {
                "temperature": "16℃~27℃",
                "weather": "晴轉多雲",
                "wind": "微風",
                "week": "星期一",
                "date": "20151012"
            },
            "day_20151013": {
                "temperature": "16℃~26℃",
                "weather": "多雲轉晴",
                ,
                "wind": "微風",
                "week": "星期二",
                "date": "20151013"
            },
            "day_20151014": {
                "temperature": "17℃~27℃",
                "weather": "晴",
                "wind": "北風微風",
                "week": "星期三",
                "date": "20151014"
            },
            "day_20151015": {
                "temperature": "17℃~28℃",
                "weather": "晴",
                "wind": "北風微風",
                "week": "星期四",
                "date": "20151015"
            },
            "day_20151016": {
                "temperature": "17℃~30℃",
                "weather": "晴",
                "wind": "北風微風",
                "week": "星期五",
                "date": "20151016"
            },
            "day_20151017": {
                "temperature": "17℃~30℃",
                "weather": "晴",
                "wind": "北風微風",
                "week": "星期六",
                "date": "20151017"
            }
        }
    },
    "error_code": 0
}

相關的實體類如下：

public class FutureDay {
    private String temperature;
    private String weather;
    private String wind;
    private String week;
    private String date;
}

public class Result {
    private Sk sk;
    private Today today;
    private Map<String,FutureDay> future;
}

public class Sk {
    private String temp;
    private String wind_direction;
    private String wind_strength;
    private String humidity;
    private String time;
}

public class Today {
    private String temperature;
    private String weather;
    private String week;
    private String city;
    private String date_y;
    private String dressing_index;
    private String dressing_advice;
    private String uv_index;
    private String comfort_index;
    private String wash_index;
    private String travel_index;
    private String exercise_index;
    private String drying_index;
}

public class Response {
    private String resultcode;
    private String reason;
    private String error_code;
    private Result result;
}

具體解析過程如下程式碼所示：

import java.io.BufferedReader;
import java.io.FileReader;
import java.util.Map;
import weather.*;
import com.google.gson.Gson;


public class GsonParseDynamicKey {
    public static  void main( String args []){
        String jsondata = readJsonFile();//從檔案中讀取出json字串，並打印出來
        Gson gson = new Gson();
        System.out.println("Start Gson parse jsondata");   
        Response response = gson.fromJson(jsondata, Response.class);        
        System.out.println(response.toString());
        System.out.println(response.getResult().getSk().toString());
        System.out.println(response.getResult().getToday().toString());

        Map<String, FutureDay> future = response.getResult().getFuture(); //對動態的key，來建立map，間接從中取出實體類futrue。
        System.out.println("Keyset method");                     //這裡取出value的方法有兩種keySet() entrySet().都給出了遍歷的方法
        for (String key:future.keySet()){                        //遍歷取出key，再遍歷map取出value。
            System.out.println("key:"+key); 
            System.out.println(future.get(key).toString());
        }

        System.out.println("Entryset method");
        for (Map.Entry<String,FutureDay> pair:future.entrySet()){//遍歷取出鍵值對，呼叫getkey()，getvalue()取出key和value。
             System.out.println("key:"+pair.getKey());
             System.out.println(pair.getValue().toString());
       }    
}

這裡順便一提遍歷Map的兩種方法keySet()，entrySet()的差別。
keySet()方法返回的是key的集合set，entrySet()返回的是鍵值對的集合set。雖然兩者從set遍歷取出元素的方法是一樣的，但是根據這個元素取出value的效率有些不同。前者取出的元素是key，還要去原map中遍歷取出value。
後者取出的元素是鍵值對，直接呼叫getkey()，getvalue()方法就能快速取出key和value。顯然在map中存在大量鍵值對時，使用entrySet()來取出value的效率更高。