一个略复杂的数据映射聚合例子及代码重构

本文内容来自真实的工作案例，因其转换略复杂，且有一定意义，故记录之。

问题###

给定一个JSON串

{
    "item:s_id:18006666": "1024",
    "item:s_id:18008888": "1024",
    "item:g_id:18006666": "6666",
    "item:g_id:18008888": "8888",
    "item:num:18008888": "8",
    "item:num:18006666": "6",
    "item:item_core_id:18006666": "9876666",
    "item:item_core_id:18008888": "9878888",
    "item:order_no:18006666": "E20171013174712025",
    "item:order_no:18008888": "E20171013174712025",
    "item:id:18008888": "18008888",
    "item:id:18006666": "18006666",
    
    "item_core:num:9878888": "8",
    "item_core:num:9876666": "6",
    "item_core:id:9876666": "9876666",
    "item_core:id:9878888": "9878888",

    "item_price:item_id:1000": "9876666",
    "item_price:item_id:2000": "9878888",
    "item_price:price:1000": "100",
    "item_price:price:2000": "200",
    "item_price:id:2000": "2000",
    "item_price:id:1000": "1000",

    "item_price_change_log:id:1111": "1111",
    "item_price_change_log:id:2222": "2222",
    "item_price_change_log:item_id:1111": "9876666",
    "item_price_change_log:item_id:2222": "9878888",
    "item_price_change_log:detail:1111": "haha1111",
    "item_price_change_log:detail:2222": "haha2222",
    "item_price_change_log:id:3333": "3333",
    "item_price_change_log:id:4444": "4444",
    "item_price_change_log:item_id:3333": "9876666",
    "item_price_change_log:item_id:4444": "9878888",
    "item_price_change_log:detail:3333": "haha3333",
    "item_price_change_log:detail:4444": "haha4444"
}

这是一个订单的两个商品的基本信息、价格信息以及计价变更信息，是从DB里的结构获取到的记录。为了企业保密性，对表、字段以及值做了简化的特殊处理。目标是将这个订单的各个商品的信息聚合在一起。 即得到：

{1024_E20171013174712025_18006666={item:id=18006666, item_price:price=100, item_price:id=1000, item_price_change_log:id=[3333, 1111], item_core:num=6, item:g_id=6666, item:item_core_id=9876666, item_price:item_id=9876666, item:order_no=E20171013174712025, item_core:id=9876666, item_price_change_log:item_id=[9876666, 9876666], item:s_id=1024, item:num=6, item_price_change_log:detail=[haha3333, haha1111]}, 1024_E20171013174712025_18008888={item:id=18008888, item_price:price=200, item_price:id=2000, item_price_change_log:id=[2222, 4444], item_core:num=8, item:g_id=8888, item:item_core_id=9878888, item_price:item_id=9878888, item:order_no=E20171013174712025, item_price_change_log:item_id=[9878888, 9878888], item:s_id=1024, item_core:id=9878888, item:num=8, item_price_change_log:detail=[haha2222, haha4444]}}

提示###

细心的读者会发现，一个商品 item 对应一个 item_core, 一个 item_price ，可能对应多个 item_price_change_log 。这三个表都是通过 item:item_core_id 来关联的，在 item 表是 item:item_core_id 字段， 在 item_price 表是 item_price:item_id 字段， 在 item_price_change_log 表是 item_price_change_log:item_id 字段。

基本思路是：

STEP1: 对于含有表 item, item_core, item_price, item_price_change_log 数据的指定JSON，构建一个总的 itemIndexMap[table:id, Map[table:field, value]]，其中 key 是 table:id, value 是一个 valMap, valMap 的键是 table:field, 值是对应的最后面的值。 table: 作为名字空间的前缀，是为了避免不同表之间的ID和field相互重合覆盖。 见方法 buildItemIndexMap 的实现。

STEP2: 从 itemIndexMap 构建 item:id 与 item:item_core_id 的映射 itemCoreId2originItemIdMap；在实际应用中， originItemId 为 oldItemId (item:id), itemCoreId 为 newItemId (item:item_core_id)；

STEP3: 对于每一个 [table:id, Map[table:field, value]], 通过 itemCoreId2originItemIdMap 拿到对应的 originitemId , 然后使用 originitemId 为键替换itemIndexMap中对应的 item_core:id, item_price:id, item_price_change_log:id，将具有相同的 originItemId 的不同的 Map[table:field, value] 进行合并后，添加到新的newItemIndexMap 中。

程序实现###

初步实现####

ItemUtil.java

package zzz.study.utils;

/**
 * Created by shuqin on 17/11/10.
 */

import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

import java.util.HashMap;
import java.util.Map;
import java.util.Set;

import static zzz.study.utils.NewMapUtil.merge;

/**
 * Created by shuqin on 17/10/23.
 */
public class ItemUtil {

  private static Logger logger = LoggerFactory.getLogger(ItemUtil.class);

  /**
   * 构建一个订单的所有商品级别的信息以及映射
   * @param multiItemInfoMapForOneOrder 一个订单下多个商品的信息
   * @return 一个订单的所有商品级别的信息以及映射
   *
   * key 是 sID + order_no + item_id ; item_id 对应 item.id, 对应 item_core.item_core_id
   */
  public static Map<String, Map<String,Object>> buildFinalOrderItemMap(Map<String,String> multiItemInfoMapForOneOrder) {
    try {
      return mergeOrderItemMap(buildItemIndexMap(multiItemInfoMapForOneOrder));
    } catch (Exception ex) {
      logger.error("failed to buildFinalOrderItemMap for: " + multiItemInfoMapForOneOrder, ex);
      return new HashMap<>();
    }

  }

  /**
   * 构建一个订单的所有商品级别的信息以及映射
   * @param itemInfoMap 商品级别的信息
   * @return 一个订单的所有商品的信息
   *
   * NOTE: itemInfoMap 是对应一个订单的所有商品的信息的映射，不要传多个订单的信息进来，可能重复
   *
   * key = table:table_unique_id , value = map [ table:field, value ]
   *
   * eg. map [ item:goods_type:1800888 = 0 , item:num:1800888 = 8 ]
   *     will be transformed into map[item:1800888 = map[item:goods_type=0 , item:num=8]]
   */
  public static Map<String,Map<String,String>> buildItemIndexMap(Map<String, String> itemInfoMap) {
    Map<String, Map<String,String>> itemIndexMap = new HashMap<>();

    itemInfoMap.forEach(
        (key, value) -> {
          String[] keyparts = key.split(":");

          // 只考虑三段式 tablename:field:id
          if (keyparts != null && keyparts.length == 3) {
            String table = keyparts[0];
            String field = keyparts[1];
            String index = keyparts[2];

            String indexKey = table+ ":" + index;
            String fieldKey = table+":"+field;
            if (itemIndexMap.get(indexKey) == null) {
              itemIndexMap.put(indexKey, new HashMap<>());
            }
            itemIndexMap.get(indexKey).put(fieldKey, String.valueOf(value));
          }
        }
    );

    return itemIndexMap;
  }

  /**
   * 聚合一个订单下的所有商品信息
   * @param itemIndexMap 一个订单所有商品的信息映射
   * @return 一个订单下的所有商品信息
   *
   * key 是 sID + order_no + item_id ; item_id 对应 item.id, item_core.item_core_id
   */
  private static Map<String, Map<String,Object>> mergeOrderItemMap(Map<String, Map<String,String>> itemIndexMap) {

    if (itemIndexMap == null || itemIndexMap.isEmpty()) {
      return new HashMap<>();
    }

    // Map[oldItemId, newItemId]
    Map<String,String> old2newItemIdMap = new HashMap<>();
    Map<String,String> new2oldItemIdMap = new HashMap<>();

    Set<Map.Entry<String,Map<String,String>>> entries = itemIndexMap.entrySet();
    String orderNo = "";
    String sID = "";
    for (Map.Entry<String,Map<String,String>> entry: entries) {
      String indexKey = entry.getKey();
      Map<String,String> value = entry.getValue();

      if (indexKey.startsWith("item:")) {
        old2newItemIdMap.put(indexKey, value.get("item:item_core_id"));
        new2oldItemIdMap.put(value.get("item:item_core_id"), indexKey);
        orderNo = value.get("item:order_no");
        sID = value.get("item:s_id");
      }
    }

    Map<String, Map<String,Object>> newItemIndexMap = new HashMap<>();

    for (Map.Entry<String,Map<String,String>> entry: entries) {
      String indexKey = entry.getKey();
      Map<String,String> value = entry.getValue();

      if (indexKey.startsWith("item:")) {
        if (newItemIndexMap.get(indexKey) == null) {
          newItemIndexMap.put(indexKey, new HashMap<>());
        }
        newItemIndexMap.get(indexKey).putAll(value);
      }
      else if (indexKey.startsWith("item_core:")) {
        String itemCoreId = indexKey.split(":")[1];
        String oldItemId = new2oldItemIdMap.get(itemCoreId);
        if (newItemIndexMap.get(oldItemId) == null) {
          newItemIndexMap.put(oldItemId, new HashMap<>());
        }
        newItemIndexMap.get(oldItemId).putAll(value);
      }
      else if (indexKey.startsWith("item_price:")) {
        // item_price 与 item_id 一对一关系
        String itemCoreId = itemIndexMap.get(indexKey).get("item_price:item_id");
        String oldItemId = new2oldItemIdMap.get(itemCoreId);
        if (newItemIndexMap.get(oldItemId) == null) {
          newItemIndexMap.put(oldItemId, new HashMap<>());
        }
        newItemIndexMap.get(oldItemId).putAll(value);
      }
      else if (indexKey.startsWith("item_price_change_log:")) {
        // item_price_change_log 与 item_id 多对一关系
        String itemCoreId = itemIndexMap.get(indexKey).get("item_price_change_log:item_id");
        String oldItemId = new2oldItemIdMap.get(itemCoreId);
        if (newItemIndexMap.get(oldItemId) == null) {
          newItemIndexMap.put(oldItemId, new HashMap<>());
        }
        Map<String,Object> srcMap = newItemIndexMap.get(oldItemId);
        newItemIndexMap.get(oldItemId).putAll(merge(srcMap, value));
      }

    }

    return buildFinalOrderItemMap(newItemIndexMap, old2newItemIdMap, orderNo, sID);

  }


  /**
   * 构建最终的订单商品信息
   * @param itemIndexMap 商品信息
   * @param old2newItemIdMap 新老itemId映射
   * @param orderNo 订单号
   * @param sID 店铺号
   * @return 订单商品扩展信息
   */
  private static Map<String, Map<String,Object>> buildFinalOrderItemMap(Map<String, Map<String,Object>> itemIndexMap,
                                                                        Map<String,String> old2newItemIdMap,
                                                                        String orderNo, String sID) {
    Map<String, Map<String,Object>> finalResult = new HashMap<>();

    Set<Map.Entry<String,Map<String,Object>>> entries = itemIndexMap.entrySet();

    for (Map.Entry<String,Map<String,Object>> entry: entries) {
      String indexKey = entry.getKey();
      Map<String,Object> value = entry.getValue();

      String itemId = indexKey.split(":")[1];

      String itemKey = sID + "_" + orderNo + "_" + itemId;
      finalResult.put(itemKey, value);
    }
    return finalResult;
  }

}

NewMapUtil.java

package zzz.study.utils;

import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;

/**
 * Created by shuqin on 17/11/10.
 */
public class NewMapUtil {

  /**
   * 将 map 的值转为字符串类型
   */
  public static Map<String,String> transMap(Map<String,Object> map) {
    if (map == null) { return null; }
    Map<String,String> result = new HashMap<>();
    map.forEach(
        (k,v) -> result.put(k, v != null ? v.toString(): null)
    );
    return result;
  }

  /**
   * 将两个 Map 里相同 key 的值合并为列表
   *
   * eg. src = ["id": 1, "detail": "haha111", "extra":"extra111"] ,
   *     dest = ["id": 2, "detail": "haha222", "another": "another222"]
   * merge 之后返回 ["id": [1,2], "detail": ["haha111", "haha222"], "extra":"extra111", "another": "another222"]
   `   */
  public static Map<String, ? extends Object> merge(Map<String,? extends Object> src, Map<String, ? extends Object> dest) {
    if (src == null || src.size() == 0) { return dest; }
    if (dest == null || dest.size() == 0) { return src; }

    Map<String, Object> result = new HashMap<>();
    src.forEach(
        (key, value) -> {
          Object valueDesc = dest.get(key);
          if (valueDesc != null) {
            result.put(key, mergeToList(value, valueDesc));
          }
          else {
            result.put(key, value);
          }
        }
    );
    dest.forEach(
        (key, value) -> {
          if (result.get(key) == null) {
            result.put(key, value);
          }
        }
    );
    return result;

  }


  public static List mergeToList(Object src, Object... args) {
    List valList = new ArrayList();
    add(valList, src);
    for (Object arg: args) {
      add(valList, arg);
    }
    return valList;
  }

  public static List add(List valList, Object src) {
    if (src == null) { return valList; }
    if (src instanceof List) {
      valList.addAll((List)src);
    }
    else {
      valList.add(src);
    }
    return valList;
  }

}

代码重构###

可以看到，初步实现虽然实现了功能，可是代码比较乱，尤其是 mergeOrderItemMap 方法，混杂了业务表的逻辑，理解和扩展起来比较麻烦。需要仔细重构下。另外，Map的遍历访问比较啰嗦，可以更简洁一些。

使用forEach进行Map遍历####

重构从简单做起。原来使用了

Set<Map.Entry<String,Map<String,Object>>> entries = itemIndexMap.entrySet();

    for (Map.Entry<String,Map<String,Object>> entry: entries) {
      String indexKey = entry.getKey();
      Map<String,Object> value = entry.getValue();

      String itemId = indexKey.split(":")[1];

      String itemKey = sID + "_" + orderNo + "_" + itemId;
      finalResult.put(itemKey, value);
    }

在 java8 中可使用 forEach 语法简洁表达：

itemIndexMap.forEach(
        (indexKey,value) -> {
          String itemId = indexKey.split(":")[1];
          String itemKey = sID + "_" + orderNo + "_" + itemId;
          finalResult.put(itemKey, value);
        }
    );

抽离通用逻辑####

里面有很多如下重复代码

if (newItemIndexMap.get(oldItemId) == null) {
          newItemIndexMap.put(oldItemId, new HashMap<>());
        }
        newItemIndexMap.get(oldItemId).putAll(value);

实际上就是将新的item_core_id 映射成 item.id ，然后填充到新的Map，可以抽离出一个函数，然后复用。

private static void putNewIndexMap(Map<String, Map<String,Object>> newItemIndexMap,
                                     String indexKey, Map<String,String> value, Function<String, String> getOriginItemIdFunc) {
    String originItemId = getOriginItemIdFunc.apply(indexKey);
    if (newItemIndexMap.get(originItemId) == null) {
      newItemIndexMap.put(originItemId, new HashMap<>());
    }
    Map<String,Object> srcMap = newItemIndexMap.get(originItemId);
    newItemIndexMap.get(originItemId).putAll(merge(srcMap, value));
}

调用： putNewIndexMap(newItemIndexMap, indexKey, value, key -> key);

配置化####

看如下代码，里面含有 item_price: , item_price:item_id 这样的业务信息，破坏了方法的通用性。

if (indexKey.startsWith("item_price:")) {
        // item_price 与 item_id 一对一关系
        String itemCoreId = itemIndexMap.get(indexKey).get("item_price:item_id");
}

可以抽离出来做成配置。好的配置可以让代码大大简化。仔细思考下整个过程：当 indexKey 包含某个表的前缀时，取它对应的 itemCoreId 的字段，然后得到 itemCoreId ，再根据 putNewIndexMap 方法将对应的 Map 添加到最终的Map中。见如下代码：

private static Map<String,String> itemIdConf = new HashMap() {
    {
      put("item", "item:item_core_id");
      put("item_core", "item_core:id");
      put("item_price", "item_price:item_id");
      put("item_price_change_log", "item_price_change_log:item_id");
    }
};

String table = indexKey.split(":")[0];
if (itemIdConf.containsKey(table)) {
   String itemCoreIdField = itemIdConf.get(table);
   String itemCoreId = itemIndexMap.get(indexKey).get(itemCoreIdField);
   putNewIndexMap(newItemIndexMap, indexKey, value,
      key -> new2oldItemIdMap.get(itemCoreId));
}
        

哈！if-elseif-elseif 语句消失了！此时，才说真正抓住了设计重点：将itemId 的映射抽离出来做成配置，然后其它的依此解析。

重构后的ItemUtil####

// 这个配置含有新老itemId映射的信息以及获取orderNo, sID 的字段信息
  private static List<String>
      itemBaseDataConf = Arrays.asList("item:", "item:item_core_id",
                                       "item:order_no", "item:s_id");

  private static Map<String,String> itemIdConf = new HashMap() {
    {
      put("item", "item:item_core_id");
      put("item_core", "item_core:id");
      put("item_price", "item_price:item_id");
      put("item_price_change_log", "item_price_change_log:item_id");
    }
  };

  /**

   * 聚合一个订单下的所有商品信息
   * @param itemIndexMap 一个订单所有商品的信息映射
   * @return 一个订单下的所有商品信息
   *
   * key 是 sID + order_no + item_id ;
   */
  private static Map<String, Map<String,Object>> mergeOrderItemMap(Map<String, Map<String,String>> itemIndexMap) {

    if (itemIndexMap == null || itemIndexMap.isEmpty()) {
      return new HashMap<>();
    }

    // Map[oldItemId, newItemId]
    Map<String,String> old2newItemIdMap = new HashMap<>();
    Map<String,String> new2oldItemIdMap = new HashMap<>();

    Set<Map.Entry<String,Map<String,String>>> entries = itemIndexMap.entrySet();
    String orderNo = "";
    String kdtId = "";

    // 构建itemID映射
    for (Map.Entry<String,Map<String,String>> entry: entries) {
      String indexKey = entry.getKey();
      Map<String,String> value = entry.getValue();

      if (indexKey.startsWith(itemBaseDataConf.get(0))) {
        old2newItemIdMap.put(indexKey, value.get(itemBaseDataConf.get(1)));
        new2oldItemIdMap.put(value.get(itemBaseDataConf.get(1)), indexKey);
        orderNo = value.get(itemBaseDataConf.get(2));
        kdtId = value.get(itemBaseDataConf.get(3));
      }
    }

    Map<String, Map<String,Object>> newItemIndexMap = aggregationAllInfoOfEachItem(itemIndexMap, new2oldItemIdMap);
    return buildFinalOrderItemMap(newItemIndexMap, old2newItemIdMap, orderNo, kdtId);

  }

  /*
   * 聚合每个商品的所有信息
   *
   * Map[item:id, Map[table:field, value]]
   */
  private static Map<String, Map<String,Object>> aggregationAllInfoOfEachItem(Map<String, Map<String,String>> itemIndexMap, Map<String,String> new2oldItemIdMap) {

    Map<String, Map<String,Object>> newItemIndexMap = new HashMap<>();

    itemIndexMap.forEach(
        (indexKey, value) -> {

          String table = indexKey.split(":")[0];
          if (itemIdConf.containsKey(table)) {
            String itemCoreIdField = itemIdConf.get(table);
            String itemCoreId = itemIndexMap.get(indexKey).get(itemCoreIdField);
            putNewIndexMap(newItemIndexMap, indexKey, value,
                           key -> new2oldItemIdMap.get(itemCoreId));
          }
        }
    );

    return newItemIndexMap;

  }

  /*
   * 将各商品信息聚合到相应的原itemId的键下
   */
  private static void putNewIndexMap(Map<String, Map<String,Object>> newItemIndexMap,
                                     String indexKey, Map<String,String> value, Function<String, String> getOriginItemIdFunc) {
    String originItemId = getOriginItemIdFunc.apply(indexKey);
    if (newItemIndexMap.get(originItemId) == null) {
      newItemIndexMap.put(originItemId, new HashMap<>());
    }
    Map<String,Object> srcMap = newItemIndexMap.get(originItemId);
    newItemIndexMap.get(originItemId).putAll(merge(srcMap, value));
  }

Java8特性###

Function####

可以看到，putNewIndexMap 使用了 Function 作为参数，让调用方指定如何去获取 originItemId，然后根据获取的originItemId进行通用处理。这里 Function 实现了模板方法模式。

协变####

注意到这个方法签名使用了 [ ? extends Object ]。 这里使用了协变特性。即对应参数 Map[String, ? extend Object]，既可以传入 Map[String, Object], 也可以传入 Map[String, String] ，或 Map[String, Entity] ，避免从 Map[String, Entity] 到 Map[String, Object] 的无聊转换。

public static Map<String, ? extends Object> merge(Map<String,? extends Object> src, Map<String, ? extends Object> dest) 

单测###

ItemUtilTest.java

package cc.lovesq.study.test.datastructure;

import org.junit.Test;

import java.util.Arrays;
import java.util.Map;

import cc.lovesq.study.test.CommonForTest;
import zzz.study.utils.ItemUtil;
import zzz.study.utils.JsonUtil;
import zzz.study.utils.NewMapUtil;

/**
 * Created by shuqin on 17/11/10.
 */
public class ItemUtilTest extends CommonForTest {

  String newItemInfoStr = "{
"
                          + "    "item:s_id:18006666": "1024",
"
                          + "    "item:s_id:18008888": "1024",
"
                          + "    "item:g_id:18006666": "6666",
"
                          + "    "item:g_id:18008888": "8888",
"
                          + "    "item:num:18008888": "8",
"
                          + "    "item:num:18006666": "6",
"
                          + "    "item:item_core_id:18006666": "9876666",
"
                          + "    "item:item_core_id:18008888": "9878888",
"
                          + "    "item:order_no:18006666": "E20171013174712025",
"
                          + "    "item:order_no:18008888": "E20171013174712025",
"
                          + "    "item:id:18008888": "18008888",
"
                          + "    "item:id:18006666": "18006666",
"
                          + "    "item_core:num:9878888": "8",
"
                          + "    "item_core:num:9876666": "6",
"
                          + "    "item_core:id:9876666": "9876666",
"
                          + "    "item_core:id:9878888": "9878888",
"
                          + "    "item_price:item_id:1000": "9876666",
"
                          + "    "item_price:item_id:2000": "9878888",
"
                          + "    "item_price:price:1000": "100",
"
                          + "    "item_price:price:2000": "200",
"
                          + "    "item_price:id:2000": "2000",
"
                          + "    "item_price:id:1000": "1000",
"
                          + "    "item_price_change_log:id:1111": "1111",
"
                          + "    "item_price_change_log:id:2222": "2222",
"
                          + "    "item_price_change_log:item_id:1111": "9876666",
"
                          + "    "item_price_change_log:item_id:2222": "9878888",
"
                          + "    "item_price_change_log:detail:1111": "haha1111",
"
                          + "    "item_price_change_log:detail:2222": "haha2222",
"
                          + "    "item_price_change_log:id:3333": "3333",
"
                          + "    "item_price_change_log:id:4444": "4444",
"
                          + "    "item_price_change_log:item_id:3333": "9876666",
"
                          + "    "item_price_change_log:item_id:4444": "9878888",
"
                          + "    "item_price_change_log:detail:3333": "haha3333",
"
                          + "    "item_price_change_log:detail:4444": "haha4444"
"
                          + "}";

  @Test
  public void testBuildItemIndexMapForNew() {
    Map<String,Object> itemInfoMap = JsonUtil.readMap(newItemInfoStr);
    Map<String,Map<String,String>> itemIndexMap = ItemUtil
        .buildItemIndexMap(NewMapUtil.transMap(itemInfoMap));

    System.out.println(itemIndexMap);

    eq("18006666", itemIndexMap.get("item:18006666").get("item:id"));
    eq("6666", itemIndexMap.get("item:18006666").get("item:g_id"));
    eq("1024", itemIndexMap.get("item:18006666").get("item:s_id"));
    eq("E20171013174712025", itemIndexMap.get("item:18006666").get("item:order_no"));
    eq("9876666", itemIndexMap.get("item:18006666").get("item:item_core_id"));

    eq("18008888", itemIndexMap.get("item:18008888").get("item:id"));
    eq("8888", itemIndexMap.get("item:18008888").get("item:g_id"));
    eq("1024", itemIndexMap.get("item:18008888").get("item:s_id"));
    eq("E20171013174712025", itemIndexMap.get("item:18008888").get("item:order_no"));
    eq("9878888", itemIndexMap.get("item:18008888").get("item:item_core_id"));

    eq("9876666", itemIndexMap.get("item_core:9876666").get("item_core:id"));
    eq("6", itemIndexMap.get("item_core:9876666").get("item_core:num"));

    eq("9878888", itemIndexMap.get("item_core:9878888").get("item_core:id"));
    eq("8", itemIndexMap.get("item_core:9878888").get("item_core:num"));

    eq("9876666", itemIndexMap.get("item_price:1000").get("item_price:item_id"));
    eq("1000", itemIndexMap.get("item_price:1000").get("item_price:id"));
    eq("100", itemIndexMap.get("item_price:1000").get("item_price:price"));

    eq("9878888", itemIndexMap.get("item_price:2000").get("item_price:item_id"));
    eq("2000", itemIndexMap.get("item_price:2000").get("item_price:id"));
    eq("200", itemIndexMap.get("item_price:2000").get("item_price:price"));

    eq("9876666", itemIndexMap.get("item_price_change_log:1111").get("item_price_change_log:item_id"));
    eq("haha1111", itemIndexMap.get("item_price_change_log:1111").get("item_price_change_log:detail"));
    eq("9878888", itemIndexMap.get("item_price_change_log:2222").get("item_price_change_log:item_id"));
    eq("haha2222", itemIndexMap.get("item_price_change_log:2222").get("item_price_change_log:detail"));

    eq("9876666", itemIndexMap.get("item_price_change_log:3333").get("item_price_change_log:item_id"));
    eq("haha3333", itemIndexMap.get("item_price_change_log:3333").get("item_price_change_log:detail"));
    eq("9878888", itemIndexMap.get("item_price_change_log:4444").get("item_price_change_log:item_id"));
    eq("haha4444", itemIndexMap.get("item_price_change_log:4444").get("item_price_change_log:detail"));
  }

  @Test
  public void testBuildFinalOrderItemMapForNew() {
    Map<String,Object> itemInfoMap = JsonUtil.readMap(newItemInfoStr);
    Map<String, Map<String,Object>> finalOrderItemMap = ItemUtil
        .buildFinalOrderItemMap(NewMapUtil.transMap(itemInfoMap));
    System.out.println(finalOrderItemMap);

    eq("18006666", finalOrderItemMap.get("1024_E20171013174712025_18006666").get("item:id"));
    eq("6666", finalOrderItemMap.get("1024_E20171013174712025_18006666").get("item:g_id"));
    eq("1024", finalOrderItemMap.get("1024_E20171013174712025_18006666").get("item:s_id"));
    eq("E20171013174712025", finalOrderItemMap.get("1024_E20171013174712025_18006666").get("item:order_no"));
    eq("9876666", finalOrderItemMap.get("1024_E20171013174712025_18006666").get("item_core:id"));
    eq("6", finalOrderItemMap.get("1024_E20171013174712025_18006666").get("item_core:num"));
    eq("9876666", finalOrderItemMap.get("1024_E20171013174712025_18006666").get("item_price:item_id"));
    eq("100", finalOrderItemMap.get("1024_E20171013174712025_18006666").get("item_price:price"));

    eq("18008888", finalOrderItemMap.get("1024_E20171013174712025_18008888").get("item:id"));
    eq("8888", finalOrderItemMap.get("1024_E20171013174712025_18008888").get("item:g_id"));
    eq("1024", finalOrderItemMap.get("1024_E20171013174712025_18008888").get("item:s_id"));
    eq("E20171013174712025", finalOrderItemMap.get("1024_E20171013174712025_18008888").get("item:order_no"));
    eq("9878888", finalOrderItemMap.get("1024_E20171013174712025_18008888").get("item_core:id"));
    eq("8", finalOrderItemMap.get("1024_E20171013174712025_18008888").get("item_core:num"));
    eq("9878888", finalOrderItemMap.get("1024_E20171013174712025_18008888").get("item_price:item_id"));
    eq("200", finalOrderItemMap.get("1024_E20171013174712025_18008888").get("item_price:price"));

    eq(Arrays.asList("haha3333", "haha1111"), finalOrderItemMap.get("1024_E20171013174712025_18006666").get("item_price_change_log:detail"));
    eq(Arrays.asList("haha2222", "haha4444"), finalOrderItemMap.get("1024_E20171013174712025_18008888").get("item_price_change_log:detail"));
  }

}

CommonForTest.java

package cc.lovesq.study.test;

import org.junit.Assert;

import java.util.List;

import static org.junit.Assert.assertEquals;

/**
 * Created by shuqin on 17/11/10.
 */
public class CommonForTest {

  public static final String NOT_THROW_EXCEPTION = "Not Throw Exception";

  public void eq(Object expected, Object actual) {
    assertEquals(expected, actual);
  }

  public <T> void eq(T[] expected, T[] actual) {
    Assert.assertArrayEquals(expected, actual);
  }

  public <T> void eq(List<T> expectedList, List<T> actualList) {
    if (expectedList == null && actualList == null) {
      return ;
    }
    assertEquals(expectedList.size(), actualList.size());
    for (int i=0; i< expectedList.size(); i++) {
      assertEquals(expectedList.get(i), actualList.get(i));
    }
  }

  public void fail(String message) {
    Assert.fail(message);
  }

}

package zzz.study.utils;

import org.codehaus.jackson.map.DeserializationConfig;
import org.codehaus.jackson.map.ObjectMapper;
import org.codehaus.jackson.map.annotate.JsonSerialize;

import java.text.SimpleDateFormat;
import java.util.HashMap;
import java.util.Map;

public class JsonUtil {

  private static final ObjectMapper MAPPER = new ObjectMapper();

  static {
    // 为保持对象版本兼容性,忽略未知的属性
    MAPPER.configure(DeserializationConfig.Feature.FAIL_ON_UNKNOWN_PROPERTIES, false);
    // 序列化的时候，跳过null值
    MAPPER.getSerializationConfig().setSerializationInclusion(JsonSerialize.Inclusion.NON_NULL);
    // date类型转化
    SimpleDateFormat fmt = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss");
    MAPPER.setDateFormat(fmt);
  }

  /**
   * 将一个json字符串解码为java对象
   *
   * 注意：如果传入的字符串为null，那么返回的对象也为null
   *
   * @param json json字符串
   * @param cls  对象类型
   * @return 解析后的java对象
   * @throws RuntimeException 若解析json过程中发生了异常
   */
  public static <T> T toObject(String json, Class<T> cls) {
    if (json == null) {
      return null;
    }
    try {
      return MAPPER.readValue(json, cls);
    } catch (Exception e) {
      return null;
    }
  }

  /**
   * 读取JSON字符串为MAP
   */
  @SuppressWarnings("unchecked")
  public static Map<String, Object> readMap(String json) {
    return toObject(json, HashMap.class);
  }

}