Hive 配置单元concat两个贴图对象

我在hive中有两个表，如下所示：

表1

key1  | value1

int   | map(int,array(int))

表2

key2  |  value2

int   | map(int,array(int))

现在我加入键上的表，我想连接两个具有相同键的映射。换句话说，决赛应该是这样的

表格

key  |  value

int  | map(int,array(int))

我在加入时尝试使用函数collect_set，如下所示：

collect_set(value1,value2) 

但它通过例外，只有一个输入是必需的。有什么想法或意见吗？ 谢谢

COLLECT\u SET（）

是一个聚合函数，因此如果您试图组合一些东西，它就不会真正有用（或有效）。您可以尝试使用

COMBINE（）

。可以在这个UDF库中找到它。假设您有如下数据：

表0：

idx     map_kv
0       {2:[1,2,3,4], 3:[5,6,7,8,9]}

idx     map_kv
0       {2:[5,6,7,8,9], 3:[1,2,3,4]}

ADD JAR /path/to/jar/brickhouse-0.7.1.jar;
CREATE TEMPORARY FUNCTION COLLECT AS 'brickhouse.udf.collect.CollectUDAF';
CREATE TEMPORARY FUNCTION COMBINE AS 'brickhouse.udf.collect.CombineUDF';

SELECT idx
  , COLLECT(map_key, arr) AS final_map
FROM (
  SELECT a.idx
    , a.map_key
    , COMBINE(map_val_0, map_val_1) AS arr
  FROM (
    SELECT idx
      , map_key
      , map_val_0
    FROM database.table0
    LATERAL VIEW EXPLODE(map_kv) exptbl0 AS map_key, map_val_0 ) a
  JOIN (
    SELECT idx
      , map_key
      , map_val_1
    FROM database.table1
    LATERAL VIEW EXPLODE(map_kv) exptbl1 AS map_key, map_val_1 ) b
  ON a.idx=b.idx AND a.map_key=b.map_key ) c
GROUP BY idx;

idx     final_map
0       {2:[1,2,3,4,5,6,7,8,9], 3:[5,6,7,8,9,1,2,3,4]}

表1：

idx     map_kv
0       {2:[1,2,3,4], 3:[5,6,7,8,9]}

idx     map_kv
0       {2:[5,6,7,8,9], 3:[1,2,3,4]}

ADD JAR /path/to/jar/brickhouse-0.7.1.jar;
CREATE TEMPORARY FUNCTION COLLECT AS 'brickhouse.udf.collect.CollectUDAF';
CREATE TEMPORARY FUNCTION COMBINE AS 'brickhouse.udf.collect.CombineUDF';

SELECT idx
  , COLLECT(map_key, arr) AS final_map
FROM (
  SELECT a.idx
    , a.map_key
    , COMBINE(map_val_0, map_val_1) AS arr
  FROM (
    SELECT idx
      , map_key
      , map_val_0
    FROM database.table0
    LATERAL VIEW EXPLODE(map_kv) exptbl0 AS map_key, map_val_0 ) a
  JOIN (
    SELECT idx
      , map_key
      , map_val_1
    FROM database.table1
    LATERAL VIEW EXPLODE(map_kv) exptbl1 AS map_key, map_val_1 ) b
  ON a.idx=b.idx AND a.map_key=b.map_key ) c
GROUP BY idx;

idx     final_map
0       {2:[1,2,3,4,5,6,7,8,9], 3:[5,6,7,8,9,1,2,3,4]}

那你就可以做了

查询：

idx     map_kv
0       {2:[1,2,3,4], 3:[5,6,7,8,9]}

idx     map_kv
0       {2:[5,6,7,8,9], 3:[1,2,3,4]}

ADD JAR /path/to/jar/brickhouse-0.7.1.jar;
CREATE TEMPORARY FUNCTION COLLECT AS 'brickhouse.udf.collect.CollectUDAF';
CREATE TEMPORARY FUNCTION COMBINE AS 'brickhouse.udf.collect.CombineUDF';

SELECT idx
  , COLLECT(map_key, arr) AS final_map
FROM (
  SELECT a.idx
    , a.map_key
    , COMBINE(map_val_0, map_val_1) AS arr
  FROM (
    SELECT idx
      , map_key
      , map_val_0
    FROM database.table0
    LATERAL VIEW EXPLODE(map_kv) exptbl0 AS map_key, map_val_0 ) a
  JOIN (
    SELECT idx
      , map_key
      , map_val_1
    FROM database.table1
    LATERAL VIEW EXPLODE(map_kv) exptbl1 AS map_key, map_val_1 ) b
  ON a.idx=b.idx AND a.map_key=b.map_key ) c
GROUP BY idx;

idx     final_map
0       {2:[1,2,3,4,5,6,7,8,9], 3:[5,6,7,8,9,1,2,3,4]}

这将产生：

输出：

idx     map_kv
0       {2:[1,2,3,4], 3:[5,6,7,8,9]}

idx     map_kv
0       {2:[5,6,7,8,9], 3:[1,2,3,4]}

ADD JAR /path/to/jar/brickhouse-0.7.1.jar;
CREATE TEMPORARY FUNCTION COLLECT AS 'brickhouse.udf.collect.CollectUDAF';
CREATE TEMPORARY FUNCTION COMBINE AS 'brickhouse.udf.collect.CombineUDF';

SELECT idx
  , COLLECT(map_key, arr) AS final_map
FROM (
  SELECT a.idx
    , a.map_key
    , COMBINE(map_val_0, map_val_1) AS arr
  FROM (
    SELECT idx
      , map_key
      , map_val_0
    FROM database.table0
    LATERAL VIEW EXPLODE(map_kv) exptbl0 AS map_key, map_val_0 ) a
  JOIN (
    SELECT idx
      , map_key
      , map_val_1
    FROM database.table1
    LATERAL VIEW EXPLODE(map_kv) exptbl1 AS map_key, map_val_1 ) b
  ON a.idx=b.idx AND a.map_key=b.map_key ) c
GROUP BY idx;

idx     final_map
0       {2:[1,2,3,4,5,6,7,8,9], 3:[5,6,7,8,9,1,2,3,4]}

---

请添加包含数据（而不是元数据）的完整示例