How to merge two json arrays by property id (SQL Server) - arrays

Question

I have two columns of json that I would like to join on id into a single select.
Sample Data
| a | b |
+------------------------------------------------+-------------------------------------+
| [{id: 1, name: "Alice"},{id:2, name: "Bob"}] | [{id: 1, age: 30}, {id:2, age: 32}] |
| [{id: 5, name: "Charlie"},{id:6, name: "Dale"} | [{id: 5, age: 20}, {id:6, age: 14}] |
Desired Output
| c |
+-------------------------------------------------------------------+
| [{id: 1, name: "Alice", age: 30},{id:2, name: "Bob", age: 32}] |
| [{id: 5, name: "Charlie", age: 20},{id:6, name: "Dale", age: 14}] |
I'd like to do something like
select
id,
name,
age
from openJson(select a from someDb) sd
with (
id int '$.id',
age int '$.age'
)
inner join (
select
id,
age
from openJson(select b from someDb)
with (
id int '$.id',
age int '$.name'
)
) x
on x.id = sd.id

Answer 1

I don't think that current versions of SQL Server support a MERGE function. The only option is JSON_MODIFY() function, that can either update the value of an existing property, insert a new key:value pair or delete a key.
But in your case, the more appropriate approach is to parse the stored JSON as tables using OPENJSON() with explicit schema, join the tables and rebuild the required JSON output again:
SELECT
c = (
SELECT a.id, a.name, b.age
FROM OPENJSON(v.a) WITH (
id int '$.id',
name varchar(50) '$.name'
) a
FULL JOIN OPENJSON(v.b) WITH (
id int '$.id',
age int '$.age'
) b ON a.id = b.id
FOR JSON PATH
)
FROM (VALUES
('[{"id": 1, "name": "Alice"}, {"id":2, "name": "Bob"}]', '[{"id": 1, "age": 30}, {"id":2, "age": 32}]'),
('[{"id": 5, "name": "Charlie"}, {"id":6, "name": "Dale"}]', '[{"id": 5, "age": 20}, {"id":6, "age": 14}]')
) v (a, b)
Result:
c
--------------------------------------------------------------------
[{"id":1,"name":"Alice","age":30},{"id":2,"name":"Bob","age":32}]
[{"id":5,"name":"Charlie","age":20},{"id":6,"name":"Dale","age":14}]