在postgres中加入列中多个json数组元素的方法。要附加所有colmns json数组以创建新数组

Question

以下是样本数据

样本表

id | data_value 1 | [{"a":"f1"},{"b":"b1"},{"c":"c1"}] 2 | [{"a":"f2"},{"b":"b2"},{"c":"c2"}] 3 | [{"a":"f3"},{"b":"b2"},{"c":"c3"}] 4 | [{"a":"f4"},{"b":"b3"},{"c":"v4"}] 5 | [{"a":"f5"},{"b":"b4"},{"c":"c4"}]

结果

`[{"a":"f1"},{"b":"b1"},{"c":"c1"},{"a":"f2"},{"b":"b2"},{"c":"c2"},"a":"f3"},{"b":"b2"},{"c":"c3"}{"a":"f4"},{"b":"b3"},{"c":"v4"},    {"a":"f5"},{"b":"b4"},{"c":"c4"}]`

Answer 1

t=# select ('['||translate(json_agg(dv)::text,'[]','')||']')::jsonb from d1;
                                                                                                jsonb

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
-------------
 [{"a": "f1"}, {"b": "b1"}, {"c": "c1"}, {"a": "f2"}, {"b": "b2"}, {"c": "c2"}, {"a": "f3"}, {"b": "b2"}, {"c": "c3"}, {"a": "f4"}, {"b": "b3"}, {"c": "v4"}, {"a": "f5"}, {"b": "b4"},
{"c": "c4"}]
(1 row)

或更整洁的SQL，但需要一些构建：

create or replace function json_agg_arr(jsonb,jsonb) returns jsonb as $$
  select case when coalesce($1::text, '') <> '' then $1 || $2 else $2 end;
$$ language sql called on null input;

create aggregate json_agg_arr (jsonb) (sfunc=json_agg_arr,stype=jsonb);

然后只聚合行：

t=# select json_agg_arr(dv::jsonb) from d1;
                                                                                            json_agg_arr

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
-------------
 [{"a": "f1"}, {"b": "b1"}, {"c": "c1"}, {"a": "f2"}, {"b": "b2"}, {"c": "c2"}, {"a": "f3"}, {"b": "b2"}, {"c": "c3"}, {"a": "f4"}, {"b": "b3"}, {"c": "v4"}, {"a": "f5"}, {"b": "b4"},
{"c": "c4"}]
(1 row)