Query with elements depended on another - sql

Question

There is a Postgres database and the table has three columns. The data structure is in external system so I can not modify it.
Every object is represented by three rows (identified by column element_id - rows with the same value in this column represents the same object), for example:
key value element_id
-----------------------------------
status active 1
name exampleNameAAA 1
city exampleCityAAA 1
status inactive 2
name exampleNameBBB 2
city exampleCityBBB 2
status inactive 3
name exampleNameCCC 3
city exampleCityCCC 3
In the query, I want to put list of some names, check if the value of row with key column status in the same object has status 'active' and return the name of this objects only if the status is 'active'.
So for this example, there are three objects in the database table. I want to put in query two 'names':
a)exampleNameAAA
b)exampleNameCCC
and the result should be:
exampleNameAAA (because I asked for two objects and only one of them has active value in status row.

Answer 1

You can use an EXISTS query:
select e1.*
from element e1
where (e1.key, e1.value) in ( ('name', 'exampleNameAAA'), ('name', 'exampleNameCCC'))
and exists (select *
from element e2
where e2.element_id = e1.element_Id
and (e2.key, e2.value) = ('status', 'active'));
Online example: https://rextester.com/JOWED21150

Answer 2

One option uses aggregation:
SELECT
MAX(CASE WHEN "key" = 'name' THEN "value" END) AS name
FROM yourTable
GROUP BY element_id
HAVING
MAX(CASE WHEN "key" = 'name' THEN "value" END) IN
('exampleNameAAA', 'exampleNameCCC') AND
SUM(CASE WHEN "key" = 'status' AND "value" = 'active' THEN 1 ELSE 0 END) > 0;
name
exampleNameAAA
Demo
This is the pivot approach, where we isolate individual keys and values for each element_id group.

Answer 3

I like expressing this as:
SELECT MAX(t.value) FILTER (WHERE t.key = 'name') AS name
FROM t
GROUP BY t.element_id
HAVING MAX(t.value) FILTER (WHERE t.key = 'name') IN ('exampleNameAAA', 'exampleNameCCC') AND
MAX(t.value) FILTER (WHERE t.key = 'status') = 'active';
All that said, the exists solution is probably more performant in this case. The advantage of the aggregation approach is that you can easily bring additional columns into the select, such as the city:
SELECT MAX(t.value) FILTER (WHERE t.key = 'name') AS name,
MAX(t.value) FILTER (WHERE t.key = 'city') as city
(Note: key is a bad name for a column because it is a SQL keyword.)