SQL Query constructing for searching values in Tables single Column

Question

I have Table USER_SERVICES with 2 Columns USER_NAME, Services. Please find below sample data on USER_SERVICES Table.
A user can be assigned with multiple services(service1, service2 etc.,). So we store data as below :
USER_NAME Services
user1 service1
user1 service2
user1 service3
user2 service1
user2 service2
user3 service3
user4 service4
I need to frame a SQL Query to 'select users assigned with Service A and Service B and Service n'.
Note: 'And' condition is used when Searching for multiple services.
Example: The SQL Query has to dynamically handle condition if i give Single Service, multiple Services(service1 and service2 e.t.c,)
If i say select users assigned with service1 and service2, from the above Table, result should be user1 and user2
If i say select users assigned with service1 and service2 and service3, from the above Table, result should be user1
If i say select users assigned with service4 from the above table, result should be user4
If i say select users assigned with service1 from the above table, result should be user2,user1
If i say select users assigned with service1 and service2 from the above table, result should be user2,user1
How do i construct the SQL Query. Any help will be appreciated.
Thanks,
Sathish
Alternate email: [email protected]

Answer 1

How it will be stored? In a string or a table or a collection type?
If you have it in a table you can do like this.
with t
as
select 'user1' user_name, 'service1' services from dual union all
select 'user1', 'service2' from dual union all
select 'user1', 'service3' from dual union all
select 'user2', 'service1' from dual union all
select 'user2', 'service2' from dual union all
select 'user3', 'service3' from dual union all
select 'user4', 'service4' from dual
t1 as
select 'service1' services from dual union all
select 'service2' from dual
select distinct user_name
  from (
select t.user_name, t.services, t1.services, count(t1.services) over(partition by t.user_name) cnt, t1.cnt cnt_1
  from t
  left join (select t1.*, count(*) over() cnt from t1) t1 on t1.services = t.services
where cnt = cnt_1