Home >Database >Mysql Tutorial >How to Select Distinct User IDs with Multiple Role IDs in SQL?
Retrieving User IDs Meeting Specific Role ID Criteria Across Multiple Rows
This article addresses the challenge of selecting unique user IDs that satisfy multiple role ID conditions spread across different rows within a database table. Let's consider a userrole
table with userid
and roleid
columns:
userid | roleid |
---|---|
1 | 1 |
1 | 2 |
1 | 3 |
2 | 1 |
The goal is to identify all distinct userid
values associated with roleid
values 1, 2, and 3. Using the example data, the expected output is userid
1. Two SQL approaches can achieve this.
Method 1: Aggregate Query with HAVING Clause
This method utilizes an aggregate function and a HAVING
clause:
<code class="language-sql">SELECT userid FROM userrole WHERE roleid IN (1, 2, 3) GROUP BY userid HAVING COUNT(DISTINCT roleid) = 3;</code>
This query first filters rows where roleid
is 1, 2, or 3. Then, it groups the results by userid
and counts the number of distinct roleid
values for each user using COUNT(DISTINCT roleid)
. The HAVING
clause filters to only include users with exactly three distinct role IDs.
Method 2: Self-Join Query
A more efficient alternative, especially for larger datasets, is a self-join approach:
<code class="language-sql">SELECT DISTINCT t1.userid FROM userrole t1 JOIN userrole t2 ON t1.userid = t2.userid AND t2.roleid = 2 JOIN userrole t3 ON t1.userid = t3.userid AND t3.roleid = 3 WHERE t1.roleid = 1;</code>
This query uses three self-joins: t1
selects users with roleid
1; t2
ensures they also have roleid
2; and t3
confirms they have roleid
3. DISTINCT
ensures only unique userid
values are returned.
The optimal approach (aggregate vs. join) depends on table size and data distribution. For scenarios with a low probability of matching rows, the self-join method tends to perform better.
The above is the detailed content of How to Select Distinct User IDs with Multiple Role IDs in SQL?. For more information, please follow other related articles on the PHP Chinese website!