Question

I am not to savvy on SQL queries but I am trying to build a query to remove any data (reported comments) based on a join to another table (user) using an ID to match, here is a sample schema below:

create table tbl_reported_comment(id int, commentId int, reported_by_userid int);

insert tbl_reported_comment values
  (1, 1, 101),
  (2, 2, 131),
  (3, 3, 101),
  (4, 4, 101),
  (5, 5, 24),
  (6, 6, 201),
  (7, 7, 1),
  (8, 8, 24),
  (9, 9, 23),
  (10, 10, 16),
  (11, 11, 31);

Create table tbl_user(userId int, Username varchar(50));

insert tbl_user values
  (1, 'admin'),
  (101, 'test1'),
  (131, 'test2'),
  (24, 'test3'),
  (201, 'test4');

What I am trying to achieve in this instance is the following:

Remove any data in the tbl_reported_comment table where the
[reported_by_userid] column doesn’t exist as an [userId] in the user
table

Here is a link to the SQLFIDDLE with this sample schema: SQLFiDDLE. I am using SQL Server as the database.

Many thanks,

Answer 1

How about this?

  DELETE FROM dbo.tbl_reported_comment
  OUTPUT DELETED.id, DELETED.commentId, DELETED.reported_by_userid
  WHERE reported_by_userid NOT IN 
             (SELECT UserID FROM dbo.tbl_user)

This will delete those rows and it will output those rows it deleted to the screen (in SQL Server Management Studio) so you can see what was deleted.

And Ben is right – once you’ve done this, you should establish a foreign key relationship between those two tables to avoid zombie data like this in the future!

-- make your "UserID" column NOT NULL
ALTER TABLE dbo.tbl_user
ALTER COLUMN UserID INT NOT NULL

-- so that we can use it as the PRIMARY KEY for that table!
ALTER TABLE dbo.tbl_user
ADD CONSTRAINT PK_user PRIMARY KEY CLUSTERED(UserID)

-- so that we can then establish a FK relationship between those two tables....
ALTER TABLE dbo.tbl_reported_comment
ADD CONSTRAINT FK_reported_comment_user
FOREIGN KEY (reported_by_userid) REFERENCES dbo.tbl_user(UserID)