SQL选择具有相同列值的行

Question

Good afternoon you everyone! Could you help me? :)

DATABASE SAMPLE :

ID  Catalog Code  Description Type  Supplier  Supplier Code

1   6083          TV LG 32    tv    lg        ud28f1137ka-ga-i2-tr  
1   6083          TV LG 32    tv    samsung   asfb1145-ssd          # select it
2   6129          Phone 5X    phone apple     mics_rp
2   6129          Phone 5X    phone htc       nco_p13 961-x
2   6129          Phone 5X    phone nokia     n_41s
3   6210          Friezer     agd   samsung   asfb1145-ssd          # found match
                                                                  it has the same
                                                                  `Supplier Code`

The above ID is assigned to Catalog Code . All fields are VARCHAR except for ID . Could be empty.

WHAT I'D LIKE TO DO:

Select eg Catalog Code = 6083 and get all other Catalog Codes IF one of Supplier Codes is the same.

So I should get: Catalog Code = 6210 (last row) because it has the same Supplier Code as 6083 (first row).

MY POOR ATTEMPT:

SELECT a.*
  FROM `TABLE` a
  WHERE EXISTS
        (SELECT 1
            FROM `TABLE` b
            WHERE b.`Catalog Code` = '6083'
            AND (
                  a.`Supplier Code` NOT IN ('') 
              AND a.`Supplier Code` IN b.`Supplier Code`
                )
        )

NEED FOR DESIGN IMPROVEMENTS

My real DB has over 100 000 rows . I feel the DB design needs improvements but which ones?

INDEXES , PRIMARY KEY , second DB with FOREIGN KEY ?

Answer 1

I would write the query more like this:

select t.*
from t
where t.supplier_code in (select t2.supplier_code
                          from t t2
                          where t2.catalog_code = '6083'
                         );

Your code is rather strange. Why would you have not in ('') ? This is redundant. Then on the second in , you need a list, not a column value. In fact, just = is sufficient.

For performance, you want an index on t(catalog_code, supplier_code) .

Answer 2

For this particular question how about you use 2 table .

1) contains catalog code details.(keep catalog code as primary key)

2) Supplier details (keep ID as primary key- 1,2,3)

3) catalog code and supplier code-ID mapping (foreign key from both tables) .. both are integer values so queries will be bit faster because of comparison.