sql - logic behind deletion of duplicate rows? -

what logic behind deletion of duplicate rows? got know query used delete duplicate rows.

delete tvsemp e rowid>(select min(rowid) tvsemp m e.ename=m.ename); 

here when divided subquery i.e first select min(m.rowid) tvsemp e, tvsemp m e.ename=m.ename; result rowid

min(m.rowid) ___________________ aaaeduaabaaakiqaap 

then outer query select ename tvsemp emp rowid>'aaaeduaabaaakiqaap' getting n-1 results(except 1 eliminated using '>' sin)

my doubt if 2 combined getting exact table without duplicate why this?

are asking if first query work?

delete tvsemp e rowid>(select min(rowid) tvsemp m e.ename=m.ename); 

the answer "yes".

as how works, subquery correlated subquery, means references outer query. oracle executes subquery each row of tvsemp e, , e.ename = m.ename in subquery limits subquery's results rows ename equals ename in outer query's current row.

getting n - 1 results want: if there 4 rows name chay, want delete 3 of them.

put way, if there 4 rows name chay, 1 of them have minimum rowid value. rowid other 3 higher minimum, delete query delete them.

i hope makes sense. there's more information subqueries (including correlated subqueries) here.