DELETE FROM tablea a
      WHERE NOT EXISTS (SELECT NULL
                          FROM tableb b
                         WHERE a.uniqueid = b.uniqueid)

delete tableA 
where uniqueid in 
  (select uniqueid from tableB 
   minus 
   select uniqueid from tableA)
/

create table test_90 (col_1  number, col_2 number);

create table test_91 (col_1  number);

insert into test_90 values (1,10);
insert into test_90 values (2,10);
insert into test_90 values (3,10);

insert into test_91 values (1);

commit;
        
merge into test_90  t90
using (select col_1 from test_91) t91
on    (t90.col_1 = t91.col_1)
when matched then update set col_2 = null delete where col_1 is not null;

select * from test_90;

rollback;

delete /*+ bypass_ujvc */
       (select t90.*
        from   test_90 t90
              ,test_91 t91
        where  t90.col_1 =t91.col_1);

select * from test_90;

Operation	Object Name				Rows	Bytes	Cost	Object Node	In/Out	PStart	PStop
	
DELETE STATEMENT Optimizer Mode=ALL_ROWS		150 M	 	608284  	 	      	             	 
  DELETE	schema.tableA	  	 	 	 	      	             	 
    MERGE JOIN ANTI					150 M	2G	608284  	 	      	             	 
      INDEX FULL SCAN	schema.uniqueid_indexA		194 M	2G	431349  	 	      	             	 
      SORT UNIQUE					44 M	293 M	176935  	 	      	             	 
        INDEX FAST FULL SCAN	schema.uniqueid_indexB	44 M	293 M	23839  	 	      	             	 

select count(*) from table_a a, table_b b
where a.unique_id = b.unique_id (+)
and
b.unique_id is null;

CREATE TABLE tablea_temp as select * from tablea where uniqueid in (select uniqueid from tableb);

INSERT /*+ append */ into tablea select * from tablea_temp;