No. It can be at table level, but this will get failed 
while referencing in another table.

A cursor is a SELECT statement that is defined within the 
declaration section of your PLSQL code. We'll take a look 
at three different syntaxes for cursors.

Cursor without parameters (simplest)
The basic syntax for a cursor without parameters is:

CURSOR cursor_name
IS
    SELECT_statement;



For example, you could define a cursor called c1 as below.

CURSOR c1
IS
    SELECT course_number
      from courses_tbl
      where course_name = name_in;

The result set of this cursor is all course_numbers whose 
course_name matches the variable called name_in.



Below is a function that uses this cursor.

CREATE OR REPLACE Function FindCourse
   ( name_in IN varchar2 )
   RETURN number
IS
    cnumber number;

    CURSOR c1
    IS
       SELECT course_number
        from courses_tbl
        where course_name = name_in;

BEGIN

open c1;
fetch c1 into cnumber;

if c1%notfound then
     cnumber := 9999;
end if;

close c1;

RETURN cnumber;

END;

Try this...Hope this will give the Required Answer
create table testsamp(a char,b char,c char);
insert into testsamp values ('A','B','C');
insert into testsamp values ('A','B','X');

select a,b,c from(
 select a , b,
 c||lead(c,1) over (partition by a,b order by a,b) c from 
testsamp ) tmp where rownum=1;

Regards,
Murali

Try this,


update testsamp1 set c=decode
(c,'MALE','FEMALE','FEMALE','MALE');


Regards,
Murali

SELECT CASE WHEN instr(TASK_NUMBER,'NN')>0 THEN 'NN' ELSE 
null END FROM PA_TASKS;

to retrieve data from the database tables is called query

A Query is a SELECT statement, which is always successful.

Implicit : Every SQL Statement performed as implicit cursor. Ex: Select * from Emp where empno = 7839 This is query automatically open the cursor and fetch the record getting the output of that record.

Explicit : In PLSql Block in we declare on declaration section we declare with name and select statement. In executable block we call the cursor and fetch the records more the one. For Ex : Declare
cursor c1 is select * from emp
getemp c1%rowtype;
begin
open c1;
loop
fetch c1 into getemp;
dbms_output.put_line (c1.col list);
end loop;
close c1;
end;  This is explicit cursor

delete from abc where rowid not in (select max(rowid) from
abc group by column_name_with_dup.values.);

delete from emp e1 where rowid >(select min(rowid) from emp 
e2 where e1.empid=e2.empid)

delete from 
<<table_name>>
where rowid in 
(select a.rowid
from <<table_name a,table_name b>>
where a.colname = b.colname
b.colname2 = b.colname2
......
...
a.colnamen = colnamen
order by .... desired colnames)

Primary key and unique are Entity integrity constraints

Primary key allows each row in a table to be uniquely 
identified and ensures that no duplicate rows exist and no 
null values are entered.

Unique key constraint is used to prevent the duplication of 
key values within the rows of a table and allow null 
values. (In oracle, one null is not equal to another null).

More or less both the keys sounds same, but the major 
diffrence is where unique will allow null values, because 
each null treated as a diffrent value internally. Incase of 
primary key it will not allow any repeated data and null 
data.

Yes.When a column is assigned primary key it means it 
contains unique value(i.e. no duplicate value is allowed) & 
it can't accept null values.
  Whereas unique key means no duplicare value is allowed, but
null value is allowed. And not null means no null value is
accepted but duplicate value can be passed.
 So when we assign primary key to a column, the column
becomes both unique & not null.

oracle 10g has introduced data pump,awr addm etc

oracle 10g has many features compare to 9i like
asm,swr,addm,flaskback recovery,data pump..
generating snap shot is much more flxible in 10 g compare to 9i

Plese see thin link: http://www.orafaq.com/wiki/PL/SQL_FAQ