Add a new line for grouping similar rows

Question

As subject, possible to do this using SQL?
For eg:
KEY       Description
1           John
1           John Smith
2           Peter
2           Peter Pan
to:
KEY      Description
1          John
1          John Smith
<newline>
2          Peter
2          Peter Pan

Answer 1

With the model clause:
SQL> with t as (
  2    select 1 KEY, 'John' Description from dual union
  3    select 1 , 'John Smith' from dual union
  4    select 3 , 'Agent' Description from dual union
  5    select 3 , 'Agent Smith' from dual union
  6    select 2 , 'Peter' from dual union
  7    select 2 , 'Peter Pan' from dual)
  8  -- end of sample data
  9  SELECT decode(m, 0, to_char(k), 1, '<newline>') as key, d as description
10    FROM t
11   MODEL PARTITION BY (key as k)
12         DIMENSION BY (description as d)
13         MEASURES (0 as m)
14   RULES UPSERT
15   (m[null] = 1)
16  /
KEY                                      DESCRIPTION
1                                        John
1                                        John Smith
<newline>                               
3                                        Agent
3                                        Agent Smith
<newline>                               
2                                        Peter
2                                        Peter Pan
<newline>                               
9 rows selected
SQL> If you don't really care about the <newline> string and the key itself will do you can skip the decode and put an explicit ORDER BY as in:
SELECT k as key, d as description
  FROM t
MODEL PARTITION BY (key as k)
       DIMENSION BY (description as d)
       MEASURES (0 as m)
RULES UPSERT
(m[null] = 1)
ORDER BY k, dI'm assuming your column description won't have nulls for your key values.