Resolve column names in a sql Query

Question

Hi Folks,
     I’m upgrading an application and there are database columns that have major changes that could affect production reports. I would like to survey v$SQL for a few days to see if these columns are affected. Here’s the problem.
Looking at a SQL query to determine which columns are used is not nearly as easy as it sounds. Are there any oracle functions or 3rd party tools that can be used to list the columns used by a sql query? (this gets extremely tricky when subqueries become involved)
For instance.
select
name, b.dep_id, employee_id
From emp a
Inner join v_department b on a.dep_id=b.dep_id
Columns used:+
Emp.name+
Departments.dep_id (under the view)+
Emp.employee_id+
Emp.dep_id (from the join)+

Answer 1

If you're on > 10g, you can use dbms_xplan with "all" option.
It shows "column projection information" which is exactly what you want.
UKJA@ukja102> set serveroutput on
UKJA@ukja102>
UKJA@ukja102> drop table t1 purge;
Table dropped.
Elapsed: 00:00:00.03
UKJA@ukja102> create table t1(c1 int, c2 int);
Table created.
Elapsed: 00:00:00.03
UKJA@ukja102>
UKJA@ukja102> explain plan for
  2  select
  3    t1.c1, t1.c2, v.cnt
  4  from t1,
  5    (select /*+ no_merge */ c1, count(*) as cnt
  6        from t1
  7        group by c1) v
  8  where
  9    t1.c1 = v.c1
10  ;
Explained.
Elapsed: 00:00:00.01
UKJA@ukja102>
UKJA@ukja102> @plan_all
UKJA@ukja102> select * from table(dbms_xplan.display(null,null,'all'))
  2  /
PLAN_TABLE_OUTPUT                                                              
Plan hash value: 536125944                                                     
| Id  | Operation            | Name | Rows  | Bytes | Cost (%CPU)| Time     |  
|   0 | SELECT STATEMENT     |      |     1 |    52 |     6  (34)| 00:00:01 |  
|*  1 |  HASH JOIN           |      |     1 |    52 |     6  (34)| 00:00:01 |  
|   2 |   TABLE ACCESS FULL  | T1   |     1 |    26 |     2   (0)| 00:00:01 |  
|   3 |   VIEW               |      |     1 |    26 |     3  (34)| 00:00:01 |  
|   4 |    HASH GROUP BY     |      |     1 |    13 |     3  (34)| 00:00:01 |  
|   5 |     TABLE ACCESS FULL| T1   |     1 |    13 |     2   (0)| 00:00:01 |  
Query Block Name / Object Alias (identified by operation id):                  
   1 - SEL$1                                                                   
   2 - SEL$1 / T1@SEL$1                                                        
   3 - SEL$2 / V@SEL$1                                                         
   4 - SEL$2                                                                   
   5 - SEL$2 / T1@SEL$2                                                        
Predicate Information (identified by operation id):                            
   1 - access("T1"."C1"="V"."C1")                                              
Column Projection Information (identified by operation id):                    
   1 - (#keys=1) "T1"."C1"[NUMBER,22], "T1"."C2"[NUMBER,22],                   
       "V"."CNT"[NUMBER,22]                                                    
   2 - "T1"."C1"[NUMBER,22], "T1"."C2"[NUMBER,22]                              
   3 - "V"."C1"[NUMBER,22], "V"."CNT"[NUMBER,22]                               
   4 - (#keys=1) "C1"[NUMBER,22], COUNT(*)[22]                                 
   5 - "C1"[NUMBER,22]                                                         
Note                                                                           
   - dynamic sampling used for this statement                                  
40 rows selected.
Elapsed: 00:00:00.01
UKJA@ukja102>
UKJA@ukja102>
UKJA@ukja102> @end
UKJA@ukja102> set echo offDion Cho