Different Outer Join query result

Similar Questions

• OUTER JOIN query returns the results of JOIN IN-HOUSE 11.2.0.1.0

  I'm data transfer in 11.2.01.0 (Windows XP 32-bit) and I wanted to compare the sizes of table with the same Table name in two different patterns, ML and SILENT, with a FULL OUTER JOIN (to account for all the tables and NULL values in a diagram).
  
  The scheme of ML has 176 tables: schema TUT a 133 tables. The use of a standard INNER JOIN gives 131 paintings.
  
  I get precisely the results with a FULL OUTER JOIN I get with an INTERNAL JOIN (not the same NULL values so I know they exist).
  
  This happens in SQL-Plus, SQL_Developer and using Oracle Wire pilot of Data_Direct.
  
  Here is the code:
  
  Login: SYS as SYSDBA or SYSTEM (same results for either)
  
  SELECT M.TABLE_NAME, M.NUM_ROWS, T.TABLE_NAME, T.NUM_ROWS
  OF SYS. ALL_TABLES M FULL OUTER JOIN SYS. ALL_TABLES T ON M.TABLE_NAME = T.TABLE_NAME
  WHERE
  M.OWNER = 'ML' AND
  T.OWNER = 'TUT';
  
  Produce the same results with LEFT OUTER joins and RIGHT OUTER joins in ASI and Oracle (+) syntax.
  
  Any thoughts?

  Hello

  If you read what I posted, forget it. MScallion (below) gave the correct answerr.

  If conditions such as

  owner   = 'ML'
  

  in the WHERE clause, and then they will reject the rows of the result set formed by the join condition.

  The inner join returns only 131 lines where the two 'paintings' have the same table_names.
  The outer joins return multiple lines (133, 176 or 178) before the place WHERE the provision is applied , but the WHERE clause eliminates all lines except the 131 found by the inner join.

  Published by: Frank Kulash, July 10, 2010 14:23

• need help for an outer join query

  Hi friends...
  I have oracle 10g...
  I have a question which involve is joining three tables...
  
  the query is as follows:

  SELECT DISTINCT MU.MKT_ID                  "PAR ID",
                     MU.MKT_ID                  "ROOT ID",
                   MSU.SPEC_ID                "ID",
                 SU.SPEC_DESC                "DESC",
             SU.SPEC_CD          "SPEC CD"
  
                FROM IPOADM_BATCH.MKT_UV1        MU,
                       IPOADM_BATCH.MKT_SPEC_UV1   MSU,
               IPOADM_BATCH.SPEC_UV1       SU
  
                 WHERE MSU.MKT_ID  = MU.MKT_ID
             AND MSU.SPEC_ID = SU.SPEC_ID
           AND SU.SPEC_GRP_OR_CMPSTN_CD = 'C';

  in my front end application has some delete operations that removes specific data of the IPOADM_BATCH. MKT_SPEC_UV1 due to which the above query is nor fetch all rows.
  the relationship between the tables is clearly from the above query...
  
  I have the data in IPOADM_BATCH. MKT_UV1 and IPOADM_BATCH. Table of SPEC_UV1 after the delete operation
  
  now, I want the query above to retrieve data from IPOADM_BATCH. MKT_UV1 with small changes to the above query even if there is no data in IPOADM_BATCH. MKT_SPEC_UV1.
  
  I thought that if I use an outer join I can get it, but here I have to join three tables-join condition is based on the table that doesn't have a data...
  
  
  so please help me guys how can I change the query to get my desired out put...

  Hello

  There are several different things you could mean by it.

  Here's how to join the tables to get one of them:

  SELECT DISTINCT  mu.mkt_id                  "PAR ID",
                   mu.mkt_id                  "ROOT ID",
                   msu.spec_id                "ID",
                   su.spec_desc               "DESC",
                   su.spec_cd                   "SPEC CD"
  FROM           ipoadm_batch.mkt_uv1        mu
  LEFT OUTER JOIN      ipoadm_batch.mkt_spec_uv1   msu  ON   msu.mkt_id               = mu.mkt_id
  LEFT OUTER JOIN  ipoadm_batch.spec_uv1       su       ON   msu.spec_id              = su.spec_id
                                                AND  su.spec_grp_or_cmpstn_cd = 'C'
  ;
  

  If it does not matter what it is that you want, and then after a small example of data (CREATE TABLE and INSERT, only relevant columns instructions) for all the tables and the results desired from these data.
  Highlight a few places where the above querry is the production of incorrect results of your sample data and explains how to get the correct results in these places.

• Help for a LEFT OUTER JOIN query

  Hello, all,.

  I'm having some trouble setting up an Oracle 11 g Server SQL query, and I could use some help.

  Let's say tableA is blogs; tableC is comments for blog entries; tableB is the associative array:

  tableA
  blogID        blogTitle       blogBody      dateEntered
  1             This is a test  More text...  2016-05-20 11:11:11
  2             More testing    Still more!   2016-05-19 10:10:10
  3             Third charm!!   Blah, blah.   2016-05-18 09:09:09
  

  tableC
  commID        userID          userText      dateEntered
  10            Bravo           I like it!    2016-05-20 11:21:31
  11            Charlie         I don't!      2016-05-20 11:31:51
  12            Alpha           Do it again!  2016-05-19 10:20:30
  13            Bravo           Still more?   2016-05-19 10:30:50
  14            Charlie         So, what?     2016-05-19 10:35:45
  15            Bravo           Blah, what?   2016-05-18 09:10:11
  16            Alpha           Magic number! 2016-05-18 09:11:13
  

  tableB
  blogID        commID
  1             10
  1             11
  1             12
  2             13
  2             14
  3             15
  3             16
  
  
  
  

  I'm trying to get blogID, blogTitle, blogBody and the number of comments for each blog entry.  But, since I'm on to_char() for date and COUNT (commID) for the total number of comments, I am not "a group by expression.

  Here is an example of pseudo-SQL of what I'm trying.

  SELECT a.blogID, a.blogTitle, a.blogBody, to_char(a.dateEntered,'YYYY-MM-DD HH24:MI:SS') as dateEntered, COUNT(c.commID) as total
  FROM tableA a LEFT OUTER JOIN tableB b ON b.blog_ID = a.blog_ID
                LEFT OUTER JOIN tableC c ON c.commID = b.commID
  WHERE a.blogID = '1'
  GROUP BY blogID, blogTitle, blogBody
  ORDER BY to_date(dateEntered,'MM-DD-YYYY HH24:MI:SS') desc
  

  I'm sure it's something simple, but I just DO NOT see it.  Can you help me?

  V/r,

  ^_^

  Try:

  GROUP BY a.blogID, a.blogTitle, a.blogBody, to_char(a.dateEntered,'YYYY-MM-DD HH24:MI:SS')
  

  See you soon

  Eddie

• Concatinating join query results

  Hello
  
  Is it possible to concatenate the value of a particular column that are the result of a join query.
  
  Employee
  

  Employee Id | Employee Name
  254         | Employee 01 
  266         | Employee 02 
  654         | Employee 03
  741         | Employee 04
  266         | Employee 05
  877         | Employee 06
  955         | Employee 07

  Project
  

  Project Id | Project Name 
  1          | Project 01   
  2          | Project 02       
  3          | Project 03               

  Project_Employee
  

  Project Id | Employee ID 
  1          | 254         
  1          | 266        
  1          | 654        
  2          | 741        
  2          | 266        
  3          | 877          
  2          | 955        

  Result
  

  Project Id | Project Name | Employees
  1          | Project 01   | Employee 01,  Employee 02,  Employee 03
  2          | Project 02   | Employee 04, Employee 05, Employee 07
  3          | Project 03   | Employee 06         

  I use oracle 10g
  
  Thank you

  Hello

  SQL> with employee as (
    2  select 254 employee_id, 'Employee 01' employee_name from dual  union all
    3  select 266, 'Employee 02' from dual  union all
    4  select 654, 'Employee 03' from dual union all
    5  select 741,  'Employee 04' from dual union all
    6  select 266, 'Employee 05' from dual union all
    7  select 877, 'Employee 06'from dual union all
    8  select  955,         'Employee 07' from dual),
    9  project as (
   10  select  1 project_id,'Project 01' project_name  from  dual union all
   11  select 2 , 'Project 02' from dual  union all
   12  select 3, 'Project 03' from dual),
   13  project_employee as(
   14  select 1 project_id,         254  employee_id from dual union all
   15  select 1 ,         266  from  dual union all
   16  select 1 ,654 from  dual union all
   17  select 2,741  from dual union all
   18  select 2 , 266   from dual  union all
   19  select 3 , 877  from dual union all
   20  select 2 ,955 from dual),
   21  a  as (select  p.project_id, p.project_name, e.employee_id, e.employee_name
  
   22              from employee e, project p, project_employee pe
   23              where pe.project_id = P.project_id
   24              and pe.employee_id = e.employee_id)
   25              select a.project_id proj_id  , a.project_name proj_name,
   26                    rtrim(xmlagg(xmlelement("enam",
   27                                         a.employee_name || ', ')
   28                    order by employee_id).extract('//text()'),', ')
   29                     as employee_names
   30              from a
   31              group by a.project_id,a.project_name;
  
     PROJ_ID PROJ_NAME
  ---------- ----------
  EMPLOYEE_NAMES
  --------------------------------------------------------------------------------
  
           1 Project 01
  Employee 01, Employee 02, Employee 05, Employee 03
  
           2 Project 02
  Employee 02, Employee 05, Employee 04, Employee 07
  
           3 Project 03
  Employee 06
  
  SQL>
  

  The output may be formatted, but the result is what you want.

• Left outer join query and check the status

  Hello

  I have two tables as tables below

  table_1

  MI_ACC_IDENTIFIER	CHARGE_START_DATE	CHARGE_END_DATE	PBA_INT_AMT
  2000000000	01 SEP-05 00.00.00	29 SEP-05 00.00.00	0.26
  2000000000	30 SEP-05 00.00.00	31 OCTOBER 05 00.00.00	1.92
  2000000000	1 NOVEMBER 05 00.00.00	NOVEMBER 30 05 00.00.00	0.34
  2000000000	1 AUGUST 06 00.00.00	31 AUGUST 06 00.00.00	0.47
  2000000000	31 MARCH 06 00.00.00	27 APRIL 06 00.00.00	0.34
  2000000000	30 DECEMBER 05 00.00.00	31 JANUARY 06 00.00.00	1.92
  2000000000	1 MARCH 05 00.00.00	31 MARCH 05 00.00.00	0.26

  Table_2

  MI_ACC_IDENTIFIER	CHARGE_START_DATE	CHARGE_END_DATE	TOT_INT_AMT_OVER_25P
  2000000000	30 SEP-05 00.00.00	31 OCTOBER 05 00.00.00	0
  2000000000	1 NOVEMBER 05 00.00.00	NOVEMBER 30 05 00.00.00	0.81756
  2000000000	1 DECEMBER 05 00.00.00	29 DECEMBER 05 00.00.00	0.64724
  2000000000	30 DECEMBER 05 00.00.00	31 JANUARY 06 00.00.00	5.51555

  Power required:

  MI_ACC_IDENTIFIER	CHARGE_START_DATE	CHARGE_END_DATE	NVL(B.PBA_INT_AMT,0)	TOT_INT_AMT_OVER_25P
  2000000000	1 NOVEMBER 05 00.00.00	NOVEMBER 30 05 00.00.00	0.34	0.81756
  2000000000	30 DECEMBER 05 00.00.00	31 JANUARY 06 00.00.00	1.92	5.51555
  2000000000	1 DECEMBER 05 00.00.00	29 DECEMBER 05 00.00.00	0	0.64724

  I have to check if TOT_INT_AMT_OVER_25P > B.PBA_INT_AMT and also required to display if there is no matching record in table_1 and exist in table_2 then display as well

  Queries for the table:

  CREATE TABLE 'TABLE_1 '.

  (

  ACTIVATE THE "MI_ACC_IDENTIFIER" NUMBER (10,0) NOT NULL,

  DATE OF THE "CHARGE_START_DATE."

  DATE OF THE "CHARGE_END_DATE."

  NUMBER (15.2) "PBA_INT_AMT".

  );

  Insert into TABLE_1 (MI_ACC_IDENTIFIER, CHARGE_START_DATE, CHARGE_END_DATE, PBA_INT_AMT) values (2000000000, to_date ('01 - SEP - 05 00.00.00','DD-MON-RR HH24.MI.)) SS'), to_date (29-SEP-05 00.00.00','DD-MON-RR HH24.MI.) SS'), 0.26);

  Insert into TABLE_1 (MI_ACC_IDENTIFIER, CHARGE_START_DATE, CHARGE_END_DATE, PBA_INT_AMT) values (2000000000, to_date (30-OCT-05 00.00.00','DD-MON-RR HH24.MI.)) SS'), to_date (31 October 05 00.00.00','DD-MON-RR HH24.MI.) SS'), 1.92);

  Insert into TABLE_1 (MI_ACC_IDENTIFIER, CHARGE_START_DATE, CHARGE_END_DATE, PBA_INT_AMT) values (2000000000, to_date (1 November 05 00.00.00','DD-MON-RR HH24.MI.)) To_date SS'), (30 November 05 00.00.00','DD-MON-RR HH24.MI.) SS'), 0.34).

  Insert into TABLE_1 (MI_ACC_IDENTIFIER, CHARGE_START_DATE, CHARGE_END_DATE, PBA_INT_AMT) values (2000000000, to_date (1 August 06 00.00.00','DD-MON-RR HH24.MI.)) SS'), to_date (31 August 06 00.00.00','DD-MON-RR HH24.MI.) SS'), 0.47);

  Insert into TABLE_1 (MI_ACC_IDENTIFIER, CHARGE_START_DATE, CHARGE_END_DATE, PBA_INT_AMT) values (2000000000, to_date (31 March 06 00.00.00','DD-MON-RR HH24.MI.)) To_date SS'), (27 April 06 00.00.00','DD-MON-RR HH24.MI.) SS'), 0.34).

  Insert into TABLE_1 (MI_ACC_IDENTIFIER, CHARGE_START_DATE, CHARGE_END_DATE, PBA_INT_AMT) values (2000000000, to_date (30 December 05 00.00.00','DD-MON-RR HH24.MI.)) To_date SS'), (31 January 06 00.00.00','DD-MON-RR HH24.MI.) SS'), 1.92);

  Insert into TABLE_1 (MI_ACC_IDENTIFIER, CHARGE_START_DATE, CHARGE_END_DATE, PBA_INT_AMT) values (2000000000, to_date (1 March 05 00.00.00','DD-MON-RR HH24.MI.)) SS'), to_date (31 March 05 00.00.00','DD-MON-RR HH24.MI.) SS'), 0.26);

  CREATE TABLE 'TABLE_2.

  (

  ACTIVATE THE "MI_ACC_IDENTIFIER" NUMBER (10,0) NOT NULL,

  DATE OF THE "CHARGE_START_DATE."

  DATE OF THE "CHARGE_END_DATE."

  "TOT_INT_AMT_OVER_25P" NUMBER (15.5)

  );

  Insert in TABLE_2 (MI_ACC_IDENTIFIER, CHARGE_START_DATE, CHARGE_END_DATE, TOT_INT_AMT_OVER_25P) values (2000000000, to_date (30-OCT-05 00.00.00','DD-MON-RR HH24.MI.)) SS'), to_date (31 October 05 00.00.00','DD-MON-RR HH24.MI.) SS'), 0);

  Insert in TABLE_2 (MI_ACC_IDENTIFIER, CHARGE_START_DATE, CHARGE_END_DATE, TOT_INT_AMT_OVER_25P) values (2000000000, to_date (1 November 05 00.00.00','DD-MON-RR HH24.MI.)) To_date SS'), (30 November 05 00.00.00','DD-MON-RR HH24.MI.) SS'), 0.81756);

  Insert in TABLE_2 (MI_ACC_IDENTIFIER, CHARGE_START_DATE, CHARGE_END_DATE, TOT_INT_AMT_OVER_25P) values (2000000000, to_date (1 December 05 00.00.00','DD-MON-RR HH24.MI.)) SS'), to_date (29 December 05 00.00.00','DD-MON-RR HH24.MI.) SS'), 0.64724);

  Insert in TABLE_2 (MI_ACC_IDENTIFIER, CHARGE_START_DATE, CHARGE_END_DATE, TOT_INT_AMT_OVER_25P) values (2000000000, to_date (30 December 05 00.00.00','DD-MON-RR HH24.MI.)) To_date SS'), (31 January 06 00.00.00','DD-MON-RR HH24.MI.) SS'), 5.51555);

  Query, I used:

  	SELECT A.MI_ACC_IDENTIFIER, A.CHARGE_START_DATE, A.CHARGE_END_DATE, NVL (B.PBA_INT_AMT, 0), a.TOT_INT_AMT_OVER_25P OF
  	TABLE_2 A, B FROM TABLE_1
  	WHERE A.MI_ACC_IDENTIFIER = B.MI_ACC_IDENTIFIER (+)
  	AND A.CHARGE_START_DATE = B.CHARGE_START_DATE (+)
  	AND A.CHARGE_END_DATE = B.CHARGE_END_DATE (+)
  	and A.TOT_INT_AMT_OVER_25P > B.PBA_INT_AMT (+);

  I have been using the syntax of ANSI join for some time. They are readable and code looks more elegant. A reason do not want to use it?

  I don't care

  SQL> select t2.mi_acc_identifier
    2       , t2.charge_start_date
    3       , t2.charge_end_date
    4       , nvl(t1.pba_int_amt, 0) pba_int_amt
    5       , t2.tot_int_amt_over_25p
    6    from table_1 t1
    7       , table_2 t2
    8   where t1.mi_acc_identifier (+)= t2.mi_acc_identifier
    9     and t1.charge_start_date (+)= t2.charge_start_date
   10     and t1.charge_end_date   (+)= t2.charge_end_date
   11     and (t1.pba_int_amt < t2.tot_int_amt_over_25p     or t1.mi_acc_identifier is null);
  
  MI_ACC_IDENTIFIER CHARGE_ST CHARGE_EN PBA_INT_AMT TOT_INT_AMT_OVER_25P
  ----------------- --------- --------- ----------- --------------------
         2000000000 01-NOV-05 30-NOV-05         .34               .81756
         2000000000 30-DEC-05 31-JAN-06        1.92              5.51555
         2000000000 01-DEC-05 29-DEC-05           0               .64724
  
  SQL>
  

• outer join query

  Hi, here is the code to test
  

  create table dm_top (site varchar2(10),
  area varchar2(10),
  machine varchar2(10),
  operation varchar2(10));
  
  create table dm_mid (site varchar2(10),
  area varchar2(10),
  machine varchar2(10),
  operation varchar2(10),
  activity varchar2(10));
  
  create table dm_tr(site varchar2(10),
  area varchar2(10),
  machine varchar2(10),
  operation varchar2(10),
  dm_tr_id number); 

  INSERT orders
  

  insert into dm_top
    (site, area, machine, operation)
  values
    ('site1','area1','machine1','oper1');
    
    insert into dm_top
    (site, area, machine, operation)
  values
    ('site1','area1','machine1','oper2');
    
  insert into dm_mid
    (site, area, machine, operation, activity)
  values
    ('site','area1','machine1','oper1','act1');
  
    
   insert into dm_mid
    (site, area, machine, operation, activity)
  values
    ('site','area1','machine1','oper1','act2');
    
    insert into dm_mid
    (site, area, machine, operation, activity)
  values
    ('site','area1','machine1','oper1','act3');
    
  
    
    insert into dm_mid
    (site, area, machine, operation, activity)
  values
    ('site','area1','machine1','oper2','act1');
  
    
   insert into dm_mid
    (site, area, machine, operation, activity)
  values
    ('site','area1','machine1','oper2','act2');
    
    insert into dm_mid
    (site, area, machine, operation, activity)
  values
    ('site','area1','machine1','oper2','act3');
    
    
  insert into dm_tr
    (site, area, machine, operation, dm_tr_id)
  values
    ('site','area1','machine1','oper1',9999);

  and my desired output, always had to print even if there is no record in dm_tr
  

  site area1 machine1 oper1 act1  9999
  site area1 machine1 oper1 act2  9999
  site area1 machine1 oper1 act3  9999
  site area1 machine1 oper2 act1 
  site area1 machine1 oper2 act2
  site area1 machine1 oper2 act3

  Thanks in advance.

  Try this

  select a.*,c.draft, b.dm_tr_id from dm_mid a,dm_tr b,dm_top c
  where a.site = b.site (+) and
  a.area = b.area (+) and
  a.machine = b.machine (+) and
  a.operation = b.operation (+) and
  a.area = c.area (+) and
  a.machine = c.machine(+) and
  a.operation = c.operation(+)
  order by 1,2,3,4,5
  /
  
  SITE       AREA       MACHINE    OPERATION  ACTIVITY   D   DM_TR_ID
  ---------- ---------- ---------- ---------- ---------- - ----------
  site       area1      machine1   oper1      act1       Y       9999
  site       area1      machine1   oper1      act2       Y       9999
  site       area1      machine1   oper1      act3       Y       9999
  site       area1      machine1   oper2      act1       N
  site       area1      machine1   oper2      act2       N
  site       area1      machine1   oper2      act3       N
  
  6 rows selected.
  
  Elapsed: 00:00:00.00
  

  Kind regards
  Prazy

• My Confusion of outer join

  Oracle 10g
  
  I have three below queries that use the outer join syntax.
  
  All three queries return exactly the same results
  
  
  (a)

  SELECT TA.ID TA
       , TB.ID TB
    FROM TA
       , TB
   WHERE TA.ID = TB.ID(+);

  (b)

  SELECT TA.ID TA
       , TB.ID TB
    FROM TA
       , TB
   WHERE TB.ID(+) = TA.ID;

  (c)

  SELECT TA.ID TA
       , TB.ID TB
    FROM TA
    LEFT OUTER JOIN TB ON TA.ID = TB.ID;

  I have the right call outer join query because the (+) sign is located on the right and outer join query (b) left because the (+) sign is on the left?
  
  Or is the left join or right determined by the join columns specified in the order select it?
  
  
  
  Create the Script

  CREATE TABLE TA
    (
      "ID" NUMBER
    );
  
  CREATE TABLE TB
    (
      "ID" NUMBER
    );
  
  INSERT INTO TA (ID) VALUES ('1');
  INSERT INTO TA (ID) VALUES ('2');
  INSERT INTO TA (ID) VALUES ('3');
  
  INSERT INTO TB (ID) VALUES ('1');
  INSERT INTO TB (ID) VALUES ('2');
  INSERT INTO TB (ID) VALUES ('4');

  Ben
  
  Published by: benton on August 15, 2012 08:16

  Hi, Ben.

  Benton says:
  ... So is it correct to say that the left or right refers to which side the null values will be displayed?

  Lol it is incorrect to say that the old syntax outer join (using the sign +) is either a left - or a right outer join.

  ... So I need to have the order of the columns in the correct SELECTION so that there is no likelihood of confusion over which side will display NULL values. If I place the columns A and B in the wrong order, that is to say B then a I'll lend to confusion about what will be returned with respect whether left or right.

  No, do not hesitate to organize columns in the select in any way will help your users the most. What is happening in the FROM and WHERE clause, in particular the order in which the tables happens to appear, may not have something to do with the order of the columns in the output.
  Many readers are more comfotable with having NULL columns at the end of a line of output, or at least not at the beginning, so maybe it's one of the reasons for ta.id first in your example. Rearrange the columns in the game any way more than makes sense for people who will look at them.

• That IS NULL is done by replacing the NOT EXISTS operator with an outer join?

  DB version: 10 gr 2
  
  An example of searchoracle.target.com
  
  The query that contains a NOT EXISTS operator

  SELECT M.ModulId, M.Modul FROM MODULE M
  WHERE NOT EXISTS
  ( SELECT PROJEKTMODUL.IdModul 
    FROM PROJEKTMODUL 
    WHERE M.ModulId = PROJEKTMODUL.IdModul 
    AND PROJEKTMODUL.IdProjekt = 23 )

  is replaced by an equivalent outer join query
  

  select distinct
         M.ModulId
       , M.Modul 
    from MODULE M
  left outer
    join PROJEKTMODUL
      on M.ModulId = PROJEKTMODUL.IdModul
     and PROJEKTMODUL.IdProjekt = 23
   where PROJEKTMODUL.IdProjekt is null

  I do not understand what

  PROJEKTMODUL.IdProjekt is null

  fact in the above rewritten query.

  He is to perform an anti-jointure. As far as I know Oracle (starting with version 10g - I thought that release 1) rewritten a Join Anti that resembles your second query of your request NOT EXISTS and NOT IN.

  Look at the results of the query without the predicate IS NULL and see what records (from this set) must be selected to show the same results as the first query. Areas of interest match the predicate IS NULL.

• Outer join does not not as expected left

  Hi all

  I have it here are three tables with values. Mentioned the under outer join query does not and behave like the inner query.

  CREATE TABLE RET_FUND_FEE

  (

  NPTF VARCHAR2 (8 CHAR),

  TPART VARCHAR2 (4 CHAR)

  );

  CREATE TABLE PART_PTF

  (

  Mf_Id VARCHAR2 (6 CHAR) NOT NULL,

  TPARTS VARCHAR2 (4 CHAR) NOT NULL

  );

  CREATE TABLE TFC_FUNDS

  (

  NPTF VARCHAR2 (8 CHAR) NOT NULL,

  MULTIFONDS_ID VARCHAR2 (6 CHAR)

  );

  INSERT INTO RET_FUND_FEE VALUES('111','A');

  INSERT INTO RET_FUND_FEE VALUES('111','D');

  INSERT INTO RET_FUND_FEE VALUES('111','E');

  INSERT INTO PART_PTF VALUES ('MF1', 'A');

  INSERT INTO PART_PTF VALUES ('MF1', 'B');

  INSERT INTO PART_PTF VALUES('MF1','C');

  INSERT INTO TFC_FUNDS VALUES('111','MF1');

  INSERT INTO TFC_FUNDS VALUES('111','MF1');

  INSERT INTO TFC_FUNDS VALUES('111','MF1');

  SELECT A.TPART, B.TPARTS, A.NPTF, B.Mf_Id, C.MULTIFONDS_ID, C.NPTF

  OF RET_FUND_FEE A, PART_PTF B, TFC_FUNDS C

  WHERE A.NPTF = C.NPTF

  AND C.MULTIFONDS_ID = B.Mf_Id

  AND A.TPART = B.TPARTS (+)

  AND C.MULTIFONDS_ID = 'MF1 '.

  AND C.NPTF = '111'

  Here, I expect all records in the RET_FUND_FEE table that I am using outer join.

  But I'm only corresponding chronogram RET_FUND_FEE, PART_PTF as an inner join. Can you get it someone please let me know what lack us.

  Is my version of oracle 11g

  SELECT

  A.TPART, B.TPARTS, A.NPTF, B.Mf_Id, C.MULTIFONDS_ID, C.NPTF

  Of

  PART_PTF B

  Join

  C TFC_FUNDS

  on (C.MULTIFONDS_ID = B.Mf_Id

  AND C.MULTIFONDS_ID = 'MF1 '.

  AND C.NPTF = '111'

  )

  right outer join

  RET_FUND_FEE HAS

  on (A.TPART = B.TPARTS

  and A.NPTF = C.NPTF) - added as correction

  TPART	TPARTS	NPTF	MF_ID	MULTIFONDS_ID	NPTF
  A	A	111	MF1	MF1	111
  A	A	111	MF1	MF1	111
  A	A	111	MF1	MF1	111
  E	-	111	-	-	-
  D	-	111	-	-	-

  or

  SELECT A.TPART, d.TPARTS, A.NPTF, d.Mf_Id, d.MULTIFONDS_ID, d.NPTF

  OF RET_FUND_FEE HAS

  , (

  Select

  *

  PART_PTF b, TFC_FUNDS C

  where B.Mf_Id = C.MULTIFONDS_ID

  AND C.MULTIFONDS_ID = 'MF1 '.

  AND C.NPTF = '111'

  ) d

  WHERE A.TPART = D.TPARTS (+)

  and A.NPTF = D.NPTF (+) - added as a correction

  Sorry had to correct the syntax oracle solution.

  The first one was bad because it would return also B lines that have no match in C.

  Sorry a correction more on these two approaches, missed the second predicate.

• Bad result in a left outer join in 12.1.0.2

  Hallo,

  We discovered a strange behaviour in a query. The query provides values in a column of outer join where there is no corresponding value in the table is attached to the outside.

  When you expand this request by the "ORDER BY" then this query gives the correct result.

  Example:

  SQL > desc tb_a
  Name                                Null?    Typ
  -------------------------------------------- ----------------------------
  ID NOT NULL NUMBER (19)

  SQL > desc tb_b
  Name                                Null?    Typ
  -------------------------------------------- ----------------------------
  CLOSED NOT NULL NUMBER (1)
  ID NOT NULL NUMBER (19)

  CCS_APPLICATION@icw01> select * from tb_a where id in (4148,4141,4195);

  ID
  ----------
  4148
  4141
  4195

  CCS_APPLICATION@icw01> select * from tb_b where id in (4148,4141,4195);

  INTERNAL ID
  ---------- ----------
  4148 0

  CCS_APPLICATION@icw01> SELECT
  2      b.id                            AS b_id,
  3      a.id                            AS a_id,
  4 b.closed AS b_closed
  5
  6 tb_a a
  7 LEFT OUTER JOIN tb_b b ON a.id = b.id
  8 WHERE a.id IN (4148, 4195, 4141)
  9 ORDER BY ASC a.id
  10;

  B_ID ALLOCATION A_ID B_CLOSED
  ---------- ---------- ----------
  4141
  4148 4148 0
  4195

  CCS_APPLICATION@icw01> SELECT
  2      b.id                            AS b_id,
  3      a.id                            AS a_id,
  4 b.closed AS b_closed
  5
  6 tb_a a
  7 LEFT OUTER JOIN tb_b b ON a.id = b.id
  8 WHERE a.id IN (4148, 4195, 4141)
  9 - ORDER BY ASC a.id
  10;

  B_ID ALLOCATION A_ID B_CLOSED
  ---------- ---------- ----------
  4148 4148 0
  4141 4141
  4195 4195

  instance parameter:

  VALUE OF TYPE NAME

  ------------------------------------ ----------- ------------------------------

  compatible string 12.1.0.2.0

  optimizer_features_enable string 12.1.0.2

  After ""alter system set optimizer_features_enable = ' 11.2.0.4 ';"  the query provides the correct result in both cases (ordered and unordered).

  Now the final question: is this a bug?

  1480970 wrote:

  Hallo!  Yes, I searched the Support of Oracle. I found some similar entries, but not an exact match. To fix some issues

  with 12.1.0.2.

  There is another interesting clue when look you on the execution plan:

  Note

  -----

  -the dynamic statistics used: dynamic sampling (level = 2)

  - This is an adaptation plan

  We have disabled (= FALSE) optimizer_adaptive_features and the query provides the correct values.

  This could be a solution for us.

  Looks like a pretty tight match for bug 18430870, even if it affects the two 12.1.0.1 and 12.1.0.2, which contradicts the Martin trial against 12.1.0.1.

  The description of the bug mentions disabling "_projection_pushdown" (set to false) should also be a viable solution, perhaps if you want to give that a try and see if it is a different bug or not.

  There are also a number of one-time fixes already available for download, maybe your version / platform is already covered, if the bug applies.

  Randolf

• outer join on the aggregate query

  This is probably a relatively simple matter, as long as I explain it well enough:

  I have two tables:

  categorycodes and properties

  categorycodes is a lookup table.

  both tables have a catcode field which is a char (1) that contains matching data (only the numbers 1 to 6)

  CREATE

  TABLE CATEGORYCODES

  (

  CATCODE CHAR (1 BYTE) NOT NULL,

  DESCRIPTION VARCHAR2 (25 BYTE) NOT NULL,

  CONSTRAINT CATEGORYCODES_PK PRIMARY KEY (CATCODE) ALLOW

  )

  catCode

  1

  2

  3

  4

  5

  6

  The properties table has approximately 600 000 records. The properties table also has a field named parcelno which is a tank (9).  It contains a string of figures and numbers only.

  What I would like is:

  catCode, count (*)

  1 580

  2 300

  3 3000

  4 235

  5 0

  6 80

  I limited the results of the query to make sure it was a game that would not all catcodes in it.  I have trouble to get the one with zero to display.  I know that this has to do with how I do the join, but I don't know what.

  It is a sample of what I've tried:

  Select i.

  Of

  (select catcode, count (*)

  property p

  where substr (parcelno, 1, 3) = ' 871 "

  Catcode group) i

  outer join right categorycodes cc

  We i.catcode = cc.catcode;

  I'm not worried about the situations where catcode is null in the properties.  Parcelno cannot be null.

  Hello

  Looks like your query should work; except that you won't COUNT (*); That would make each issue at least 1.  COUNT (*) means that count the total number of lines, no matter what is on them, so he'll see the line with just the catcode of the lookup table that matches nothing and which count as 1.  You want to count the number of rows in the table of properties, so expect a column of the properties that cannot be NULL.

  Here is a slightly different way

  SELECT c.catcode

  EARL of (p.catcode) AS cnt

  OF categorycodes c

  P ON p.catcode = c.catcode LEFT OUTER JOIN properties

  AND SUBSTR (p.parcelno

  1

  3

  ) = ' 871 "

  ;

  If the condition about 871' ' part of the join condition, then you don't need a subquery.

  .

  I hope that answers your question.
  If not, post a small example data (CREATE TABLE and only relevant columns, INSERT statements) for all of the tables involved and also publish outcomes from these data.
  Point where the above statement is erroneous results, and explain, using specific examples, how you get the right result of data provided in these places.

  Always say what version of Oracle you are using (for example, 11.2.0.2.0).

  See the FAQ forum: https://forums.oracle.com/message/9362002

• outer join does not give the expected results (missing documents)

  Hello
  I have problem with outer join in OBIEE 10 g.
  
  I'm testing it on the model of simplified database
  
  STAT table with some statistics of TYPEs
  stat_date, id_type, num
  ..
  ..
  26/03/2003, 20: 1
  26/03/2003, 21: 1
  26/03/2003, 23: 1
  26/03/2003, 24, 1
  26/03/2003, 25, 1
  ..
  27/03/2003, 22, 1 <-different date here
  ..
  ..
  
  Table TYPES
  ID_type
  ..
  ..
  20
  21
  22
  23
  24
  25
  ..
  ..
  
  
  Physical model
  STAT >-TYPES
  
  
  AM.
  F1 (only Digital STAT table column (agg. County))
  D1 (STAT_DATE of the STAT table column only)
  D2 (source TYPES)
  
  D1 - inner join - < F1
  D2 - left outer join - < F1
  
  
  I would like to create simple report filtered by date with number of each type.
  I want to get this result
  ID_type County
  ..     ..
  ..     ..
  20 1
  21 1
  22 null
  23 1
  24 1
  25 1
  ..     ..
  ..     ..
  
  
  But I get this
  ID_type County
  ..     ..
  ..     ..
  20 1
  21 1
  23 1
  24 1
  25 1
  ..     ..
  ..     ..
  
  ID_type 22 is missing
  
  
  !!!
  
  I NORMALLY GET RESULT ONLY IF I DELETE RECORD ' 27/03/2003 22' OF MY FACT TABLE.
  
  !!!
  
  
  I have the filter on the date column D1
  STAT_DATE = 26/03/2003 or STAT_DATE is null
  
  
  BI Server running this SQL query
  
  
  Select distinct D1.c1 as c1,
  D1. C2 C2
  Of
  (select count (T17840. NUM) in c1.
  T17867.ID_TYPE C2
  Of
  Left outer join of TYPES T17867
  STAT T17840 on T17840.ID_TYPE = T17867.ID_TYPE
  where (T17840. STAT_DATE in (TO_DATE ('2003-03-26', 'YYYY-MM-DD')) or T17840. STAT_DATE is null)
  T17867.ID_TYPE group
  ) D1
  
  
  Can you give me some advice?
  
  Thank you

  Hi cardel,

  I think that you can achieve your goal if you build a dimension of the array STAT form, you can create a table Alias in the physical layer and then to join the table of facts (external to the left). In this way, I think that you can solve your problem.

  Let me know.

  Kind regards
  Gianluca

• FULLT OUTER JOIN on two different selects

  Hello
  
  I am facing a problem where I don't just get the handle to solve.
  
  I have two pieces of music with different return values. This value should be compared and joined. Simply put, I do an example where the return of the selection is represented as the table1 and table2
  Here is the example:

  table1
  abc_comp_nr, abc_cnt_emp
  12000, 15
  12500, 10
  13000, 30
  13500, 10
  
  table2
  xyz_comp_nr, xyz_cnt_emp
  12000, 15
  13000, 30
  14000, 20
  
  Output should be this:
  comp_nr, abc_cnt_emp, xyz_cnt_emp
  12000, 15, 15
  12500, 10, 0
  13000, 30, 30
  13500, 10, 0
  14000, 0, 20

  comp_nr is the combination of abc_comp_nr or xyz_comp_nr
  
  I tried this but did not work as expected:
  Select?, abc_cnt_emp, xyz_cnt_emp
  Of
  Table1 FULL OUTER JOIN table2
  ON abc_comp_nr = xyz_comp_nr
  
  Hope that someone was in trouble similar and found a good way to work it.
  
  Thanks in advance
  
  Tobias

  Hi, Tobias,.

  Tobias Arnhold wrote:
  Hello

  I am facing a problem where I don't just get the handle to solve.

  I have two pieces of music with different return values. This value should be compared and joined. Simply put, I do an example where the return of the selection is represented as the table1 and table2
  Here is the example:

  table1
  abc_comp_nr, abc_cnt_emp
  12000, 15
  12500, 10
  13000, 30
  13500, 10
  
  table2
  xyz_comp_nr, xyz_cnt_emp
  12000, 15
  13000, 30
  14000, 20
  

  Whenever you have a problem, post CREATE TABLE and INSERT statements for the sample data.
  See the FAQ forum {message identifier: = 9360002}

  Output should be this:
  comp_nr, abc_cnt_emp, xyz_cnt_emp
  12000, 15, 15
  12500, 10, 0
  13000, 30, 30
  13500, 10, 0
  14000, 0, 20
  

  comp_nr is the combination of abc_comp_nr or xyz_comp_nr

  I tried this but did not work as expected:
  Select?, abc_cnt_emp, xyz_cnt_emp
  Of
  Table1 FULL OUTER JOIN table2
  ON abc_comp_nr = xyz_comp_nr

  Instead of??? you want to

  NVL (abc_comp_nr, xyz_comp_nr)
  

  The first column is supposed to be the common identifier, abc_comp_nr, or xyz_comp_nr. It doesn't matter which, since the join condition

  ON abc_comp_nr = xyz_comp_nr
  

  said they will be the same, except that it's a FULL OUTER JOIN, if one of them might be null. If you want to display the non NULL when one of them is NULL and or the other (any) when they are both present. It's just that NVL.

  You can use NVL on other columns, too, to display 0 instead of NULL when there is no match:

  SELECT   NVL ( abc_comp_nr, xyz_comp_nr)     AS comp_nr
  ,        NVL ( abc_cnt_emp, 0)               AS abc_cnt_emp
  ,      NVL ( xyz_cnt_emp, 0)               AS xyz_cnt_emp
  FROM              table1
  FULL OUTER JOIN  table2  ON  abc_comp_nr  = xyz_comp_nr
  ;
  

  However, this will display 0 instead of NULL for abc_cnt_emp and xyz_cnt_emp when these columns are NULL for a reason, not just at the time when they do not reach the join condition. If the original paintings contain NULL values and you should keep NULL values in the result set, but you want NULLs which were created by the outer join to show as 0, then you may want something like this:

  SELECT   NVL ( abc_comp_nr, xyz_comp_nr)     AS comp_nr
  ,        NVL2 ( abc_comp_nr, abc_cnt_emp, 0)     AS abc_cnt_emp
  ,      NVL2 ( xyz_comp_nr, xyz_cnt_emp, 0)     AS xyz_cnt_emp
  FROM              table1
  FULL OUTER JOIN  table2  ON  abc_comp_nr  = xyz_comp_nr
  ;
  

  Whenever you use FULL OUTER JOIN, there is an excellent chance that you will want to use NVL (or his brother younger, stronger, COALESCE, or their cousin NVL2) as well.

  Published by: Frank Kulash on 3 March 2013 14:49

• How do the query select outer join to a report of the APEX

  Hi all
  
  I'm Ann.
  
  I have a select statement that is used to calculate statistics for a month (October 2012 in this example)
  Select ph.phase_number
  sum ((case
  WHEN ph.date_finished IS NULL or ph.date_finished > last_day (TO_DATE (' ' Oct 2012 ', ' MY YYYY' "))
  THEN last_day (TO_DATE (' ' Oct 2012 ', ' MY YYYY' "))
  Of OTHER ph.date_finished
  END)
  (-ph.date_started + 1) / count (def.def_id) as avg_days
  Ph phase_membership
  inner join court_engagement this on ph.mpm_eng_id = ce.engagement_id
  join in-house defendant def on ce.defendant_id = def.def_id
  where def.active = 1
  and ph.date_started < = last_day (TO_DATE (' ' Oct 2012 ', ' MY YYYY' "))
  and ph.active = 1
  and UPPER (ce.court_name) LIKE '% '.
  Rollup Group (phase_number)
  ;
  
  The result is as below
  Phase_Number AVG_DAYS
  Phase One 8.6666666666666667
  Phase two 14.6
  Phase three 12
  11.4615365
  
  I have another list of selection mainly the list of months between two date value.
  Select to_char (which_month, 'LUN YYYY') as display_month
  de)
  Select add_months (to_date (' ' August 2012 ', ' MY YYYY' "), rownum-1) which_month
  of object
  where
  rownum < = months_between (to_date (' ' Oct 2012 ', ' MY YYYY' "), add_months (to_date (' ' August 2012", "MY YYYY"), - 1))
  order of which_month)
  
  The query result is as below
  
  DISPLAY_MONTH
  
  AUGUST 2012
  SEP 2012
  OCT 2012
  
  Is it possible I can join these two select statement above to generate a comparable result:
  
  Days of month Phase number Avg
  August 2012 Phase One 8.666
  Sep 2012 Phase One 7.66
  Oct 2012 Phase One 5,66
  August 2012 Phase two 8.666
  Sep 2012 Phase two 7.66
  Oct 2012 Phase two 5,66
  August 2012 Phase three 8.666
  Sep 2012 Phase three 7.66
  Oct 2012 Phase three 5,66
  
  Or
  Days of month Phase number Avg
  August 2012 Phase One 8.666
  August 2012 Phase two 7.66
  August 2012 Phase three 5,66
  Sep 2012 Phase One 8.666
  Sep 2012 Phase two 7.66
  Sep 2012 Phase three 5,66
  Oct 2012 Phase One 8.666
  Oct 2012 Phase two 7.66
  Oct 2012 Phase three 5,66
  
  And it can be controlled by Phase number or month.
  My other colleague suggested I should use a left outer join, but after having tried many ways, I'm still stuck.
  
  I tried select is
  Select a.display_month, b.* in)
  Select to_char (which_month, 'LUN YYYY') as display_month
  de)
  Select add_months (to_date (' ' August 2012 ', ' MY YYYY' "), rownum-1) which_month
  of object
  where
  rownum < = months_between (to_date (' ' Oct 2012 ', ' MY YYYY' "), add_months (to_date (' ' August 2012", "MY YYYY"), - 1))
  order which_month)) a left outer join
  
  (Select to_char (ph.date_finished, 'MY YYYY') as join_month, ph.phase_number)
  sum ((case
  WHEN ph.date_finished IS NULL or ph.date_finished > last_day (TO_DATE (a.display_month, 'MY YYYY'))
  THEN last_day (TO_DATE (a.display_month, 'MY YYYY'))
  Of OTHER ph.date_finished
  END)
  (-ph.date_started + 1) / count (def.def_id) as avg_days
  Ph phase_membership
  inner join court_engagement this on ph.mpm_eng_id = ce.engagement_id
  join in-house defendant def on ce.defendant_id = def.def_id
  where def.active = 1
  and ph.date_started < = last_day (TO_DATE (a.display_month, 'MY YYYY'))
  and ph.active = 1
  and UPPER (ce.court_name) LIKE '% '.
  To_char (ph.date_finished, 'MY YYYY'), group (phase_number) rollup) b
  On a.display_month = b.join_month
  
  but then I get an error
  SQL error: ORA-00904: "A." "" DISPLAY_MONTH ": invalid identifier
  
  I need to view a report on the APEX with option for people to download at least format CSV.
  Already 1 inteactive report in the page, so I don't think adds another interactive report without using the iframe trick.
  
  If any of you have any ideas, please help.
  
  Thank you very much.
  
  Ann

  Hello Ann,.

  Frank has done a very good job. I am also impressed.

  Is in regard to your correction to his question, the problem is on this replacement you did

  last_day(TO_DATE(am.which_month,'MON YYYY'))
  

  AM.which_month is already a date type, and you don't need to convert it to this day.
  Here is the correct way:

  last_day(am.which_month)
  

  There are also sometimes with the data you've posted have no line for this month. So I also added a function NVL to display 0 under avg_days for these cases.

  Here is my corrected query:

  DEFINE startmonth = "Aug 2012";
  DEFINE endmonth   = "Oct 2012";
  WITH  all_months  AS
  (
     SELECT ADD_MONTHS(to_date('&startmonth','MON YYYY'), ROWNUM-1) AS which_month
     ,      ADD_MONTHS(to_date('&startmonth','MON YYYY'), ROWNUM  ) AS next_month
     from all_objects
     where
     rownum <= months_between(to_date('&endmonth','MON YYYY'), add_months(to_date('&startmonth','MON YYYY'), -1))
  )
  SELECT TO_CHAR (am.which_month, 'Mon YYYY')  AS month
       , ph.phase_number
       , NVL(sum ( (CASE
                   WHEN ph.date_finished IS NULL OR ph.date_finished > last_day(am.which_month)
                   THEN last_day(am.which_month)
                   ELSE ph.date_finished
                END
               ) - ph.date_started + 1
             ) / count(def.def_id), 0) as avg_days
    FROM all_months am
         LEFT OUTER JOIN  a_phase_membership  ph  PARTITION BY (ph.phase_number)
            ON  am.which_month <= ph.date_started
            AND am.next_month  >  ph.date_started
            AND ph.date_started <= last_day(am.which_month)  -- May not be needed
            AND ph.active = 1
         LEFT OUTER JOIN  a_engagement  ce
            ON  ph.mpm_eng_id = ce.engagement_id
            AND ce.court_name IS NOT NULL  -- or something involving LIKE
         LEFT OUTER join  a_defendant     def
            ON  ce.defendant_id = def.def_id
            AND def.active = 1
   GROUP BY ROLLUP(phase_number, am.which_month)
   ORDER BY  am.which_month
          ,  ph.phase_number
  ;
  
  The output is:
  MONTH    PHASE_NUMBER           AVG_DAYS
  -------- -------------------- ----------
  Aug 2012 PHASE ONE                     0
  Aug 2012 PHASE THREE                   0
  Aug 2012 PHASE TWO                     0
  Sep 2012 PHASE ONE                    12
  Sep 2012 PHASE THREE                   1
  Sep 2012 PHASE TWO                     9
  Oct 2012 PHASE ONE                     8
  Oct 2012 PHASE THREE                   0
  Oct 2012 PHASE TWO                    14
           PHASE ONE                    11
           PHASE THREE                   1
           PHASE TWO                  11.5
                                9.71428571
  

  I don't know if that's really what you want. In the case check it and let me know.

  Kind regards.
  Al