outer join on the aggregate query

Similar Questions

• SSRS for lack of outer join with the Oracle data source

  It seems to be a problem with the Oracle driver used in the Reporting SERVICES query designer.

  When you use an Oracle data source, if I create an outer join in the graphic designer, it automatically inserts '{OJ' before the join and '} ' after her.  This is an incorrect syntax for Oracle and refuses to start.  The curly braces and the JO editable in designer text, but if I go back to the graphic designer and immediately to reintegrate them.

  Only, this has started to happen a year or two ago - before that it worked, but with the old (+) syntax.

  Can it not be healed?  It makes things very difficult.

  -Geoff

  Hi Geoff,

  Thanks for posting in the Microsoft Community.

  However, the question you posted would be better suited in the Forums of the Oracle Support; We recommend that you post your query in Oracle Support Forums to get help:

  https://forums.Oracle.com/forums/main.jspa;JSESSIONID=8d92100c30d8fb401bcbd10b46c38c9ddf1a3242549a.e34SbxmSbNyKai0Lc3mPbhmSc3aNe0? CategoryID = 84

  If you have any other questions or you need Windows guru, do not hesitate to post your questions and we will be happy to help you.

• Outer join on the Filter Condition

  Hello

  I would like to show all rows in a query for two conditions to filter on the 'status '. I used an outer join on the outer query to return the values inside the inline query, but it returns not to return all rows.

  create table (test)
  ID number,
  start_date date,
  status varchar2 (1).
  number of amount,
  number of cust_type
  );

  
  Insert test values (001, June 1, 2014 ", am ', 189, 78");
  Insert test values (001, March 26, 2014 ", 'R', 175, 4");
  Insert test values (001, December 1, 2014 ", 'R', 89, 91");
  Insert test values ("001, 13 August 2014 ', 'J', 19, 2);
  Insert test values (001, 12 August 2014 ', 'E', 19, 2);

  Insert test values (002, January 1, 2014 ', 'R', 17, 4);
  Insert test values (002, 26 June 2014 ", 'R', 175, 4");
  Insert test values (' 002, February 1, 2014 ', 'J', 9, 8);
  Insert test values (002, 13 October 2014 ', ', 190, 2);

  Insert test values (' 003, June 1, 2014 ', 'J', 189, 78);
  Insert test values (003, March 26, 2014 ", 'R', 175, 4");
  Insert test values (003, December 1, 2014 ", 'R', 89, 91");
  Insert test values (' 003, 13 August 2014 ', 'J', 19, 2);
  Insert test values (' 003, 12 August 2014 ', 'J', 19, 2);

  commit;

  Select i.ids, i.start_date, i.cust_type, i.amount as gross_amount, i.amount + ii.amount as net_amount
  test I have,
  (select id, amount)
  of the test
  where (status = status or 'R' = 'J')) ii
  where i.cust_type in (4, 78, 91, 2, 4, 8)
  and i.ids (+) = ii.ids
  and i.status (+) to (', 'E')
  "and i.start_date between 1 January 14 ' and 31-dec-2014."

  The query above exclude ID 003 because there is that no status (M / E) butI want always display all codes, even if the condition is not consistent

  Hello

  Want results from an inner join to match identifiers, but the results of an outer join for identifiers that are not?  In other words, you want to include the rows of the table I only when one of the following is true:

  The line of table I have ticked all the boxes, or

  No line in not table i with the same ID meets all conditions

  ?

  If so, here's a way to do it:

  WITH got_rnk AS

  (

  Select i.ids, i.start_date, i.cust_type

  , nvl2 (ii.ids, i.amount, 0) as gross_amount

  , i.amount + nvl (ii.amount, 0) as net_amount

  dense_rank () over (partition of i.ids

  order of nvl2 (ii.ids, 'A', 'B')

  ) as rnk

  I have test

  (in left outer join

  SELECT ID, amount

  of the test

  where status ('R', 'J')

  ) ii on i.ids = ii.ids

  and i.cust_type in (4, 78, 91, 2, 4, 8)

  and i.start_date between to_date (January 1, 2014 ", 'dd-mon-yyyy')

  and to_date (31-dec-2014', 'dd-mon-yyyy')

  and i.status in (', 'E')

  )

  SELECT ID, start_date, cust_type

  gross_amount, net_amount

  of got_rnk

  where rnk = 1

  ;

  Results of your sample data:

  ID START_DATE GROSS_AMOUNT NET_AMOUNT CUST_TYPE

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

  1 1 June 2014 78 189 364

  1 1 June 2014 78 189 278

  1 1 June 2014 78 189 208

  1 12 August 2014 2 19 38

  1 12 August 2014 2 19 108

  1 12 August 2014 2 19 194

  2 13 October 2014 2 190 207

  2 13 October 2014 2 190 365

  2 13 October 2014 2 190 199

  3 March 26, 2014 4 0 175

  3 1 June 2014 78 0 189

  3 12 August 2014 2 0 19

  3 August 13, 2014 2 0 19

  3 1 December 2014 91 0 89

• outer join in the view criteria

  Hello world

  I use jdeveloper 11.1.1.6.0

  I want to create outer join in the view criteria, so I had to substitute getCriteriaItemClause ViewObjImpl.java method

  my query is like this:

  SELECT t1. row_no, t2.type FROM table_1 t1, t2 from table_2

  WHERE t1. row_no (+) = t2. row_no

  and t1.user_Id (+) =: bindVar

  table_1 (row_no, user_Id)

  table_2 (row_no, type)

  the part " t1.user_Id (+) =: bindVar " should be put in the display criteria

  the method:

  @Override

  public String getCriteriaItemClause (ViewCriteriaItem viewCriteriaItem) {}

  String newQeury = this.getEntityDef (0) .getAliasName () + "." + viewCriteriaItem.getColumnName () + "(+) =:bindVar .

  Return newQeury;

  }

  but when I run the criteria see exception below raises:

  oracle.jbo.expr.JISyntaxError: Houston-36000: an unexpected token expression is found.

  When I remove the sign "(+)" of newQeury

  String newQeury = this.getEntityDef (0) .getAliasName () + "." + viewCriteriaItem.getColumnName () + "=:bindVar .  remove the sign "(+)" before =:bindVar

  raises the exception below

  Houston-29000: Unexpected exception caught: oracle.jbo.expr.JIEvalException, msg = Houston-25077: t1 name not found in the given object: ViewRow [[5-1-5] oracle.jbo.Key].

  Habib

  This is a duplicate of https://forums.oracle.com/thread/2577092

  Please do not post your questions several times.

  A possible solution posted to the original thread.

  Timo

• using outer joins if the two column is null? Use only (+)

  Hi all

  create the table xxc_tr_num (tl_number number, tr_no number tl_no_id);

  insert into xxc_tr_num values (123,100,222);

  insert into xxc_tr_num values (124,100,333);

  create the table xxc_od_tab (tl_number number, tl_id number);

  insert into xxc_od_tab values (123,001);

  insert into xxc_od_tab values (null, null);

  create table xxc_oth_tab (name varchar2 (10), number of tl_id);

  insert into xxc_oth_tab values('abc',,001);

  insert into xxc_oth_tab values (null, null);

  Wait it out put

  tr_no tl_no_id name

  100 222 abc

  100 333

  using outer joins if the two column is null? use only please of outer joins

  And I tried to use outer joins on both tl_id column but not get values and I use have County (tr_no ) > 1

  Rajesh123 wrote:

  Thank you Kiss it is not possible to use having clause?

  You need to understand the functioning of the group. If you will not be asked this question.

  Check this box

  SQL> select tr_no,
    2         tl_no_id,
    3         count(*)
    4    from xxc_tr_num a,
    5         xxc_od_tab b,
    6         xxc_oth_tab c
    7   where a.tl_number = b.tl_number(+)
    8     and b.tl_id = c.tl_id(+)
    9   group
   10      by tr_no
   11       , tl_no_id;
  
       TR_NO   TL_NO_ID   COUNT(*)
  ---------- ---------- ----------
         100        333          1
         100        222          1
  

  See what returns the count? You have grouped according to TR_NO and TL_NO_ID. You must take into consideration the TL_NO_ID just put COUNT (TR_NO) does not increase the NUMBER of the whole group. To get the NUMBER on the whole group, I used the analytical function and did. Like this, see the number of the analytical function here

  SQL> select tr_no,
    2         tl_no_id,
    3         count(*),
    4         count(*) over(partition by tr_no)
    5    from xxc_tr_num a,
    6         xxc_od_tab b,
    7         xxc_oth_tab c
    8   where a.tl_number = b.tl_number(+)
    9     and b.tl_id = c.tl_id(+)
   10   group
   11      by tr_no
   12       , tl_no_id;
  
       TR_NO   TL_NO_ID   COUNT(*) COUNT(*)OVER(PARTITIONBYTR_NO)
  ---------- ---------- ---------- ------------------------------
         100        222          1                              2
         100        333          1                              2
  

  So to answer your question, yes you can't do in the HAVING clause...

• How to specify an outer join in the query designer

  I quickly built a large query with the query designer, and it works very well, but does not produce 9 disks on 2200 - so I need to change to an outer join.
  
  My problem is that I can't find a way to specify an outer join. I tried to change the query, query builder formed, but it won't let me.
  
  I know that I can capture the query, edit and run it in sql developer and that works fine, but I want to use the Query Builder as it is the fastest way I know to quickly add and remove items - my users don't know exactly what that it to include.
  
  I know it's kind of a trivial question, but I searched the forum, manual, the web and no luck
  
  Thanks for any help you can give
  
  Glenn

  Hi Glenn,.

  When you use the query designer to create the sql statement, you must click on the empty box to the right of the fields in each table to create the join. When you are finished, you should see a line drawn between the two tables. Click on this line, and a small window appears allowing you to select inner outer or right outer joins, left.

  When the report was created, however, you must change the statement. For example, if you use the query designer to join DEPT at EMP based on the DEPTNO field on the two tables, you get:

  select      "DEPT"."DEPTNO" as "DEPTNO",
        "DEPT"."DNAME" as "DNAME",
        "DEPT"."LOC" as "LOC",
        "EMP"."EMPNO" as "EMPNO",
        "EMP"."ENAME" as "ENAME",
        "EMP"."JOB" as "JOB",
        "EMP"."MGR" as "MGR",
        "EMP"."HIREDATE" as "HIREDATE"
   from      "EMP" "EMP",
        "DEPT" "DEPT"
   where   "DEPT"."DEPTNO"="EMP"."DEPTNO"
  

  This is to change:

  select      "DEPT"."DEPTNO" as "DEPTNO",
        "DEPT"."DNAME" as "DNAME",
        "DEPT"."LOC" as "LOC",
        "EMP"."EMPNO" as "EMPNO",
        "EMP"."ENAME" as "ENAME",
        "EMP"."JOB" as "JOB",
        "EMP"."MGR" as "MGR",
        "EMP"."HIREDATE" as "HIREDATE"
   from      "EMP" "EMP"
        LEFT OUTER JOIN "DEPT" "DEPT" ON "DEPT"."DEPTNO"="EMP"."DEPTNO"
  

  Andy

• Outer joins to the left... Please help!

  Hmm. Ive had trouble with this one for a while and have read and re-read docs etc.

  IM under 11.2 XE and its Apex 4.2 application...

  Its a simple query with outer joins... but I can't simply to do it right after several hours of trying! ID be very grateful if anyone can lend a hand...

  For the query below. I am looking for the following:

  Complete list of entities, a sum of the values if they exist (and obviously empty if they don't), filtered by a table of SUM choice only the records that match a value in the lookup table.

  Select

  e.ENTITY as ENTITY,

  Sum (p.amount) as Forecast_Income,

  Sum (BR.bri_credit) as Actual_Income,

  Sum (BR.bri_debit) as Actual_Expenses

  Of

  e entities

  LEFT OUTER JOIN payments p

  WE (e.entity_id = p.entity_id)

  LEFT OUTER JOIN pmt_stat_lookup ps

  WE (p.status_id = ps.status_id

  and ps.forecast = 'Y')

  LEFT OUTER JOIN bri_recon br

  WE (e.entity_id = br.entity_id)

  E.entity group

  My problem with the join where I join p.status_id for ps.status_id and ps.forecast = 'Y '...   If I run the above statement, it seems does not account for this filter criteria and summarizes all records of PAYMENTS, regardless of...  If I move outside the outer join clause, it only brings back the records of PAYMENTS where there is a join. Ive tried to join in a different order etc... but in vain...

  If someone could help, Id be very grateful.

  Kind regards

  Richard

  Hello Richard

  Try this query:

  SELECT e.entity as ENTITY

  sum (PO.amount) as Forecast_Income

  sum (br.bri_credit) as Actual_Income

  sum (br.bri_debit) as Actual_Expenses

  Of

  E ENTITIES

  LEFT OUTER

  JOIN (SELECT P.entity_id )

  P.amount

  PAYMENTS P

  JOIN PMT_STAT_LOOKUP PS

  ON p.status_id = ps.status_id

  AND ps.forecast = 'Y '.

  ) IN.

  WE e.entity_id = in. entity_id

  LEFT OUTER

  JOIN BR BRI_RECON

  ON e.entity_id = br.entity_id

  E.entity GROUP

  ;

  I hope it helps.

  Best regards, David

  Post edited by: David Berger

• outer join with the additional constraint

  Hello

  With the help of Oracle 11 g R2.

  I would of outer join tables 2 together and put down restrictions on the types of records that are returned in the query result. Here's a mock-up of the tables and data.

  create table aaa (col1 number not null, col2 number not null)

  create table bbb (col1 number not null, col2 number not null)

  insert into values of aaa (1: 80)

  insert into values aaa (2, 90)

  insert into values aaa (3, 80)

  insert into values aaa (4, 90)

  insert into values aaa (5, 80)

  
  insert into bbb values (3, 600)

  insert into values of bbb (4, 700)

  This is the query

  
  select a.col1, a.col2, b.col1, b.col2
  from aaa a, bbb b
  where a.col1 = b.col1 (+)
  and   (a.col2, b.col2) <> ((90, 700))
  

  The result of the query is as follows.

  col1 col1 col2 col2

  1 80

  3 80 3 600

  5 80

  Where col1 = 4 has been deleted, which is an expected result. However, where col1 = 2 has also been removed, which is not a desired outcome. Your response is appreciated.

  Hello

  Here is a way that works for the given sample data:

  SELECT *.

  AAA a

  LEFT OUTER JOIN bbb b ON a.col1 = b.col1

  WHERE the NVL (a.col2, 0) <> 90

  OR NVL (b.col2, 0) <> 700

  ;

  I don't know if that will satisfy your requirements with other data, since you didn't say what your needs are.

  Whenever you have a WHERE clause is applied after the outer join, all columns of the table in option (table bbb in this example) must be used in an NVL, NVL2 or something like a CASE expression that takes into account null values; otherwise, the effect will be the same as an inner join.

• outer join on the left, needs improvement

  Two table t1 and t2 where t1.col1 = t2.col2

  create table t1 ( col1 varchar2(20));
  insert into t1 values('aa');
  insert into t1 values('bb');
  insert into t1 values('cc');
  insert into t1 values('dd');
  insert into t1 values('ee');
  
  
  create table t2 ( col2 varchar2(20) , col3 varchar2(20));
  insert into t2 values('aa','q_aa');
  insert into t2 values('bb',' ');
  insert into t2 values('cc','q_cc');
  insert into t2 values('dd',' ');
  
  
  here is my join query:-
  select t1.col1,t2.col2,t2.col3,
  case when t2.col3 is not null then t2.col3
       when t2.col3 is null then t1.col1 end as name1                                   
  from t1 left outer join t2
  on t1.col1 =t2.col2

  real to come result
  COL1 ^ ^ COL2 ^ ^ COL3 ^ ^ NAME1
  AA ^ ^ ^ aa ^ ^ ^ q_aa ^ ^ ^ q_aa
  BB ^ ^ ^ bb
  CC ^ ^ ^ cc ^ ^ ^ q_cc ^ ^ ^ q_cc
  JJ ^ ^ ^ JJ
  ee^^^^^^^^^^^^^^^^ee
  
  requirement of result
  COL1 ^ ^ COL2 ^ ^ COL3 ^ ^ ^ NAME1
  AA ^ ^ ^ aa ^ ^ ^ q_aa ^ ^ ^ q_aa
  BB ^ ^ ^ bb ^ ^ ^ bb
  CC ^ ^ ^ cc ^ ^ ^ q_cc ^ ^ ^ q_cc
  JJ ^ ^ ^ DD ^ ^ ^ JJ
  ee^^^^^^^^^^^^^^^^^ ee
  
  
  and how can copy paste the result of sql * more... box when I copy paste here it automatically omit the space, as above

  The problem is values col3 to the 'bb' and 'dd' lines are not null, because they have a space in them. You can either remove the space and use:

  nvl(t2.col3,t1.col1) as name1
  

  or tim space out:

  nvl(trim(t2.col3),t1.col1) as name1
  

• left outer join and the where clause for the table to the right

  I want to join two tables a and b, where a is a must and b is a result set in option. When I use a left outer join to a to b, I want to achieve:
  
  1. Select a single column, two columns of b (not the join columns)
  2 - even if theres no friendly on the join column does not return data from one.
  3. If there is a match applies when the criteria on column b (table in option)
  
  so, how can I avoid no_data_found in this case? When I apply where criteria for b, so it does not return the data from one, which is a must.

  Sounds like a regular outer join to me...

  select a.col1, b.col2, b.col2
  from   tableA a
         left outer join tableB b
         on (a.id = b.id and b.colX = 'X')
  

• outer join and OR condition

  Hello world
  
  I have a query that is not working properly. I'm trying to join two tables and I or condition and outer join
  I get the error message

   ORA-01719: outer join operator (+) not allowed in operand of OR or IN
  
  my query is as follow
  
  SELECT l.* FROM table1 l, table2 map
  WHERE (NVL(l.id, '-99') = NVL(map.id(+), NVL(l.id, '-99'))
          AND  l.TYPE = 'TKE')
        ) OR 
      (NVL(l.id, '-99') = NVL(map.id(+), NVL(l.id, '-99'))
  AND NVL(l.type, '-99') = NVL(map.type(+), NVL(l.type, '-99'))
  AND NVL(l.pos_22, '-99') = NVL(map.pos_22(+),NVL(l.pos_22, '-99'))
  AND NVL(l.pos_27, '-99') = NVL(map.pos_27(+),NVL(l.pos_27, '-99'))
  AND NVL(l.pos_17, '-99') = NVL(map.pos_17(+),NVL(l.pos_17, '-99'))
  AND NVL(l.pos_18, '-99') = NVL(map.pos_18(+),NVL(l.pos_18, '-99'))
  AND NVL(l.pos_12, '-99') = NVL(map.pos_12(+),NVL(l.pos_12, '-99'))
  AND NVL(l.pos_16, '-99') = NVL(map.pos_16(+),NVL(l.pos_16, '-99'))
  AND NVL(l.pos_34, '-99') = NVL(map.pos_34(+),NVL(l.pos_34, '-99'))
  AND NVL(l.pos_43, '-99') = NVL(map.pos_43(+),NVL(l.pos_43, '-99'))
  AND NVL(l.pos_14, '-99') = NVL(map.pos_14(+),NVL(l.pos_14, '-99'))
  AND NVL(l.status_cd, '-99') = NVL(map.status(+),NVL(l.status_cd, '-99'))
  AND NVL(l.action_cd,'-99') = NVL(map.action_cd(+), NVL(l.action_cd,'-99'))
  AND NVL(l.sys,'-99') = NVL(map.sys(+), NVL(l.sys,'-99'))
  AND NVL(l.ind,'-99') = NVL(map.ind(+), NVL(l.ind,'-99'))
    )

  So, basically in the above query I say that if the type = "TKE" do not join on all columns of pos. If type is not TKE then do all the joints on all columns POS etc.
  How do I rewrite query so that I can use OR and the outer JOIN in the same query?

  Hello

  It is one of several good reasons to use the ANSI join syntax. Move all the join conditions in the clause on, not the WHERE clause, and you do not use the sign +.

  SELECT  l.*
  FROM           table1 l
  LEFT OUTER JOIN  table2 map  ON   (    NVL(l.id, '-99') = COALESCE (map.id, l.id, '-99')
                                      AND  l.TYPE = 'TKE'
                        )
                               OR  (     NVL (l.id, '-99') = COALESCE (map.id, l.id, '-99')
                            AND   NVL (l.type, '-99') = COALESCE (map.type), l.type, '-99')
                       ...
  

  Instead of nesting the NVL statements, use COALESCE, as illustrated above. It is independent of the issue of the join.

• Oracle 8i outer join

  The following query is supposed to return all the rows in kpi_reports where the conditions are met and its values corresponding to kpi_master and kpi_report_requests, where the values exist. However, it only returns the lines containing the values kpi_report_request, doing so as an inner join.
  
  SELECT kr.report_id, kr.report_name, kr.kpi_id, km.kpi_description, krr.requested_date
  OF kpi_reports kr, kpi_master km, kpi_report_requests krr
  WHERE kr.kpi_id = km.kpi_id
  AND kr.dsp_id = 9
  AND kr.report_name IS NOT NULL
  AND kr.report_id = krr.report_id (+)
  AND krr.dealer_id = 26402
  ORDER BY kr.report_name
  
  I have reviewed a number of tutorials online brush upward on this thus shown a colleague, and I think Miss me something, maybe you can not do an inner join and outer join in the same query? Don't seem to make sense to me, but all the examples I've seen consist of two tables with an outer join.
  
  Thanks in advance!

  Something like this maybe? (Not sure about Oracle 8, but I do similar things in 9i)

  Phil

• He had to know the right outer join using...

  He had to know the right outer join using...

  For example: first query left outer join for the emp table: SELECT EMPNO, ENAME, D.DEPTNO FROM EMP E, Department D WHERE the E.DEPTNO = D.DEPTNO)

  Second query left outer join for the Dept table: SELECT EMPNO, ENAME, D.DEPTNO FROM EMP E, Department D WHERE the D.DEPTNO = E.DEPTNO)

  In the example above I just Exchange where condition condition to get an outer join of two table with a left outer join itself. Wat is use right outer join, instead, I can swap the status of table name for the result. Please suggest...

  Hello

  chan001 wrote:

  He had to know the right outer join using...

  For example: first query left outer join for the emp table: SELECT EMPNO, ENAME, D.DEPTNO FROM EMP E, Department D WHERE the E.DEPTNO = D.DEPTNO)

  Second query left outer join for the Dept table: SELECT EMPNO, ENAME, D.DEPTNO FROM EMP E, Department D WHERE the D.DEPTNO = E.DEPTNO)

  In the example above I just Exchange where condition condition to get an outer join of two table with a left outer join itself. Wat is use right outer join, instead, I can swap the status of table name for the result. Please suggest...

  The two examples above use the old syntax outer join of Oracle. (I guess there should be a sign inside the parentheses, e.g.. +)

  ...  WHERE E.DEPTNO = D.DEPTNO (+)

  )

  The LEFT OUTER JOIN and RIGHT OUTER JOIN terms apply only to the ANSI join syntax, e. g.

  .

  .. FROM EMP E

  DEPT LEFT OUTER JOIN D ON E.DEPTNO = D.DEPTNO

  As Blushadow said above, there's no real point in having LEFT OUTER JOIN and RIGHT OUTER JOIN;  What you can do with one (or a combination of both) can be done with the other.  Most people use LEFT OUTER JOIN systematically and never use RIGHT OUTER JOIN.

  There are situations where using a combination of the two would mean a little less striking, but only a little less and this kind of situation is very common, and one may wonder if the somewhat shorter code is more specific.  I suggest that forget you RIGHT OUTER JOIN.

• Outer join to find the missing?

  I have 2 tables:
  
  with stud_courses as
  (
  Select studentid 111, 2, 2201 coursecategory courseid, 'A' GRADE of double
  Union
  Select studentid 111, 2, 2203 coursecategory courseid, CATEGORY "B" of the double
  Union
  Select studentid 111, 4 coursecategory, 2121 courseid, 'A' GRADE of double
  Union
  Select studentid 222, 3 coursecategory, 1111 courseid, 'A' GRADE of double
  Union
  Select studentid 222, 3 coursecategory, 1112 courseid, 'A' GRADE of double
  )
  course_categories as
  (
  Select 2 coursecategory, 2201 courseid double
  Union
  Select 2 coursecategory, 2202 courseid double
  Union
  Select 2 coursecategory, 2203 courseid double
  Union
  Select 2 coursecategory, 2204 courseid double
  Union
  Select 3, 1110 coursecategory courseid double
  Union
  Select 3 coursecategory, 1111 courseid double
  Union
  Select 3 coursecategory, 1112 courseid double
  Union
  Select 3 coursecategory, 1113 courseid double
  Union
  Select 4 coursecategory, 2120 courseid double
  Union
  Select 4 coursecategory, 2121 courseid double
  Union
  Select 4 coursecategory, 2122 courseid double
  Union
  Select 4 coursecategory, 2123 courseid double
  )
  
  For each student, wants to add the missing courseids for the coursecategory he takes. So student because 111 is in coursecategory 2 I should be able to show all results for him, including the ID of course 2202 and 2204 of category 2 and 2120, 2122, 2123 for this child
  
  My end result would be something like:
  
  StudentID, courseid, coursecategory, grade
  111,2,2201, "A".
  111,2,2201, '
  111,2,2203, 'B '.
  111,2,2201, '
  111,4,2120, '
  111,4,2121, "A".
  111,4,2122, '
  111,4,2123, '
  222,3,1110, '
  222,3,1111, "A".
  222,3,1112, "A".
  222,3,1113, '
  
  I know how to do it with a Union and not exists, but I was wondering if I could do it with an outer join.
  
  I would appreciate your help.
  
  Thank you

  Hello

  Yes, you can do it with an outer join. The tricky thing is you need outside to join the table real stud_courses to something that has valid all combinations of the studnents and prices. There is no single table that contains it; the following query will receive these combinations by making a cross join:

  WITH     all_students     AS
  (
       SELECT DISTINCT     studentid
       FROM          stud_courses
  )
  ,     got_category_cnt     AS
  (
       SELECT     all_s.studentid
       ,     c.coursecategory
       ,     c.courseid
       ,     NVL ( s.grade
                , 'M'
                )               AS grade
       ,     COUNT (s.grade) OVER ( PARTITION BY  all_s.studentid
                                   ,          c.coursecategory
                           )      AS category_cnt
       FROM           all_students         all_s
       CROSS JOIN      course_categories  c
       LEFT OUTER JOIN      stud_courses         s  ON    s.studentid     = all_s.studentid
                                           AND   s.courseid          = c.courseid
  )
  SELECT       studentid
  ,       coursecategory
  ,       courseid
  ,       grade
  FROM       got_category_cnt
  WHERE       category_cnt     > 0
  ORDER BY  studentid
  ,       coursecategory
  ,       courseid
  ,       grade
  ;
  

  The results I get are:

  `STUDENTID COURSECATEGORY   COURSEID G
  ---------- -------------- ---------- -
         111              2       2201 A
         111              2       2202 M
         111              2       2203 B
         111              2       2204 M
         111              4       2120 M
         111              4       2121 A
         111              4       2122 M
         111              4       2123 M
         222              3       1110 M
         222              3       1111 A
         222              3       1112 A
         222              3       1113 M
  

  These are a little different from what you have posted.

  user9207291 wrote:
  ...
  StudentID, courseid, coursecategory, grade
  111,2,2201, "A".
  111,2,2201, '
  111,2,2203, 'B '.
  111,2,2201, '
  111,4,2120, '
  111,4,2121, "A".
  111,4,2122, '
  111,4,2123, '
  222,3,1110, '
  222,3,1111, "A".
  222,3,1112, "A".
  222,3,1113, '

  Is it because there are errors in what you posted? Do you really want to have three lines of output for the same studentid (111) and courseid (2201)?

  Looks like there should be a slick way to do this with a partitioned outer join.
  A partitioned outer join matches each row of table A with the lines in table B, but, if there is no correspondence between a given A line and a line to a given partition of B, that the rank of A and the B partition is included in the game anyway. Here, we want the two partition tables.

• You can force BI outer join in w/o logical SQL Anwser in the Advanced tab

  In the repository of BI, we Protocol, region, Site of Protocol, Dimension of the object and a table of facts containing the physical tables Protocol, region, site and subject as a source in MDB.
  
  The relationship is that a protocol can have 0, n regions, region 0, n sites and a site 0, n subject, with the fact table of the region contains Protocol WID and done Site contains the Protocol and Feds region, subject is contains the Protocol, the region and Site Feds.
  
  In reporting on the Protocol and the region, BI users would like to have all the protocols appear in the report, without worrying if a protocol contains regions or not. Even as the Protocol-region-Site reports, etc. I can create these reports with SQL logic in the Advanced tab to force BI joins external Protocol, region or Site, etc.. However, users of BI would be difficult to write these SQLs.
  
  It seems to me that BI server always uses an inner join in the SQLs generated when the user selects the attributes of Protocol, region and Sites, etc. I played with constant implicit approach.
  
  I know that I can use outer join to define the logic Table Source in the layer of MDB. Would this tangle Protocol, region and site, tables etc. in MDB? What is the best practice to the MDB template when you want compatible outer join?
  
  Is there a way to force the BI server to generate outer join in the answers?
  
  Thanks in advance,
  
  Shi-ning

  Joins using external in the layer MDB is not recommended because it brings in the tables in the query and also impact performance. The right way to handle this scenario is the ETL.

  Add a record to the dimension tables with 0 for the ROW_WID and NULL values for the rest of the columns
  All of fact records without valid WID values for the dimension of research must have a value of 0 instead
  Now when we try to do an INTERNAL JOIN between the fact and any dimension, even those made documents that donot have a valid WID value will also be part of the result. It is essentially an OUTER JOIN operation.