Trying to UNION two views with nested tables

I am using Oracle Database 10g Enterprise Edition Release 10.2.0.4.0 - Prod, and my objective is to generate some XML. What I have been doing in the past, is to create a view, and then pass that view to the DBMS_XMLQuery routine. This gives me a CLOB that I pass along to the application. A typical object would be something like:
create or replace type XML_UGroup_Member_Type as object (
     username          varchar2(128),
     group_key          varchar2(128));
create or replace type XML_UGroup_Member_List
    as table of XML_UGroup_Member_Type;
show errors
Create or replace type XML_UGroup_Type as object (
     Group_Space     varchar2(32),
     group_key     varchar2(128),
     group_name     varchar2(255),
     members          XML_UGroup_Member_List);
/This particular application will be doing stuff with group information. Pretty simply types - a group with a name and a few other keys, and a list of members. A sample view would be like:
create or replace view xml_department_ugroup_view of xml_ugroup_type
  with object identifier ( group_key ) as
  Select 'Department',
      'Department:' || orgn_code,
      'Department_' || orgn_name
      cast ( multiset (
           select  Username, person_id,
                    'Department:' ||  effective_orgn
            from directory_master dm, logins l
           where effective_orgn = dd.orgn_code
             and dm.person_id = l.owner)
            as XML_UGroup_Member_List )
         as members
   from directory_departments dd
  where dept_include = 'Y';
can select from this view, and I can pass it to the DBMS_XMLQUERY package and get a nice XML document. What I want to is essentially put several of these views together, like
Select * from XML_Department_Ugroup_View
Union
Select * from XML_Portfolio_Ugroup_View;But UNION does not work with nested tables:
ERROR at line 1: ORA-00932: inconsistent datatypes: expected - got SIMON.XML_UGROUP_MEMBER_LIST
What I was hoping to do, was to develop a set of sub views for each of the group "spaces", and then generate a view/object that includes all of the sub views, and be able to operate with that single object.
Thoughts on how to make this work, or on alternate approaches? I have considered calling XMLQuery on each of the sub views and concatenating the XML documents and returning that to the application, but I was hoping to find a more "unified" approach.

I modified my approach a bit - I changed the base views to return an XMLType object, like:
create or replace view xml_portfolio_ugroup_xml
as
Select XMLElement("group",
         xml_ugroup_type(
      'Portfolio',
         'Portfolio:' || coas_code || ':' || orgn_Code,
      cast ( multiset (
           select  Username,
            from directory_master dm, logins l
           where ( effective_orgn, effective_coas ) in
           where effective_orgn = dd.orgn_code
              and dm.person_id = l.owner
            as XML_UGroup_Member_List ))) as grp
   from directory_departments dd
  where dept_include = 'Y';and then I combined them with something like the following (I expect to adjust this as I learn more about the XML DB support)
create or replace view xml_ugroup_xml as
  select xmlconcat(
     (select xmlagg(grp) from xml_department_ugroup_xml),
     (select xmlagg(grp) from xml_portfolio_ugroup_xml)
     ) as GROUPS
     from dualand as I define more group branches, I can add them into the XMLConCat stanzas. I am still open to suggestions as to ways of doing this better or cleaner, but this at least gets the project moving forward again.

Similar Messages

  • Object view  with nested table and member functions????????????????

    HI frds:)
    I need some help regarding writeing soem queries..
    I have to use view in retreving data and by useing member functions of object.
    1) i have to create a nested table by useing type object.
    2) i have to create a object wtih member functions inorder to create view of taht nested table,.
    3) by useing this objectt view and by useing methods i have to write soem queries ...
    If any one know or any information regarding this please reply me... as i have searched in net alot but i was unable to figure out..
    Thanks....

    // first createing object
    create type emp_det as object
    (empname varcahr2(20),start_date date,end_date date);
    // creating table of that object
    create table emp_detai_table as table of emp_det;
    // creating nested table
    create table empl (emp_no number,emp_detail emp_detail_table,dep_no number)nested table emp_detail store as s;
    //now i want to create view.. inorder to create view i need to create object. in that object i want to create methods.. these methods should work with date attributes..
    after creating object view i need to select or write some queries by useing member methods..
    This is the tast i have to perform.,. i have no materials ...if u have any link ..forward me..
    i have to do it as soon as possible..
    waiting for your reply...

  • Ability to query on all columns from a view with multiple tables

    I have view with 4 tables. view has about 50 columns.
    hypothetically, I want to build a form to give the user an ability to query on any of those 50 columns. user can search on one or more fields.
    what is the best way to write the query to retrieve the results without performance impact.
    please let me know if the question is not clear.

    If you want to permit them to query any of 10 fields with no restrictions, you could create 10 separate single-column indexes. The optimizer would have to try to pick the best of these indexes if you specified multiple search criteria. But since you have a view which presumably means that you've got multiple other indexes involved for the underlying join conditions, you'd probably want/need to combine the column lists of the indexes you're adding for searches with the indexes you've created for joins, which requires looking at the text of the view and figuring out the best way to optimize the 10 single-column searches. Of course, it is possible that the optimizer could chose to do something like a b-tree to bitmap conversion on multiple indexes which could allow it to use multiple independent indexes for your queries, but that's generally not ideal performance-wise.
    All this is a fancy way of saying that there probably isn't a cut and dried answer to your question. You need to figure out the queries you're trying to optimize (say, the 10 single-condition queries) and you need to figure out what set of indexes on the various underlying tables provides the best performance weighing those benefits against the cost on other side of maintenance during DML. Which is going to depend on the text of the view, the underlying tables, the existing indexes, etc.
    Justin

  • Import tables with nested table : ORA-00600

    In Oracle 9.2
    Create object, type as table, and table with nested table (store as syms_ntab) are successfully.
    Also its export.
    In process of import on another server (also 9.2, 'fromuser=one touser=two') shows errors:
    . . importing table "SYMS_NTAB"
    IMP-00058: ORACLE error 600 encountered
    ORA-00600: internal error code, arguments: [kokeeafi1], [2], [2], [], [], [], [], []
    IMP-00075: Warning: The nested table may contain partial rows or duplicate rows
    But for all that table is created and error occur on phase inserting strings.
    What is this?
    In Oracle 8.0.5 i perform similar operation without error.

    From Oracle error messages and codes manual:
    ORA-00600 internal error code, arguments: [string], [string], [string], [string], [string], [string], [string], [string]
    Cause: This is the generic internal error number for Oracle program exceptions. It indicates that a process has encountered a low-level, unexpected condition. Causes of this message include:
    * timeouts
    * file corruption
    * failed data checks in memory
    * hardware, memory, or I/O errors
    * incorrectly restored files
    The first argument is the internal message number. Other arguments are various numbers, names, and character strings. The numbers may change meanings between different versions of Oracle.
    Action: Report this error to Oracle Support Services after gathering the following information:
    * events that led up to the error
    * the operations that were attempted that led to the error
    * the conditions of the operating system and databases at the time of the error
    * any unusual circumstances that occurred before receiving the ORA-00600 message
    * contents of any trace files generated by the error
    * the relevant portions of the Alter files
    Note: The cause of this message may manifest itself as different errors at different times. Be aware of the history of errors that occurred before this internal error.

  • PL/SQL add procedure with nested table - Duplicate Thread

    Hi,
    I am trying to do a procedure to input information for one order and another for 2 orders.
    The information I have so far is as follows:
    Drop table Orders cascade constraints;
    Drop type item_type;
    Drop type Item_nested;
    Create or Replace Type item_type AS Object (
    Cat_code Varchar2(6),
    Amount_ord Number(3),
    Cost Number(5,2) );
    Create or Replace Type item_nested as table of item_type;
    Create Table Orders (
    Order_no Varchar2(8) constraint pkorder primary key,
    Customer_name Varchar2(30),
    AddressLine1 Varchar2(20),
    AddressLine2 Varchar2(20),
    AddressLine3 Varchar2(20),
    Town Varchar2(20),
    Postcode Varchar2(10),
    Country Varchar2(20),
    Order_items item_nested,
    Order_date Date)
    Nested Table Order_items
    Store as nested_items return as locator;
    This has so far worked but I have not managed the insert procedure.
    I am using Oracle SQL*plus
    Thanks
    SG
    Edited by: user10689875 on 11-Jan-2009 03:39

    Duplicate thread ->
    PL/SQL add procedure with nested table
    Please remove it & marked it as duplicate.
    Regards.
    Satyaki De.

  • Problem when expanding Tree - Tree with nested table column

    Hi, i have created the tree using the Tree with nested table column.
    I have created a node called TREE_ROOT in the context.
    This node has few attributes which includes children_loaded, is_leaf, is_expanded.
    I have created the recursive node TREE_SUB for the above node TREE_ROOT.
    In the view, i have created the table with the master column. The above attributes have been mapped accordingly. I have created the action handler for load_children.
    In this action handler method, i receive the context_element correctly. In this method, i determine the children of the selected element and the resulting children are attached to this context_element.
    But the problem is: when i add elements to context_elements in the method load_children, these
    elements get added to the node TREE_ROOT as well.
    Please help.
    thanks and best regards,
    Pramod

    I just use some types defined in this user... Well, I hope you know what is the type definition of d_period_sec,
    don't you ? I didn't ask to provide all types existed now, only types you are
    using.
    Anyhow you have been granted with execute privilege for types you are using:
    SQL> conn tau_tll/tau_tll;
    Connected.
    SQL> create or replace type d_period_sec as object (date# date);
      2  /
    Type created.
    SQL> grant execute on d_period_sec to public with grant option;
    Grant succeeded.
    SQL> conn scott/tiger
    Connected.
    SQL> CREATE OR REPLACE TYPE unit_function AS OBJECT (
      2  xi NUMBER,
      3  yi NUMBER,
      4  xe NUMBER,
      5  ye NUMBER,
      6  xm NUMBER,
      7  ym NUMBER,
      8  v NUMBER,
      9  a NUMBER,
    10  f NUMBER,
    11  descr VARCHAR2 (20)
    12  );
    13  /
    Type created.
    SQL> grant execute on unit_function to master;
    Grant succeeded.
    SQL> CREATE OR REPLACE TYPE unit_moving_point AS OBJECT
      2  (
      3  p tau_tll.d_period_sec, -- from user TAU_TLL
      4 
      5  m unit_function
      6  )
      7  /
    Type created.
    SQL> grant execute on unit_moving_point to master;
    Grant succeeded.
    SQL> CREATE OR REPLACE TYPE moving_point_tab AS TABLE OF unit_moving_point;
      2  /
    Type created.
    SQL> grant execute on moving_point_tab to master;
    Grant succeeded.
    SQL> CREATE OR REPLACE TYPE moving_point AS OBJECT (u_tab moving_point_tab);
      2  /
    Type created.
    SQL> grant execute on moving_point to master;
    Grant succeeded.
    SQL> conn master/master
    Connected.
    SQL> CREATE TABLE MPOINTS (
      2  id NUMBER,
      3  mpoint scott.Moving_Point)
      4  NESTED TABLE mpoint.u_tab store as moving_tab;
    Table created.Rgds.

  • Partition exchange error on table with nested table

    On Oracle 11.2.0.1, I have a partitioned table with some partitions that need to be 'archived' (in terms of moving partitions to an 'archive' table).
    I have a source table like:
    CREATE TABLE IS_PODACI245
      ID_OBJEKTA_IDENTIFIKACIJA  NUMBER(10),
      ID_OBJEKTA                 NUMBER(20),
      DATUM                      TIMESTAMP(6)       NOT NULL,
      TZ                         NUMBER(3),
      DATA1                      NUMBER(10),
      DATA2                      NUMBER(6),
      DATA3                      NUMBER(10),
      DATA4                      NUMBER,
      DATA5                      T_NTCIP_CLIMATE_TABLE
    NESTED TABLE DATA5 STORE AS IS_PODACI245_STORE_TABLE
    TABLESPACE DATA
    PARTITION BY RANGE (DATUM)
      PARTITION P_201107 VALUES LESS THAN (TIMESTAMP' 2011-08-01 00:00:00')
        LOGGING
        NOCOMPRESS
        TABLESPACE DATA, 
      PARTITION P_MAXVALUE VALUES LESS THAN (MAXVALUE)
        LOGGING
        NOCOMPRESS
        TABLESPACE DATA
    NOCOMPRESS
    NOCACHE
    NOPARALLEL
    MONITORING;
    CREATE INDEX IDX_IS_PODACI245_KOMPLEKS ON IS_PODACI245
    (ID_OBJEKTA_IDENTIFIKACIJA, ID_OBJEKTA, DATUM)
      TABLESPACE DATA
    LOCAL ( 
      PARTITION P_201107
        LOGGING
        NOCOMPRESS
        TABLESPACE DATA, 
      PARTITION P_MAXVALUE
        LOGGING
        NOCOMPRESS
        TABLESPACE DATA
    NOPARALLEL;
    CREATE OR REPLACE TYPE t_ntcip_climate_table as table of t_ntcip_climate_fmt;
    CREATE OR REPLACE TYPE t_ntcip_climate_FMT as object
    (  dev_index number(6)
    ,   dev_description varchar2(512)
    ,   dev_type number(10)
    ,   dev_status number(10)
    ,   dev_mfr_status varchar2(512)
    ,   dev_active number(3)
    ,   dev_test_activation number(10)
    /I would like to make exchange partition using stage table, and everything is going fine on all tables, but only on a few of them (listed source is one of them, and they're only tables with nested tables wihin), where I get an error.. but sometimes ;)
    on a statement like:
    ALTER TABLE IS_PODACI245_ARH EXCHANGE PARTITION P_201106  WITH TABLE IS_PODACI245_STAGE EXCLUDING INDEXES  WITHOUT VALIDATION;I got an error:
    ORA-00001: unique constraint (TXV.SYS_C0032911) violated
    it's an unique index between parent and nested table.
    what could cause that problem?

    Dear,
    I suppose that the unique constraint
    ORA-00001: unique constraint (TXV.SYS_C0032911) violatedis the one you 've created on the nested table IS_PODACI245_STORE_TABLE
    If so, why not disable that constraint and try again.
    I have never exchanged such a kind of partitioned table having a nested table in it. But, I could imagine that the cloned non partitioned table IS_PODACI245_STAGE should at least be the exact image of the partitioned table IS_PODACI245_ARH (of course without the partition part) but with the nested table part and including all indexes
    In addition, if you have a parent/child relationship between your partitioned tables, then there is a chronological order of exchange starting by the child and then finishing by the parent
    see the following link for more information about this order of exchange (and comment 2 for an example also)
    http://jonathanlewis.wordpress.com/2006/12/10/drop-parent-partition/#more-65
    Hope this helps
    Mohamed Houri

  • ANSI Standard Join with Nested Table

    Does anyone know how to (or whether you actually can) use ansi standard table joins with nested tables.
    Non-ansi standard would look something like this
    SELECT e.empno
    FROM departments d, TABLE(d.employees) e
    WHERE d.deptno = 10;
    Where d.employees is a nested table.
    But if I try ansi-standard I like such:
    SELECT e.empno
    FROM departments d
    JOIN TABLE(d.employees) e
    WHERE d.deptno = 10;
    I get
    ORA-00905: missing keyword
    because I have nothing to join it on.
    Your help is very much appreciated

    Both replies worked fine.
    I think I will go with the NATURAL JOIN as it seems the cleanest option.
    Thanks Guru 2748

  • Fill datagridview with Nested Table Object Type ???

    Hello everybody, please you help me resolve my problem?
    I follow this example: *(A Sample Application using Object-Relational features)* http://download-west.oracle.com/docs/cd/B14117_01/appdev.101/b10799/adobjxmp.htm
    And this tutorial: *(Using Oracle User-Defined Types with .NET and Visual Studio)* http://www.oracle.com/technology/obe/hol08/dotnet/udt/udt_otn.htm
    Now I have 3 Object Table: Stock, Customer and Purchase Order. With the tutorial it's OK to show the data of Stock and Customer Table [there is no nested table in], but I can't do this with table Purchase Order, the tutorial don't show how to work with Nested Table type [Missing or something ?]
    When I try to display the data of table Purchase Order, I get the error:
    typeName='LINEITEMLIST_NTABTYP'' is not specified or is invalid*
    Follow the tutorial, I generate custom class for this UDT and get another error:
    this.STOCK_REF = ((object)(Oracle.DataAccess.Types.OracleUdt.GetValue(con, pUdt, "STOCK_REF")));*
    Argument Exception Unhandle: Object attribute is not mapped to a custom type member.*
    Can You show me how to do this ? Show the data of the Nested Table in Visual Studio, do I have something wrong ??? Thanks and Best Regards.

    I think you need to go through the tutorial more carefully. I think you are missing steps.
    You are using a REF. Are you creating an object table for the objects to be stored in? Are you dereferencing the REF? This is covered by the Oracle by Example tutorial.
    http://www.oracle.com/technology/obe/hol08/dotnet/udt/udt_otn.htm#t11
    When you receive the nested table, it will be inside a .NET class. You can create this class by using the Create Custom Class menu item on the nested table type in server explorer.
    This class will contain a ToString method and a ToXML method.
    The question is, where are you attempting to display the data from this nested table? In a grid?
    Does it make sense to display a nested table inside one cell of a grid? No. You would need to construct a special UI that will show the
    nested table. Maybe the user clicks on the cell on the grid, which opens a new grid that displays the data. Or maybe there is a second grid on the form that always shows the nested table for the row that the first grid currently has selected.
    I realize I am not providing you the code... but it sounds like you need to design a more sohisticated UI than what the tutorial is using -- one that can handle displaying one additional table per row.

  • How to create table view with reference table

    Hi experts,
    How to create table view with reference table in SE11, plz gve me stp by stp procedure.
    pints grnded for hlp.

    Hi
    Go to Tcode se11 choose view and enter the name and create a popup opens up choose database view option
    enter the description
    On the left hand side choose the table name.
    Click on view fields tab and choose your table fields.Here you can choose which fields you want in your view.
    Save and then activate.
    Hope this helps.
    Regards,
    Harish

  • Join two tables to one recordset with nested table?

    Hello all!
    I want to "de-normalize" two tables into one for presentation reasons. For example:
    CREATE TABLE foo(id number)
    CREATE TABLE bar(foo_id number, value varchar2(4))
    And with some data:
    FOO:
    1
    2
    BAR:
    1, 'gaz'
    1, 'boz'
    2, 'blah'
    Now I want to create a view that holds the value of the BAR table in a nested table, if possible and efficient enough..
    VIEW:
    1, nested_table('gaz,'boz')
    2, nested_table('blah')
    Any clue for creating such a view? I would prefer a view, because this view will be joined with other tables later.
    Thanks in advice!

    Or
    SQL> with foo as
      2  ( select 1 id from dual union all
      3    select 2 id from dual
      4  ),
      5  bar as
      6  (
      7    select 1 foo_id, 'gaz'  value from dual union all
      8    select 1 foo_id, 'boz'  value from dual union all
      9    select 2 foo_id, 'blah' value from dual
    10  )
    11  --
    12  --
    13  select id,
    14         cast(multiset((select value from bar where id=foo_id)) as sys.dbms_debug_vc2coll) value
    15    from foo
    16  /
            ID VALUE
             1 DBMS_DEBUG_VC2COLL('gaz', 'boz')
             2 DBMS_DEBUG_VC2COLL('blah')or from 10g on
    SQL> with foo as
      2  ( select 1 id from dual union all
      3    select 2 id from dual
      4  ),
      5  bar as
      6  (
      7    select 1 foo_id, 'gaz'  value from dual union all
      8    select 1 foo_id, 'boz'  value from dual union all
      9    select 2 foo_id, 'blah' value from dual
    10  )
    11  --
    12  --
    13  select id,
    14         cast(collect(value) as sys.dbms_debug_vc2coll) value
    15    from foo, bar
    16    where id=foo_id
    17    group by id
    18  /
            ID VALUE
             1 DBMS_DEBUG_VC2COLL('gaz', 'boz')
             2 DBMS_DEBUG_VC2COLL('blah')Edited by: michaels2 on Oct 8, 2008 2:27 PM

  • Help needed with nested tables

    I have two identical relations R and S defined as:
    CREATE TYPE table_typ AS TABLE OF VARCHAR2(8);
    CREATE TABLE R(
    a INTEGER,
    b table_type)
    NESTED TABLE b STORE as b_1;
    CREATE TABLE S(
    a INTEGER,
    b table_type)
    NESTED TABLE b STORE as b_2;
    Both have two tuples each:
    INSERT INTO r VALUES (1, table_type('a','b'));
    INSERT INTO r VALUES (2, table_type('d'));
    INSERT INTO s VALUES (1, table_type('b','c'));
    INSERT INTO s VALUES (3, table_type('e'));
    Would it be possible to write a query that "unions" R and S so that nested tables in tuples (in R and S) that agree on the A attribute are merged together, while tuples that do not agree on A still appear in the result? That is, the result I am looking for is:
    A B
    1 TABLE_TYPE('a','b','c')
    2 TABLE_TYPE('d')
    3 TABLE_TYPE('e')
    I've tried a simple union, but it does not work. Any help on this would be greatly appreciated.
    Thanks,
    Laura

    I don't have 10g, but the following is a 9i solution. I know there has to be a better way. About all I can say for the code below is that it works. I am only posting it because it is better than nothing, until somebody with 10g posts something better. I have included the stragg function by Tom Kyte.
    scott@ORA92> -- test data:
    scott@ORA92> CREATE OR REPLACE TYPE table_type AS TABLE OF VARCHAR2 (8);
      2  /
    Type created.
    scott@ORA92> CREATE TABLE r(
      2    a INTEGER,
      3    b table_type)
      4    NESTED TABLE b STORE as b_1
      5  /
    Table created.
    scott@ORA92> CREATE TABLE s(
      2    a INTEGER,
      3    b table_type)
      4    NESTED TABLE b STORE as b_2
      5  /
    Table created.
    scott@ORA92> INSERT INTO r VALUES (1, table_type('a','b'));
    1 row created.
    scott@ORA92> INSERT INTO r VALUES (2, table_type('d'));
    1 row created.
    scott@ORA92> INSERT INTO s VALUES (1, table_type('b','c'));
    1 row created.
    scott@ORA92> INSERT INTO s VALUES (3, table_type('e'));
    1 row created.
    scott@ORA92> commit
      2  /
    Commit complete.
    scott@ORA92> -- start of code from Tom Kyte:
    scott@ORA92> create or replace type string_agg_type as object
      2  (
      3       total varchar2(4000),
      4 
      5       static function
      6            ODCIAggregateInitialize(sctx IN OUT string_agg_type )
      7            return number,
      8 
      9       member function
    10            ODCIAggregateIterate(self IN OUT string_agg_type ,
    11                        value IN varchar2 )
    12            return number,
    13 
    14       member function
    15            ODCIAggregateTerminate(self IN string_agg_type,
    16                          returnValue OUT  varchar2,
    17                          flags IN number)
    18            return number,
    19 
    20       member function
    21            ODCIAggregateMerge(self IN OUT string_agg_type,
    22                      ctx2 IN string_agg_type)
    23            return number
    24  );
    25  /
    Type created.
    scott@ORA92> create or replace type body string_agg_type
      2  is
      3 
      4  static function ODCIAggregateInitialize(sctx IN OUT string_agg_type)
      5  return number
      6  is
      7  begin
      8        sctx := string_agg_type( null );
      9        return ODCIConst.Success;
    10  end;
    11 
    12  member function ODCIAggregateIterate(self IN OUT string_agg_type,
    13                             value IN varchar2 )
    14  return number
    15  is
    16  begin
    17        self.total := self.total
    18        || ','
    19        || value;
    20        return ODCIConst.Success;
    21  end;
    22 
    23  member function ODCIAggregateTerminate(self IN string_agg_type,
    24                               returnValue OUT varchar2,
    25                               flags IN number)
    26  return number
    27  is
    28  begin
    29        returnValue := ltrim(self.total,',');
    30        return ODCIConst.Success;
    31  end;
    32 
    33  member function ODCIAggregateMerge(self IN OUT string_agg_type,
    34                           ctx2 IN string_agg_type)
    35  return number
    36  is
    37  begin
    38        self.total := self.total || ctx2.total;
    39        return ODCIConst.Success;
    40  end;
    41 
    42 
    43  end;
    44  /
    Type body created.
    scott@ORA92> CREATE or replace
      2  FUNCTION stragg(input varchar2 )
      3  RETURN varchar2
      4  PARALLEL_ENABLE AGGREGATE USING string_agg_type;
      5  /
    Function created.
    scott@ORA92> -- end of code from Tom Kyte
    scott@ORA92> -- usage of above function:
    scott@ORA92> COLUMN c FORMAT A15
    scott@ORA92> SELECT a, stragg (column_value) AS c
      2  FROM   (SELECT t1.a, t1.column_value
      3            FROM   (SELECT r.a, c.column_value
      4                 FROM   r, TABLE (r.b) c) t1,
      5                (SELECT s.a, d.column_value
      6                 FROM   s, TABLE (s.b) d) t2
      7            WHERE  t1.a = t2.a
      8            UNION
      9            SELECT t1.a, t2.column_value
    10            FROM   (SELECT r.a, c.column_value
    11                 FROM   r, TABLE (r.b) c) t1,
    12                (SELECT s.a, d.column_value
    13                 FROM   s, TABLE (s.b) d) t2
    14            WHERE  t1.a = t2.a)
    15  GROUP  BY a
    16  UNION ALL
    17  SELECT a, stragg (column_value) AS c
    18  FROM   (SELECT t1.a, t1.column_value
    19            FROM   (SELECT r.a, c.column_value
    20                 FROM   r, TABLE (r.b) c) t1,
    21                (SELECT s.a, d.column_value
    22                 FROM   s, TABLE (s.b) d) t2
    23            WHERE  t1.a = t2.a (+)
    24            AND    t2.a IS NULL
    25            UNION
    26            SELECT t2.a, t2.column_value
    27            FROM   (SELECT r.a, c.column_value
    28                 FROM   r, TABLE (r.b) c) t1,
    29                (SELECT s.a, d.column_value
    30                 FROM   s, TABLE (s.b) d) t2
    31            WHERE  t1.a (+) = t2.a
    32            AND    t1.a IS NULL)
    33  GROUP  BY a
    34  /
             A C
             1 a,b,c
             2 d
             3 e

  • Data pump import error with nested tables

    So the problem is somewhat long :)
    Actually the problems are two - why and how oracle are treating OO concept and why data pump doesn't work?
    So scenario for the 1st one:
    1) there is object type h1 and table of h1
    2) there is unrelated object type row_text and table of row_text
    3) there is object type h2 under h1 with attribute as table of row_text
    4) there is table tab1 with column b with data type as table of h1. Of course column b is stored as nested table.
    So how do you think - how many nested tables Oracle will create? The correct answer is 2. One explicitly defined and one hidden with system
    generated name for the type h2 which is under type h1. So the question is WHY? Why if I create an instance of supertype Oracle tries to adapt
    it for the subtype as well? Even more if I create another subtype h3 under h1 another hidden nested table appears.
    This was the first part.
    The second part is - if I do schema export and try to import it in another schema I got error saying that oracle failed to create storage table for
    nested table column b. So the second question is - if Oracle has created such a mess with hidden nested tables how to import/export to another
    schema?
    Ok and here is test case to demonstrate problems above:
    -- creating type h1 and table of it
    SQL> create or replace type h1 as object (a number)
      2  not final;
      3  /
    Type created.
    SQL> create or replace type tbl_h1 as table of h1;
      2  /
    Type created.
    -- creating type row_text and table of it
    SQL> create or replace type row_text as object (
      2    txt varchar2(100))
      3  not final;
      4  /
    Type created.
    SQL> create or replace type tbl_row_text as table of row_text;
      2  /
    Type created.
    -- creating type h2 as subtype of h1
    SQL> create or replace type h2 under h1 (some_texts tbl_row_text);
      2  /
    Type created.
    SQL> create table tab1 (a number, b tbl_h1)
      2  nested table b
      3  store as tab1_nested;
    Table created.
    -- so we have 2 nested tables now
    SQL> select table_name, parent_table_name, parent_table_column
      2  from user_nested_tables;
    TABLE_NAME                     PARENT_TABLE_NAME
    PARENT_TABLE_COLUMN
    SYSNTfsl/+pzu3+jgQAB/AQB27g==  TAB1_NESTED
    TREAT(SYS_NC_ROWINFO$ AS "GINTS"."H2")."SOME_TEXTS"
    TAB1_NESTED                    TAB1
    B
    -- another subtype of t1
    SQL> create or replace type h3 under h1 (some_texts tbl_row_text);
      2  /
    Type created.
    -- plus another nested table
    SQL> select table_name, parent_table_name, parent_table_column
      2  from user_nested_tables;
    TABLE_NAME                     PARENT_TABLE_NAME
    PARENT_TABLE_COLUMN
    SYSNTfsl/+pzu3+jgQAB/AQB27g==  TAB1_NESTED
    TREAT(SYS_NC_ROWINFO$ AS "GINTS"."H2")."SOME_TEXTS"
    SYSNTfsl/+pz03+jgQAB/AQB27g==  TAB1_NESTED
    TREAT(SYS_NC_ROWINFO$ AS "GINTS"."H3")."SOME_TEXTS"
    TAB1_NESTED                    TAB1
    B
    SQL> desc "SYSNTfsl/+pzu3+jgQAB/AQB27g=="
    Name                                      Null?    Type
    TXT                                                VARCHAR2(100)OK let it be and now I'm trying to export and import in another schema:
    [oracle@xxx]$ expdp gints/xxx@xxx directory=xxx dumpfile=gints.dmp logfile=gints.log
    Export: Release 11.2.0.1.0 - Production on Thu Feb 4 22:32:48 2010
    <irrelevant rows skipped>
    Processing object type SCHEMA_EXPORT/TABLE/STATISTICS/TABLE_STATISTICS
    . . exported "GINTS"."TAB1"                                  0 KB       0 rows
    . . exported "GINTS"."SYSNTfsl/+pz03+jgQAB/AQB27g=="         0 KB       0 rows
    . . exported "GINTS"."TAB1_NESTED"                           0 KB       0 rows
    . . exported "GINTS"."SYSNTfsl/+pzu3+jgQAB/AQB27g=="         0 KB       0 rows
    Master table "GINTS"."SYS_EXPORT_SCHEMA_01" successfully loaded/unloaded
    ******************************************************************************And now import. In order to create types transformation of OIDs is applied and also remap_schema
    Although it fails to create the table.
    [oracle@xxx]$ impdp gints1/xxx@xxx directory=xxx dumpfile=gints.dmp logfile=gints_imp.log remap_schema=gints:gints1 transform=OID:n
    Import: Release 11.2.0.1.0 - Production on Thu Feb 4 22:41:48 2010
    Copyright (c) 1982, 2009, Oracle and/or its affiliates.  All rights reserved.
    Connected to: Oracle Database 11g Release 11.2.0.1.0 - Production
    Master table "GINTS1"."SYS_IMPORT_FULL_01" successfully loaded/unloaded
    Starting "GINTS1"."SYS_IMPORT_FULL_01":  gints1/********@xxx directory=xxx dumpfile=gints.dmp logfile=gints_imp.log remap_schema=gints:gints1 transform=OID:n
    Processing object type SCHEMA_EXPORT/USER
    ORA-31684: Object type USER:"GINTS1" already exists
    Processing object type SCHEMA_EXPORT/SYSTEM_GRANT
    Processing object type SCHEMA_EXPORT/ROLE_GRANT
    Processing object type SCHEMA_EXPORT/DEFAULT_ROLE
    Processing object type SCHEMA_EXPORT/PRE_SCHEMA/PROCACT_SCHEMA
    Processing object type SCHEMA_EXPORT/TYPE/TYPE_SPEC
    Processing object type SCHEMA_EXPORT/TABLE/TABLE
    ORA-39083: Object type TABLE:"GINTS1"."TAB1" failed to create with error:
    ORA-02320: failure in creating storage table for nested table column B
    ORA-00904: : invalid identifier
    Failing sql is:
    CREATE TABLE "GINTS1"."TAB1" ("A" NUMBER, "B" "GINTS1"."TBL_H1" ) SEGMENT CREATION IMMEDIATE PCTFREE 10 PCTUSED 40 INITRANS 1 MAXTRANS 255 NOCOMPRESS LOGGING STORAGE(INITIAL 65536 NEXT 1048576 MINEXTENTS 1 MAXEXTENTS 2147483645 PCTINCREASE 0 FREELISTS 1 FREELIST GROUPS 1 BUFFER_POOL DEFAULT FLASH_CACHE DEFAULT CELL_
    Processing object type SCHEMA_EXPORT/TABLE/TABLE_DATA
    Processing object type SCHEMA_EXPORT/TABLE/INDEX/INDEX
    Processing object type SCHEMA_EXPORT/TABLE/INDEX/STATISTICS/INDEX_STATISTICS
    ORA-39083: Object type INDEX_STATISTICS failed to create with error:
    ORA-01403: no data found
    ORA-01403: no data found
    Failing sql is:
    DECLARE I_N VARCHAR2(60);   I_O VARCHAR2(60);   c DBMS_METADATA.T_VAR_COLL;   df varchar2(21) := 'YYYY-MM-DD:HH24:MI:SS'; BEGIN  DELETE FROM "SYS"."IMPDP_STATS";   c(1) :=   DBMS_METADATA.GET_STAT_COLNAME('GINTS1','TAB1_NESTED',NULL,'TREAT(SYS_NC_ROWINFO$ AS "GINTS"."H2")."SOME_TEXTS"',1);  DBMS_METADATA.GET_STAT_INDNAME('GINTS1','TAB1_NESTED',c,1,i_o,i_n);   INSERT INTO "
    ORA-39083: Object type INDEX_STATISTICS failed to create with error:
    ORA-01403: no data found
    ORA-01403: no data found
    Failing sql is:
    DECLARE I_N VARCHAR2(60);   I_O VARCHAR2(60);   c DBMS_METADATA.T_VAR_COLL;   df varchar2(21) := 'YYYY-MM-DD:HH24:MI:SS'; BEGIN  DELETE FROM "SYS"."IMPDP_STATS";   c(1) :=   DBMS_METADATA.GET_STAT_COLNAME('GINTS1','TAB1_NESTED',NULL,'TREAT(SYS_NC_ROWINFO$ AS "GINTS"."H3")."SOME_TEXTS"',1);  DBMS_METADATA.GET_STAT_INDNAME('GINTS1','TAB1_NESTED',c,1,i_o,i_n);   INSERT INTO "
    Processing object type SCHEMA_EXPORT/TABLE/STATISTICS/TABLE_STATISTICS
    Job "GINTS1"."SYS_IMPORT_FULL_01" completed with 4 error(s) at 22:41:52So any idea how to make export/import of such tables?
    TIA
    Gints

    Tom Kyte has said it repeatedly ... I will repeat it here for you.
    The fact that Oracle allows you to build object tables is not an indication that you should.
    Store your data relationally and build object_views on top of them.
    http://www.morganslibrary.org/reference/object_views.html
    If you model properly, and store properly, you don' have any issues.

  • Urgent Help Please, Error with Nested Table

    Hello All,
    I'm trying to create data mining models based on some medical data. I have combined two tables (Patients) and (tumors) into one table (cancer_patients). In the (cancer_patients) table there is one column (tumors) as nested table column. So I got the following error (ORA-22913) about the nested table. I need an advice about how to fix this error in ODM please
    This is the code of the tables
    CREATE TABLE "DSS2_MINING"."PATIENTS"
       (     "PATIENT_ID" VARCHAR2(10 BYTE) NOT NULL ENABLE,
         "REGISTRY_ID" NUMBER(10,0),
         "RACE" VARCHAR2(2 BYTE),
         "***" VARCHAR2(1 BYTE),
         "BIRTHDATE_YEAR" NUMBER(4,0),
         "NUMBER_OF_PRIMARIES" NUMBER(1,0),
         "VITAL_STATUS_RECORD" VARCHAR2(1 BYTE),
         "CAUSE_OF_DEATH" VARCHAR2(5 BYTE),
         "SURVIVAL_TIME" VARCHAR2(4 BYTE),
         "SURVIVAL_TIME_FINAL" NUMBER,
         "SURVIVAL_VARIABLE" VARCHAR2(1 BYTE),
          CONSTRAINT "PATIENTS_PK" PRIMARY KEY ("PATIENT_ID");
    CREATE TABLE "DSS2_MINING"."TUMORS"
       (     "TUMOR_ID" NUMBER NOT NULL ENABLE,
         "PATIENT_ID" VARCHAR2(10 BYTE),   -- FK
         "SEER_RECORD_NUMBER" NUMBER,       -- This column contain a sequance number of the records for each patients
         "MARITAL_STATUS" VARCHAR2(1 BYTE),
         "AGE" NUMBER,
         "DATE_OF_DIAGNOSIS" DATE,
         "HISTOLOGY_GROUP" VARCHAR2(2 BYTE),
         "BEHAVIOR" VARCHAR2(1 BYTE),
         "GRADE" VARCHAR2(1 BYTE),
         "DERIVED_AJCC_STAGE_GROUP" VARCHAR2(2 BYTE),
         "STAGE_OF_CANCER" VARCHAR2(2 BYTE),
         "RADIATION" VARCHAR2(1 BYTE),
         "CS_SCHEMA" VARCHAR2(2 BYTE),
         "FIRST_PRIMARY_IND" VARCHAR2(1 BYTE),
         "TUMOR_SIZE" NUMBER(4,1),
         "TUMOR_EXTENSION" VARCHAR2(2 BYTE),
         "LYMPH_NODES" VARCHAR2(1 BYTE),
         "NODES_POSITIVE" NUMBER,
         "ESTROGEN" VARCHAR2(3 BYTE),
         "PROGESTERONE" VARCHAR2(3 BYTE),
         "SURGERY" VARCHAR2(2 BYTE),
          CONSTRAINT "TUMORS_PK" PRIMARY KEY ("TUMOR_ID");
    create or replace type tumor_object AS
    object(
    tumor_id VARCHAR2(1),  
    marital_status VARCHAR2(1),  
    age NUMBER(3),  
    date_of_diagnosis DATE, 
    cs_schema VARCHAR2(2),  
    histology_group VARCHAR2(2),  
    behavior VARCHAR2(1),  
    grade VARCHAR2(1),  
    first_primary_ind VARCHAR2(1),  
    tumor_size NUMBER(4,   1),  
    tumor_extension VARCHAR2(2),  
    lymph_nodes VARCHAR2(1),  
    nodes_positive NUMBER(4),  
    surgery VARCHAR2(2),
    radiation VARCHAR2(1)
    create or replace type tumor_table as table of tumor_object;
      CREATE TABLE "DSS2_MINING"."CANCER_PATIENTS"
       (     "PATIENT_ID" VARCHAR2(10 BYTE) NOT NULL ENABLE,
         "RACE" VARCHAR2(2 BYTE),
         "***" VARCHAR2(1 BYTE),
         "NUMBER_OF_PRIMARIES" NUMBER(1,0),
         "TUMORS" "DSS2_MINING"."TUMOR_TABLE" ,
         "VITAL_STATUS_RECORD" VARCHAR2(1 BYTE),
         "CAUSE_OF_DEATH" VARCHAR2(5 BYTE),
         "SURVIVAL_TIME_FINAL" NUMBER,
         "SURVIVAL_VARIABLE" VARCHAR2(1 BYTE),
          CONSTRAINT "CANCER_PATIENTS_PK" PRIMARY KEY ("PATIENT_ID")
       NESTED TABLE "TUMORS" STORE AS "TUMORS_STOR_TABLE"Then, I have transferred the data using
    INSERT
      INTO CANCER_PATIENTS
      SELECT  PATIENT_ID,
              RACE,
              NUMBER_OF_PRIMARIES,
               SELECT  CAST(
                            COLLECT(
                                    TUMOR_OBJECT(
                                                 TUMOR_ID,
                                                 MARITAL_STATUS,
                                                 AGE,
                                                 DATE_OF_DIAGNOSIS,
                                                 CS_SCHEMA,
                                                 HISTOLOGY_GROUP,
                                                 BEHAVIOR,
                                                 GRADE,
                                                 FIRST_PRIMARY_IND,
                                                 TUMOR_SIZE,
                                                 TUMOR_EXTENSION,
                                                 LYMPH_NODES,
                                                 NODES_POSITIVE,
                                                 SURGERY ,
                                                 RADIATION
                            AS TUMOR_TABLE
                 FROM  "TUMORS" T
                 WHERE T.PATIENT_ID = P.PATIENT_ID
              VITAL_STATUS_RECORD,
              CAUSE_OF_DEATH,
              SURVIVAL_TIME_FINAL,
              SURVIVAL_VARIABLE
        FROM  PATIENTS PThanks
    A.L
    Edited by: user9003901 on Nov 30, 2010 12:23 AM

    Hi
    i was using AWM since 9i just because it creates OLAP metadata with a new API called CWM2, and Oracle Enterprise Manager was using a propietary API .
    With AWM 10.1.0.4 building your OLAP metadata is very simple, you can define your Dimensions, cubes, measures and even you can load your data from your source system.
    You can refer to the Oracle® OLAP Application Developer's Guide where you can find more information on this tool.
    And relative to your error i will recommend that you build on a dimension-by-dimension basis and try to validate your OLAP Metadata.
    Regards

  • "member of " not working with nested table of dates

    Can someone explain why this doesn't work. The last select does not return any row. When trying the same with a table of number, it works fine.
    drop table test;
    drop type date_tab;
    create type date_tab as table of date;
    create table test(dates date_tab) nested table dates store as dates;
    insert into test values (date_tab(to_date('10-jan-2007'),to_date('15-jan-2007'),to_date('15-jan-2007')));
    commit;
    select * from test where to_date('10-jan-2007') member of dates;
    Line above should find the row, but does not return anything.

    > With 10G Oracle said that these two engines are now sharing the same source code
    Is that documented somewhere? Regarding database versions, it was 9.0.1 that claimed an integrated parser. I don't see any update for 10g.
    > So in the old days one had to do a [SELECT sysdate INTO d FROM dual] to assign a function value to a PL/SQL var - as the function did not exist in PL/SQL.
    Then (from Oracle 7 onwards?) these functions were also available in PL/SQL. However, the two engines did not share common code. So functions in SQL did not always behave like function in PL/SQL and vice versa.
    I don't recall that limitation in PL/SQL v1 (Forms 3 to 4.5, and database v6, though I doubt many people actually used PL/SQL on the database because (1) it was separately licensed, and (2) it didn't have stored procedures) - but then it was a while ago so I could be mistaken.
    I know USER, SYSDATE and others used to be implicit queries of dual (i.e. the supplied PL/SQL function was just a wrapper containing SELECT SYSDATE INTO dt FROM dual; RETURN dt;) although that's probably just confusing the issue.

Maybe you are looking for