Hierarchical EJB-QL queries

Hello.
I'm wondering if anyone has experience writing hirarchicar queries in EJB-QL and could help with something.
In my project there is an entity, Users, which posesses a many-to-one relation to itself, via the parent property. So, every user has two properties, Users parent, and Set<Users> children.
I would like to write an EJB-QL query that, given a Users, fetches all of it's descendants (children, grandchildren, and so on).
It would work somewhat like the following Oracle SQL query:
SELECT *
FROM USERS
CONNECT BY PRIOR USERS_ID = PARENT_ID
START WITH USERS_ID = <something>
Thanks in advance for any help
Hugo Oliveira

There is a discussion about deploying date and timestamp parameters to Application Server 7, see the following thread.
http://swforum.sun.com/jive/thread.jspa?threadID=15228&tstart=15

Similar Messages

How to create hierarchies in Bex queries?

Hi All,
I am new to query designer tool,
can you please help me how to create hierarchies in Bex queries using query designer tool?
Thanks in advance...

Hi,
directly you can't create hierarchy at bex level.
First you have to maintain hierarchy data source at source or bw side, based on your needs.
Later you can load data to hierarchy, after that at bex level you can activate hierarchy options to related info object properties.
To maintain hierarchy we need to enable hierarchy options at related info object(RSD1, hierarchy tab, select with hierarchies) maintenance level.
For example: info object - 0GL_ACCOUNT and you need hierarchy structure on gl account.
For that we need to have hierarchy data source and will load data from ecc.
Later we can develop reports on gl account in hierarchy fashion.
On what object infoo bject you want create hierarchy report and have loaded hierarchy data?
Thanks

Hierarchies in SPM-Queries

In the SPM Queries there are Hierarchies enabled for certain characteristics.
So far we have not maintained these in the backend. Now in certain SPM reports or dashboards we get errors due to that hierarchies not being available.
Therefore I would like to understand if these hierarchies are kind of mandatory for SPM, or if I can just delete them in the queries if they would never be created anyways?
Furthermore I would like to understand in general how SPM deals with hierarchies. Is there any documentation available on that topic?
Thanks again in advance,
Tobias

Hi Tobias,
Yes, you are right for certain characteristics in SPM Queries have hierarchies. And these dimensions are part of certain out of the box reports. So removing these hierarchies from the query will not fix the out of the box reports, as these reports are expecting the hierarchies to be present. But when users are creating new reports they will not receive that error anymore.
The only difference the way BW hierarchies are displayed is only one level is displayed in SPM UI at any given time.
Regards,
Rohit

EJB General Queries

Hi,
I have a few queries related to EJB. Please Help
1. Does every Bean Class have a seperate method pool of its own or all EJBs share a common method pool?.
2. Can a EJBObject serve multiple clients simultaneously. Also, Is there a single EJBObject for each bean or there can be multiple EJBObjects for a single bean.

http://forum.java.sun.com/thread.jspa?threadID=721058&tstart=0

Time Dependent Hierarchies in Bex Queries.

We are using Time depenedent Hierarchy structures.
year and period will be entered by the user in the query
Is there any simple solution where in based on the User entered year and period, corresponding hierarchy should be used in the query execution.

Hi Shashi,
One of the ways:
Create a variable type customer exit - in properties of the IO - in the 'select hierarchy' window there is a 'Variables' checkbox for key date.
Calculate the last date of the entered period/year in the exit.
Best regards,
Eugene

Using hierarchies from SAP BW BEx Query in BO Universe

Hello,
I'm trying to use a hierarchy (e.g. a simple customer class hierarchy) which is a part of a BEx Query in a BO Universe. But until now I have not found a way to get a hierarchy from a BEx Query working in a BO Unvierse.
1. Is it generally possible to use hierarchies in BEx Queries with a BO universe? (at the moment I can see all characteristics and keyfigures from the query... but I can't see the hierarchy levels for the characteristic)
2. If 1) is possible: I want to use a hierachy variable to select a hierarchy depnding on the user who is using a report... that's not a problem in BEx. But: Is it possible to use this query/scenario with a BO universe? Does the BO universe update itself automatically if the hierarchy has changed in a BEx query?
I've tried it with SAP BW 3.5 and SAP BI 7.0. BO Release is XI 3.0. Unfortunatelly without result.
Regards,
Thorsten

Hi Thorsten,
1) Hierarchies are mapped in the same way as custom-made universe hierarchies. When a hierarchy is defined on a characteristic in the BW Query, Designer creates one hierarchical structure in the universe, with a subclass for each level in the hierarchy. The structure depends on the current BW Query definition.
2) If the number of levels in a hierarchy changes in the BW query, you must update the universe. In the universe connection select *Use Single Sign On when refreshing
reports at view time* to allow the user to benefit from SAP SSO.
Hope this helps
Jacques

EJB-QL and Date

Hi!
I read in several books, datatype Date is not supported in EJB-QL.
I tried what's happening when using datatype java.util.Date in CMP Entity-Beans implementing a person with birthday and I could use queries like "findByBirthday", "findWhereBirthdayBefore", ...
So what exactly does "not supported" for Datatype "Date" mean? What are the not supported queries or am I automatically using application server dependant stuff then?
Thanks for your answer!
Martin

I use java.sql.Date and java.sqlTimestamp in my EJB-QL queries
This is example for java.sql.Timestamp
         <query>
            <query-method>
                <method-name>findByPeriod</method-name>
                <method-params>
                    <method-param>java.sql.Timestamp</method-param>
                    <method-param>java.sql.Timestamp</method-param>
                </method-params>
            </query-method>
            <ejb-ql><![CDATA[
                SELECT OBJECT(a)
                FROM action a
                WHERE a.moment > ?1 and a.moment < ?2
                ]]>
            </ejb-ql>
         </query>

Which of the following are true about abstract methods in EJB 2.0

Hi guys I'm beginner to EJB and i got some unanswered questions.
Can any one of you please.. give answers?
Thanks if you do...
Which of the following are true about abstract methods in EJB 2.0
CMP?
Choose all correct answers:
1. Abstract accessor methods should not be exposed in the EJB
component's interface
2.Abstract accessor/mutator methods are used to access and modify
persistent state and relationship information for entity objects
3.Abstract Accessor/Mutator methods do not throw exceptions
4.The EJB developer must implement the Accessor/Mutator methods
5.Abstract accessor methods may or may not be exposed in the EJB
component's interface
2.Which ONE of the following is true?
Choose the best answer:
1.Local interfaces cannot have a relationship with other Entity
components
2.Local interfaces cannot be used for Stateless Session EJB
3.Local interfaces can be a part of Object's persistent state
4.Local interfaces have the same functionality as that of a
stateless Session EJB
3.Which of the following describe the <cmr-field> in a EJB 2.0
descriptor?
Choose all correct answers:
1.A Local interface/Entity can be a value of a <cmr-field>
2.There is no <cmr-field> in EJB 2.0 descriptor
3.It is used to represent one meaningful association between any
pair of Entity EJBs, based on the business logic of the Application
4.It provides a particular mapping from an object model to a
relational database schema
5.It allows the Local Entity interfaces to participate in
relationships
4.Which of the following are the advantages of using Local interfaces
instead of dependent value classes?
Choose all correct answers:
1.Local Entity Interfaces can participate in Relationships
2.The life cycle of Local Entity Interfaces is managed by EJB
container, intelligently
3.Local Entity Interfaces can be used in EJB QL Queries
4.Local Entity Interfaces can be a part of the <cmp-field> but not
<cmr-field>
5.Which of the following are true about Local interfaces
1.A local interface must be located in the same JVM to which the EJB
component is deployed
2.Local calls involve pass-by-reference.
3.The objects that are passed as parameters in local interface
method calls must be serializable.
4.In general, the references that are passed across the local
interface cannot be used outside of the immediate call chain and must
never be stored as part of the state of another enterprise bean.
6.Which of the following specifies the correct way for a client
to access a Message driven Bean?
Choose the best answer:
1. via a Remote interface
2. via Home interface
3. Message driven bean can be accessed directly by the client
4. both 1 & 2
5. none of the above
------------------------------------------------------------------------7.Which of the following statements are true about message-driven
bean Clients?
------------------------------------------------------------------------Choose all correct answers:
They can create Queue and QueueConnectionFactory objects
They can create Topic and TopicConnectionFactory objects
They can lookup the JNDI server and obtain the references for
Queue and Topic and their connection Factories
Only 1 and 2 above

Hi guys I'm beginner to EJB and i got some unanswered
questions.
Can any one of you please.. give answers?
Thanks if you do...
Which of the following are true about abstract methods
in EJB 2.0
CMP?
Choose all correct answers:
1. Abstract accessor methods should not be exposed
d in the EJB
component's interfacefalse
2.Abstract accessor/mutator methods are used to
access and modify
persistent state and relationship information for
entity objectstrue
>
3.Abstract Accessor/Mutator methods do not throw
exceptionstrue
>
4.The EJB developer must implement the
Accessor/Mutator methodsfalse
5.Abstract accessor methods may or may not be exposed
in the EJB
component's interfacetrue
2.Which ONE of the following is true?
Choose the best answer:
1.Local interfaces cannot have a relationship with
other Entity
componentsfalse
2.Local interfaces cannot be used for Stateless
Session EJBfalse
3.Local interfaces can be a part of Object's
persistent statefalse
4.Local interfaces have the same functionality as
that of a
stateless Session EJBtrue
3.Which of the following describe the <cmr-field> in a
EJB 2.0
descriptor?
Choose all correct answers:
1.A Local interface/Entity can be a value of a
<cmr-field>true
2.There is no <cmr-field> in EJB 2.0 descriptorfalse
3.It is used to represent one meaningful association
between any
pair of Entity EJBs, based on the business logic of
the Applicationtrue
4.It provides a particular mapping from an object
model to a
relational database schematrue
5.It allows the Local Entity interfaces to
participate in
relationshipstrue
4.Which of the following are the advantages of using
Local interfaces
instead of dependent value classes?
Choose all correct answers:
1.Local Entity Interfaces can participate in
Relationshipsis
2.The life cycle of Local Entity Interfaces is
managed by EJB
container, intelligentlyis
3.Local Entity Interfaces can be used in EJB QL
Queriesnot
4.Local Entity Interfaces can be a part of the
<cmp-field> but not
<cmr-field>not
>
>
5.Which of the following are true about Local
interfaces
1.A local interface must be located in the same JVM
M to which the EJB
component is deployedtrue
2.Local calls involve pass-by-reference.true
3.The objects that are passed as parameters in local
l interface
method calls must be serializable.false
4.In general, the references that are passed across
s the local
interface cannot be used outside of the immediate
e call chain and must
never be stored as part of the state of another
r enterprise bean.true
>
6.Which of the following specifies the correct way for
a client
to access a Message driven Bean?
Choose the best answer:
1. via a Remote interfacefalse
2. via Home interfacefalse
3. Message driven bean can be accessed directly by
the clientfalse
4. both 1 & 2false
5. none of the abovetrue.
>
----------------7.Which of the following statements
are true about message-driven
bean Clients?
----------------Choose all correct answers:
They can create Queue and QueueConnectionFactory
objectsthe container can, dunno bout clients
>
They can create Topic and TopicConnectionFactory
objectsthe container can, dunno bout clients
>
They can lookup the JNDI server and obtain the
references for
Queue and Topic and their connection Factories
true
Only 1 and 2 abovefalse
somebody correct me if i'm wrong

Putting sql code in a ejb cmp method

hi i was wondering if there are any performance implications with creating a method in a cmp bean that performs an sql query as opposed to putting the query in the ejb deployment file. the reason i did this is that the ejb is quering a different table other than the one its mapped to. e.g. my cmp ejb is mapped to a table called Users (hence the bean is UserBean), and its querying a table called Address . What would be the best practice? Should I create an Address bean? The method in the UserBean e.g. getAddressForUser(String user_id) will be called quite frequently. What is the recommended best practice in this situation?
Thanks for your input.
Reed

hi i was wondering if there are any performance
implications with creating a method in a cmp bean
that performs an sql query as opposed to putting the
query in the ejb deployment file.You mean deployment descriptor ? You cannot write any SQL there.
Only EJB-QL.
the reason i did
this is that the ejb is quering a different table
other than the one its mapped to. e.g. my cmp ejb is
mapped to a table called Users (hence the bean is
UserBean), and its querying a table called Address .
What would be the best practice? Should I create an
n Address bean? The method in the UserBean e.g.
getAddressForUser(String user_id) will be called
quite frequently. What is the recommended best
practice in this situation?Create Address bean and use EJB-QL to get address for a user.
If you use CMP beans then you should have a bean for each table.
You can even use some nice tool (Middlegen) to generate entity beans.
But do use SQL for dynamic , complex database searches and bulk updates.
best regards,
Maris Orbidans

Universal Entity EJB find/select method for single db table

Hi there.
I'm quite new in bea spheres and my question reflects my small knowledge:
Is there any way to perform "universal" sql query on single database table via find/select methods? This means, that I would like to pass whole WHERE clause to these methods, not just some fields values.
What do I mean?
Normally we define find/select method that looks like this one:
SELECT OBJECT(o) FROM Cities AS o WHERE population>=?1 AND state=?2
We call this a method this way:
finder1 (1000000, 'France')
Method returns collection with objects holding data about cities in France with more than 1000000 inhabitans..
But, I want to define method, that could be called this way:
finder1 ("o.population>=1000000 AND o.state='France'")
This method should be able to perform any query on 'Cities' database table.
I tried with defining find method for following query:
SELECT OBJECT(o) FROM Cities AS o WHERE ?1
but it does not work, of course.
Is there possibility to define method that could be called this way?
I use Bea Weblogic Workshop 8.1 SP3
Message was edited by akla at Oct 23, 2004 5:21 PM

Hi,
The 'Dynamic Query' feature of WebLogic 8.1's EJB container can help you here. This feature enables you to execute complete EJB QL queries that are constructed on the fly. Check the WebLogic 8.1 documentation for more details.
-thorick

EJB QL and hints

Does OC4J ejb ql support oracle database hints? How can I optimize ejb ql queries?

Hi,
Why not just creating another 2 additional columns in Java Dictionary table in long format, while still maintain the timestamp for display? Ur search will base on long columns.

Dynamic Query

Dynamic Query
When I run the following dynamic query in the stored procedure and try to build the report then I get the following error Rep:1401 Fatal PL/SQL error occured, Ora-00936 Missing expression. When I am generating the report standalone using the following query , everything is working fine but in Stored procedure , it's giving me the above error, pls help.
create or replace package REP_DELIVERY_LOC_STATISTICS is
TYPE REP_DELIVERY_LOC_STAT_rec IS RECORD(
p_coordinating_facility_id sis_education_pgm_session.coordinating_facility_id%TYPE,
p_PSE_MUNICIPALITY_ID pse_education_pgm_session.PSE_MUNICIPALITY_ID%TYPE,
p_pse_education_pgm_session_id sis_education_pgm_session.pse_education_pgm_session_id%TYPE,
p_course_session_id course_session.course_session_id%TYPE);
TYPE REP_DELIVERY_LOC_STAT_cur IS REF CURSOR RETURN REP_DELIVERY_LOC_STAT_rec;
Type REP_DELIVERY_LOC_STAT_cur_Dyn is REF CURSOR;
FUNCTION REP_DELIVERY_LOC_STATISTICS(p_college_id IN pse_education_pgm_session.delivery_institution_id%TYPE,
p_start_date IN pse_education_pgm_session.session_strt_dte%TYPE,
p_end_date IN pse_education_pgm_session.session_strt_dte%TYPE,
p_coord_facility_join IN varchar2) return REP_DELIVERY_LOC_STAT_cur_Dyn;
end REP_DELIVERY_LOC_STATISTICS;
create or replace package body REP_DELIVERY_LOC_STATISTICS is
FUNCTION REP_DELIVERY_LOC_STATISTICS(p_college_id IN pse_education_pgm_session.delivery_institution_id%TYPE,
p_start_date IN pse_education_pgm_session.session_strt_dte%TYPE,
p_end_date IN pse_education_pgm_session.session_strt_dte%TYPE,
p_coord_facility_join varchar2) return REP_DELIVERY_LOC_STAT_cur_Dyn IS
REP_DELIV_LOC_STAT_CUR_Dyn_Q REP_DELIVERY_LOC_STAT_cur_Dyn;
BEGIN
OPEN REP_DELIV_LOC_STAT_CUR_Dyn_Q FOR
'SELECT distinct
seps.coordinating_facility_id,
peps.PSE_MUNICIPALITY_ID,
seps.pse_education_pgm_session_id,
cs.course_session_id
FROM pse_education_pgm_session peps,
sis_education_pgm_session seps,
course_session cs,
facility fac,
desc_sis_reporting_area depa,
desc_sis_reporting_type dept,
pse_education_program pep
WHERE peps.delivery_institution_id (+) = ' || p_college_id
|| ' and peps.session_strt_dte between ' ||p_start_date||' and ' || p_end_date
|| ' and seps.coordinating_facility_id = fac.facility_id (+)
and seps.pse_education_pgm_session_id (+) = peps.pse_education_pgm_session_id
and seps.pse_education_pgm_session_id = cs.pse_education_pgm_session_id (+)
and pep.pse_education_program_id = peps.pse_education_program_id
and dept.desc_sis_reporting_area_id = depa.desc_sis_reporting_area_id
and pep.desc_sis_reporting_type_id = dept.desc_sis_reporting_type_id
and dept.desc_sis_reporting_area_id = depa.desc_sis_reporting_area_id
and fac.desc_facility_id in (2,3)' || NVL (p_coord_facility_join, '1=1');
return REP_DELIV_LOC_STAT_CUR_Dyn_Q;
END;
end REP_DELIVERY_LOC_STATISTICS;

I would sugest to grab a book like
Inside Microsoft SQL Server 2008 T-SQL Querying (chapter 12 - Graphs, Trees, Hierarchies and Recursive Queries) or
Joe Celko's Trees and Hierarchies in SQL for Smarties to get familiar with how model hierarchies using SQL.
What is the meaning of the asterisks in the path?
Here is a dirty example using the Adjacency List model.
SET NOCOUNT ON;
USE tempdb;
GO
CREATE TABLE dbo.T1 (
id int NOT NULL UNIQUE CLUSTERED,
name varchar(50),
rootid int NULL
INSERT INTO dbo.T1 (id,name,rootid)
SELECT 1, 'animals', NULL
UNION ALL
SELECT 2, 'mamal', 1
UNION ALL
SELECT 3, 'cephalopod mollusc', 1
UNION ALL
SELECT 4, 'salt water fish', 1
UNION ALL
SELECT 5, 'fresh water fish', 1
UNION ALL
SELECT 6,'whale', 2
UNION ALL
SELECT 7,'pirana', 5
UNION
SELECT 8,'shark', 4
UNION
SELECT 9, 'octopus', 3;
GO
WITH Tree AS (
SELECT
id, name, rootid,
CAST('/' + [name] + '/' AS varchar(900)) AS mat_path
FROM
dbo.T1
WHERE
rootid IS NULL
UNION ALL
SELECT
C.id, C.name, C.rootid, CAST(P.mat_path + C.name + '/' AS varchar(900))
FROM
Tree AS P
INNER JOIN
dbo.T1 AS C
ON P.id = C.rootid
SELECT
FROM
Tree
ORDER BY
mat_path;
GO
DROP TABLE dbo.T1;
GO
Excuse my poor knowledge of zoology or animal biology.
AMB
Some guidelines for posting questions...
AYÚDANOS A AYUDARTE, guía básica de consejos para formular preguntas

Hibernate OR EclipseLink...Which is best with Weblogic and Oracle DB?

Hi All,
In my company, we are using Oracle DB and Weblogic application server. So in the process to upgrade or switch to new ORM, we shortlisted two options - Hibernate and EclipseLink.
I gathered following summary regarding both ORMs -
Hibernate:
1.     When you need to train people, like we are going to do next week – most of the companies have Hibernate experts.
2.     When you hire new developers, most of them come with specific Hibernate experience.
3.     When you need to consult with experts, both in the internet or consultants, you have LOTS of options. Endless forums and communities all regarding Hibernate.
4.     Hibernate is an open source which has a huge community. This means that it will be improved all the time and will push the ORM market forward.
5.     Hibernate is an open source which means you have the code to handle, and in case needed, fit it to your needs.
6.     There are lots of plugins to Hibernate, such as validations tool, audit tools, etc. These becomes standard as well and dismiss you from impl. things yourself.
7.     One most important thing with ORM tool, is to configure it according to your application’s needs. Usually the default setting doesn’t fit to your needs.
For that sake, when the market has a huge experience with the tool’s configuration, and lots of experts (see point 1 and 3) – most of chances you will find similar cases and
lots of knowledge about how to configure the tool and thus – your application.
EclipseLink:
1. Fully supported by Oracle. Hibernate no. In case of pb, it could be cumbersome to prove that it is a pure Weblogic one. Concretely, we will have to prove it (waste of time and complexity).
2. Eclipse link is developed by Oracle and the preferred ORM in the Weblogic /Oracle DB world.
3. Even if at a certain time EclipseLink was a bit late compared to Hibernate (feature), EclipseLink evolved very fast and we can consider now that they close the gap.
4. No additional fee as soon as you have Weblogic license. You will need to pays additional fee if you want some professional support on Hibernate.
5. We are currently relying on Hibernate for our legacy offer and are facing pb in second level cache (JGroups). Today, we are riding off this part!. Consequences are limitation in clustering approach (perf)
6. On EclipseLink side we do succeed to manage first and second level cache in a clustering approach.
7. Indeed Hibernate is open source, so you can imagine handling it. In reality, the code is so complex that it is nearly impossible to modify it. Moreover as it is LGPL, you need to feedback all the modified sources to the community systematically.
8. All tests performed by Oracle concerning Weblogic are using EclipseLink. Moreover, Oracle says that some specific optimizations are done to manage Oracle DB.
9. Hibernate comes from JBoss community.
Right now we are preferring Hibernate but there are concerns/reasons like EclipseLink developed by Oracle and preferred ORM in Webogic/ Oracle DB world (compatibility of ORM with DB and App. server), support comparison with both ORM, which are preventing to finalize the decision.
Please help me with you views and opinions and share you experience with us so that we can make a perfect decision.
If you want you can also reply to me @ [email protected].
Thanks.

The way the ORMs are designed, integration with application servers are relatively simple, and all provides the same features. Also since WebLogic have been around for a while, all ORMs are all well tested in this configuration.
Hibernate has lot more users, and is likely very often used with Oracle DB, so you can expect not much bug against Oracle DB, maybe even less bug than EclipseLink, which is not much used. EclipseLink does provide support for some esoteric Oracle DB features like hierarchical and flashback queries.
OpenJPA and DataNucleus are also JPA compliant. It’s likely that Open JPA has a higher user base than EclipseLink, so less unknown bugs.
Oracle paying support is well known to be a bad joke. It’s a negative return to use this channel, even if they would be free. So in reality, you end up to use the open (free) forum to get support.
What’s was lacking with Hibernate before is Dynamic Fetch Planning, but they now have some support, see http://opensource.atlassian.com/projects/hibernate/browse/HHH-3414. OpenJPA was the first to implement this must have.
EclipseLink has query in memory, which can be used, but the API do not help to leverage it, and EclipseLink’s leadership made it clear that they are not going to make it better, instead they want to push Coherence cache.
Hibernate has an open API for second level cache, which mean you can get out of problem by using another implementation, for example, EHCache seems to be professionally tested, so I would be surprise you find obvious bugs.
I cannot comment on Hibernate source code quality, but I can tell you that locking mechanism in EclipseLink is used to be very fragile, and many concepts are dispersed over the code base.
The runtime monitoring of Hibernate have always been great due to the fact that JBoss have always been strong on JMX, EclipseLink has not much usable features on this.
If I would be you, I would consider OpenJPA or Hibernate instead of EclipseLink, the main reason is that because EclipseLink has a so low user base, I have found lot of obvious bugs in production, like if I was the only user of it. Then, when I submitted bugs to the small development team, which do not encourage user base contribution, they were too busy trying to keep up adding the JPA interfaces on top of their existing proprietary APIs.

Columns for XML gallery

Hi! I'm nearly there but I can't get this scipt to work!! Its
a basic gallery I'm building that imports images via xml with
thumnails. Problem is I started trying to make columns and it wont
work! I'm nearly there though I can smell it! The Column bit is
declaired at the top and in a for statement at the bottom! Can you
take a look?
Thanks Gavin

I could generate the mixed case columns with no problem. It was my mistake and sorry for the inconvinience.
However I am running into problems modelling a nested hierarchical set of queries with levels more than 2.
Please advise of any sample code available anywhere.
For example I could do :
create table x1 ( id number , f1 varchar2(10));
creat table x2 ( id number , id_x1 number references x1(id) , f1 varchar2(10)) ;
create table x3 ( id number , id_x2 number references x2(id), f1 varchar2(10)) ;
To model this, I did
create type x3_typ as object ( id number, id_x2 number , f1 varchar2(10)) ;
create type x3_typ_t is table of x3_typ ;
create type x2_typ as object ( id number, id_x1 number , f1 varchar2(10), x3_list x3_typ_t ) ;
create type x2_typ_t is table of ref x2_typ ;
create type x1_typ as object ( id number
, f1 varchar2(10) , x2_list x2_typ_t ) ;
create or replace view x3_x2 as
select id , f1 , cast(multiset(select * from x3 ) as x3_typ_t ) as x3_list
If I try to use a view again like as given below, I get the Oracle inconsistent datatypes error.
create or replace view x2_x1 as
select id , f1 , cast(multiset(select * from x3_x2 ) as x2_typ_t ) as x2_list
I there a better way? Am I missing something? Please help.

Exception condition "HIERARCHY_NOT_FOUND"

Hello,
we have scenario in Xcelsius where we used QAAWS with different queries.
after upgrading to the EhP1 (BW System) we strated getting WIS errors in QAAWS tool wrt 0country hierarchy even though there are no hierarchies activated in queries.
I rebuilt all the connections but the error is not eliminated - when I try to create a new universe the error is triggered:
Error at quering the hierarchies for dimension 0COUNTRY (Cube xxx/xxxxxx):Exception condition "HIERARCHY_NOT_FOUND" raised.
tried to repair the InfoObjects as well, but it didn't work
any other suggestions?
thanks

Hi Ingo,
we don't want to use the hierarchy in the universe (thus it's not activated in the BW Query)
- i can run BW Query in BEX Web normally without any errors
- when i perform MDXTEST in BW system the exception HIERARCHY_NOT_FOUND is trigerred:
Auslösestelle des Laufzeitfehlers
    Programm                                SAPLRRH1
    Include                                 LRRH1U03
    Zeile                                   69
    Modultyp                                (FUNCTION)
    Modulname                               RRH1_HIERARCHY_IN_TO_EX
I already repaired hierarchies and tables of the 0COUNTRY InfoObject in RSRV , but it didn't help.
Hierarchy is not found in rshiedir table, even though it's there in A version - how can I repair this issue?

Hierarchical EJB-QL queries

Similar Messages

Maybe you are looking for