Migrate custom folders over to database views

Hi,
Over time one of our business areas has had a number of custom folders added.
Whilst the SQL is fine, I would like to put these into the database as views to reduce the length of SQL statements been parsed by Discoverer and hopefully reduce network traffic.
The problem is that a number of reports have been written based upon the existing custom folders.
Is there any way to refresh an existing custom folder so that it becomes a database folder, without having to reassign items in the existing reports.
Ambitious I know, but sooner or later I know I'm going to get asked the question.
Thanks.

Hi,
Is there any way to refresh an existing custom folder so that it becomes a database folder, without having to reassign items in the existing reports.No, there isn't. But what you could try doing is delete the custom folder and then create a base folder with exactly the same name, identifier, items and joins based on the new view. Hopefully, the workbook would then find the same items in the folder.
If you cann't get that to work, you could just change the custom folder to select * from your custom view.
Rod West

Similar Messages

Custom function in a database view makes performance slow in OBI?

Hi,
I am facing a major performance problem.
I have a oracle view which is calling a database function ( which I created).
CREATE OR REPLACE VIEW ISRM_ECOX_NAK_REPLAY (ret)
as
select APP.ISRM_ECOX_NAK_REPLAY (CASHFLOW_MESSAGE.EXTERNAL_DEAL_NUMBER) ret from CASHFLOW_MESSAGE
It runs in the database in a couple of seconds. But when I import the view in the OBI repository the query keeps running for hours without returning any data.
If there are very few records in the database table used on the view then it runs in OBI after a couple of minutes, but not otherwise.
When I pick the query from Manage Sessions and run it in the database, again it runs fast
The OBIEE version is 10.1.3.4.1
thanks and regards,
Gaurav
Edited by: Gaurav on 22-Sep-2011 02:43

Hi guarav,
may be an idea...why not creating a materialized view in the db which you make use of in obi?
Can't that solve your perf problems?
Kr,
A

Complex Folders Vs Database Views - Discoverer 10g

Hello All,
The way complex folders works is almost similar to creating a database view and using that to create a custom folder.
So could someone advise wrt peformance etc which is a better option ?
Creating complex folders gives the flexibility to change whatever changes are rqd using disco admin and does not reqd even logging to which incase of DB views would need to be done to recreate a view incase of any changes. But would like to weigh down both the options on other parameters.
One thing i see is implementing oracle apps security is easier using DB views...but again..i could create custom folders using select * from table where gl_security_pkg concept and then create complex folders using these folders ?
Any thoughts would be helpful
Thanks.
Message was edited by:
vishalg

Vishalg.
The way I've always considered it is as follows:
1. I prefer not to be 'married' to one tool (ie: Disco) and therefore always recommend writing database views over custom / complex folders if possible. That way, database tools such as TOAD, SQL Nav, SQL Dev, etc. can also use the view (not to mention Oracle Reports, etc. as well).
With that in mind, I try to have most folders just pointing to a database view and all the 'fine tuning' done in the folder (ie: column order, hierarchies, security, etc.).
2. I do use custom folders when it's a LOV that's used in an item class. For example, you could create a simple LOV in a folder on an item, by just using that item for the source and the output. However, if it's a big dataset or you want to go to another table that's small and just has the distinct values for example, a custom folder is perfect for creating (I always put a _LOV extension on the folder name so I know) and setting an item class against.
Sometimes though, I've created a LOV for the list and then have just brought it in to a folder by simply pointing to it. It's the same kind of consideration in 1. above - if I'm creating a system that may be used by other tools - or God forbid - a client wants to move to another query tool like Cognos, BO, etc. then all the views are already there in the database and you just have to build their version of the middle layer (ie: catalog, universe, etc.).
3. I don't think I've ever had to use a complex folder. The concept is good though as in you can have 2 folders that are joined, but you want to create another folder that only has certain fields from each of the original 2 folders and that's the one a user uses.
But same consideration as above. Do I want to completely customize the Disco EUL, or use it more as a 'delivery device' (hey ... good term!).
One other big consideration for all this is that if you work in a shop where they have a big (read obnoxious) DBA who doesn't "allow views in his database" (been there), then you can always do the custom / complex folder thing from within Disco Admin and keep the DBA happy. However, once he's fired for being obnoxious and then new "touchie-feely" DBA comes on board, you may curse why you had to create a custom EUL in the first place.
All just your point of view, but I'm a view guy all the way.
Russ

Migrating database views between instances

I've discoverer folders based out of database views?(around 30-40) and i'm planning to migrate all the VIEW's from one instance to another
I'm currently executing 1 by 1 and since, these views are growing in number with the requirements i was wondering if there a easy way/tool to achieve this.

Hi
In addition to the advice already given, are you also changing the owner for the views? If you are then you will need to update Discoverer to tell it who now owns them.
Should you move a view from one schema to another, use this workflow to update Discoverer.
1. Launch Discoverer Admin.
2. Open the business area containing a view that had its owner changed.
3. Right-click on the folder and from the pop-up select Properties.
4. Click on Owner and from the list of users, select the new owner
5. Click OK to return to the Folder Properties dialog box.
6. Click OK to close the Folder Properties dialog box.
7. Right-click on the folder again; and from the pop-up list select Refresh.
8. As you only have one option, click Finish.
9. Discoverer will now compare the data definitions for the object and hopefully you will have no differences.
If you do have differences this means there is something different in the new view and you will have to determine what that is and then fix it.
Does this help?
Regards
Michael

How to view messages in custom folders in PC Suite...

Hi,
I've got a 7710 cell phone, and have organized my messages in some custom folders, but when I connect my cell phone to my PC I can not access these custom folders, and only INBOX, SENT, DRAFTS and OUTBOX are seen.
Is there any way to see custom folders via PC Suite and phone browser?
Is it any plan for future versions of PC Suite to support this?

No. Is like that. The message operation of PCSuite is on my opinion the weakest part of it. Is very limited. Follow the new version on support pages, and wait for new releases, because is always improved from version to version.

Question - are we not able to create indexed items in custom folders?

Just a question here. I would like to use the index/Value option for parameters in a workbook. This is where the List of Values drop down box for the parameter would show both the index number and the value and let you choose to select based on the index or on the value. An obvious example of this would be with months, where you would have January (1), February (2), March (3), and so forth. To do this, you have to set up indexed items in Discoverer Administrator.
That is where I am running into a problem/issue. I am not being able to pick an item and click to look at its properties, scroll down to the Indexed Item property, click on that property, and get a list of items to pick from. On the indexed item property I see a value of <None>, and when I click on that property, no list of items pops up.
I have defined indexed items before, but that was always done on regular folders (imported from the database). In this case, I am working with a custom folder.
The SQL that I have keyed in for this custom folder is this -
SELECT flex_value, description, flex_value_set_id, flex_value_id
FROM fnd_flex_values_vl
WHERE flex_value_set_id = 1011187
ORDER BY flex_value;
FND_FLEX_VALUES_VL is an Oracle view, that combines Oracle tables FND_FLEX_VALUES and FND_FLEX_VALUES_TL.
My goal would be to show both the FLEX_VALUE and the DESCRIPTION in the parameter to the users, so they can better decide what value(s) they want to pick. This is a general ledger application that I am working with, with the General Ledger account number segments and the Flexfield value set for each segment (we are using 7 segments to define our general ledger account number).
Am I not able to define indexed items when working with a custom folder? The documentation is not real clear. It talks about simple folders and complex folders, but does not mention custom folders, at least that I can find. Or perhaps I need to do something different with how I code the SQL for the custom folder to enable me to define indexed items in the custom folder?
Hope this makes some sense. Always a challenge to explain the situation in a clear manner. Thanks for any insight/assistance anyone can give me on this.
John Dickey

Ahh, those picky little details. Once I created an Item Class LOV on my item, then I could click on the Indexed Item property and get the list of fields in the folder to show up and pick my item (to be the indexed item) from that list. I went back to the user guides and the documentation sure does not indicate this requirement. For example, in the Discoverer Administrator guide, you have the section (page 9-30 in the version I have) titled "How to assign an indexed item to an item". Nowhere in that section does it say that you have to create an Item Class LOV first to be able to do this. Though it does show how if you do not have an LOV on the item you pick to be the indexed item, you will be prompted to create an LOV for that item. In the past I had my LOVs created first. This time I was trying to do the indexed item first before creating my LOVs. So lesson learned - you have to create an Item Class LOV first before you can assign an indexed item to an item.
Thanks for the assistance.
John Dickey

Database migration query not executing in database

We moved over our SQL database to another server. We have a few .cfm template files that have queries to get 'content' from our database. These queries in the .cfm template files work fine and execute properly. However, for some reason, since we moved over our database, the queries in the actual database don't execute/run anymore. In the old server we didn't have this problem. Our queries in the database would run properly then. The data would display correctly on the webpages.
Is there some kind of properties setup that was overlooked when migrating over our database?
The output we now see when viewing our pages is the actual query code.
Example:
SELECT lname
FROM table
This is actually displayed on the website and therefore the query is not executing/running. Our code was never changed. So why is it not outputting the same?
Thanks for any help provided.

Sorry for the confusion.
The coldfusion templates and database queries both worked fine on the previous old server.
Now since the migration onto the new server, only the coldfusion templates work fine, but not the queries in the database.
So for example, a query in the coldfusion template will run fine, and execute the intial pull of data. But once the content is pulled from the database, the queries inside the database (including stored procedures) don't execute.
We do get the expected results when we run our stored procedure in Management Studio and if we put the stored procedure on a standalone web page we do see the records.
We only see the sql code when we have the stored procedure inside of our w_content table of our database, and its being called by the first query in the template. For some odd reason the stored procedure does not execute if it is inside the database.
Hopefully that clarifies our problem.
Thank you for helping!

Progammatic update a table with a database view in the page

Hi All,
I am using JDev 11g. With FOD database schema, I have one database view Products which comes from two tables Products_Base and Product_Transactions. I created three EOs (ProductEO, ProductsBaseEO, ProductTransactionsEO) and three VOs (ProductVO, ProductsBaseVO, ProductTransactionsVO) based on their EOs respectively.
Here is my scenario. I have an ADF form which is based on the database view Products and is dragged and dropped from Data Controls->ProductVO. When an existing record is submitted, a backing bean method will be called to update the data against the table Products_Base (and the table Product_Transactions at the same time) programmatically. An update method updateProductPrice() is added into the Application Module and published it to UI Client. The submit button in the page is created by directly dragging and dropping Data Controls->updateProductPrice into the page. When I run it, I got the following error message,
Failed to post data to database during "Update": SQL Statement "UPDATE PRODUCTS ProductEO SET COST_PRICE=:1 WHERE PRODUCT_ID=:2".
What I don't understand here is that, in my update method updateProductPrice(), it supposes to update the table Products_Base. But from the error, it appears that it is trying to update the view Products. Can anyone give me a help on what I did wrong here? When I try to debug it, it throws an exception to this line in the method updateProductPrice(),
getDBTransaction().commit();
Here are my codes,
The method which got called in the backing bean
public String cb6_action() {
DCBindingContainer bc = (DCBindingContainer)getBindings();
FacesCtrlAttrsBinding ProductId = (FacesCtrlAttrsBinding)bc.get("ProductId");
FacesCtrlAttrsBinding CostPrice = (FacesCtrlAttrsBinding)bc.get("CostPrice");
JUCtrlActionBinding action =
(JUCtrlActionBinding)bc.findCtrlBinding("updateProductPrice");
DCDataControl dc = action.getDataControl();
ApplicationModule am = (ApplicationModule)dc.getDataProvider();
AppModule service = (AppModule)am;
service.updateProductPrice(new Long(ProductId.toString()), CostPrice.toString());
return null;
public BindingContainer getBindings() {
return BindingContext.getCurrent().getCurrentBindingsEntry();
The update method defined in the Application module (AppModuleImpl.java)
public void updateProductPrice(long productId, String costPrice) {
ProductsBaseEOImpl product = retrieveProductById(productId);
if (product != null) {
try {
product.setCostPrice(new Number(costPrice));
getDBTransaction().commit();
catch (JboException ex) {
getDBTransaction().rollback();
throw ex;
catch (SQLException ex1) {
getDBTransaction().rollback();
private ProductsBaseEOImpl retrieveProductById(long productId) {
EntityDefImpl productDef = ProductsBaseEOImpl.getDefinitionObject();
Key productKey = ProductsBaseEOImpl.createPrimaryKey(new DBSequence(productId));
return (ProductsBaseEOImpl)productDef.findByPrimaryKey(getDBTransaction(),productKey);
Edited by: john wang on Oct 27, 2009 7:14 AM

or
merge into test
using (select rowid rid
            , id
            , sub_id
            , startdate
            , lead (startdate) over (order by id, sub_id) - 1 ed
       from test) x
on (x.rid = test.rowid)
when matched then
   update set end_date = x.ed
;

HR transactions and functionality and creation of Database view

Hi Guys
    I am new to HR module . can anybody send me the list of transaction codes with HR functionalities.
I also have to made a database view from hrp1001 and hrp1010 table to list all chief positions who are head of business group , Business unit , Site etc .
Could you help me in creating this databse view.

Hi,
These are HR Infotypes
0000    Events
0001    Org assignment
0002    Personal info
0003    Payroll data
0007    Work time
0008    Basic pay
0014    Reoccurring pay
0015   1 X pay
0027   Cost Center
0041   Event Dates
0057   Membership dues
0165   Over ride to limits on deductions
0167   Health
0168   Insurance
0169   Savings
0170   Spending
0194   Garnishment reduction
0195   Garnishment order
0207   Residence Tax
0208   Work Tax
0209   Unemployment Tax
0210   Withholding
0216   Garnishment adjustment
0221   Adjustment
0267   Off cycle
2005   OT
2010   Catts direct to cluster
1000   Infotypes 1000 1999 are PD Relationship infotypes
Logical Database (PNP) ·
Programming with Infotypes ·
Processing Time infotypes/Cluster ·
Processing Payroll infotypes/Cluster
In OM level hrp 1001 and 1010 will be there.
If u want total transactions goto SE11->TSTC Table.
****POINTS TO BE CONSIDERD
Thanks
P.SRIKANTH

EUL Export errors - comments in Custom Folders

In the spirit of sharing.
I recently had some major problems releasing a Discoverer EUL through our database release process, using the EUL Export functionality (i.e. creating an .eex file).
I have a large complex Custom Folder in the EUL, comprised of 14 large unioned select statements.
The EUL would seem to import OK into the new database, but a Workbook based on the Custom Folder continually came up with an item dependancy error.
We traced the problem back to the Custom Folder, which was invalid in the Admin Tool, and therefore not visible to the EUL Users - with a workbook based on the offending Custom Folder.
The moral of the story is DO NOT put single line comments "--" in Discoverer Custom Folder SQL queries - especially large queries. The EUL Export process will unformat your queries, and the single line comments will end up inside the valid SQL - which will invalidate it.
Only use block comments "/* Comment */" in your Custom Folders.
Hope this helps someone else save some time.

Just a note to thank u for this post. This is one of the answers I was searching for today and now I see the light!

Users unable to create or delete folders in migrated public folders using Outlook

We have an Exchange 2013 CU3 environment migrated from Exchange 2007.
The public folder migration was completed over the weekend.
The environment has several public folder mailboxes.
Post migration users are unable to create/delete new sub folders, or modify permissions using Outlook on any of the migrated public folders. Users can however create new top level folders using outlook. They can also create and delete new posts
in migrated public folders.
Admins are able to create folders and set permissions on migrated folders using the EAC.
Test User accounts used for testing are set to use the Primary Hierarchy mailbox as their default public folder mailbox.
Test Users have been given Owner permissions from the root down on the folders we are testing with.
We have tested with Outlook 2010 and 2013 getting the same "Cannot create the folder" error.

Further testing
We moved a top level folder from a secondary PF mailbox to the primary PF mailbox using the New-PublicFolderMoveRequest command in powershell.
After the move completed we could create new folders under the moved top level folder.
The top level folder that was moved had it's own sub-folders that we did not move to the primary mailbox. (We didn't move the whole branch.. Just the top level folder)
We still cannot create or modify the existing sub-folders after moving the top level folder.
We then moved the newly created sub-folder to a secondary PF mailbox.
At that point we could no longer create sub-folders in the folder form outlook.
From what I can tell you can only create new sub-folders in folders homed to the primary PF mailbox when using outlook.
Is this a bug or as designed?
According to this Tech ed presentation Clients connecting to a secondary PF mailbox should have folder changes proxy to the primary PF mailbox.. (See slide 10)
http://channel9.msdn.com/Events/TechEd/NorthAmerica/2013/OUC-B329#fbid=
Update : 2-5-2014
Not much new to report other than it appears that users need to be homed to the primary PF mailbox to delete calendar meetings.
Opened a ticket with Premier support.
Update 2-11-2014
Premier support continues to look into the issue. No ideas as to the cause yet.
2-24-13
Still no resolution from support or even a clue what is causing this.

Query Rewrite with regular database Views

Hi all,
I'm trying to make my programmer's life easier by creating a database view for them to query the data, so they don't have to worry about joining tables. However, query rewrite doesn't work no matter how I define the MV and View. Here's an example:
I've Sales tables with columns: PDate, CustCode, Amount
and Customer table with columns: CustCode, CustDesc
I create a view SALES_V with columns: PDate, CustCode, CustDesc, Amount by joining Sales table with Customer table as follows:
create or replace view SALES_V as
select PDate, c.CustCode, c.CustDesc, Amount
from Sales s
join Customer c on (s.CustCode=c.CustCode);
For the sake of speed, I create a materialized view SALES_TOT_MV with columns: PDate, Amount with the following SQL:
create materialized view SALES_TOT_MV
enable query rewrite
as select PDate, sum(Amount) Amount from Sales
group by PDate;
When I run the following query, I expect it to be rewritten to make use of SALES_TOT_MV:
select PDate, sum(Amount) from SALES_V
group by PDate;
However, explain plan always tell me it's using SALES table, not the SALES_TOT_MV.
Can somebody tell me it's a limitation of Oracle optimizer or I'm just missing something for this?
Thanks in advance!!
- Andrew
Edited by: blackhole001 on Jan 28, 2010 12:34 PM

blackhole001 wrote:
Hi all,
I'm trying to make my programmer's life easier by creating a database view for them to query the data, so they don't have to worry about joining tables. This sounds like a pretty horrible idea. I say this because you will eventually end up with programmers that know nothing about your data model and how to properly interact with it.
Additionally, what you will get is a developer that takes one of your views and see's that of the 20 columns in it, it has 4 that he needs. If all those 4 columns comes from a simple 2 table join, but the view has 8 tables, you're wasting a tonne of resources by using the view (and heaven forbid they have to join that view to another view to get 4 of the 20 columns from that other view as well).
Ideally you'd write stored routines that satisfy exactly what is required (if you are the database resource and these other programmers are java, .net, etc... based) and the front end developers would call those routines customized for an exact purpose.
Creating views is not bad, but it's by no means a proper solution to having developers not learn or understand SQL and/or the data model.

How to write CMP on Database VIEWs

Hi
In my application i have database views.
i want to write a CMP bean for each VIEW.
is it possible To write a CMP Bean on VIEWs.
if possible Pl let me know and send some code too.
thank u

There should be no problem writing CMPs over views as long as the views are updateable. As for an example, you
might want to try the examples on OTN using a view over the correspodning tables.
Thanks -- Jeff

CMP Entity Bean from dataBase views

Hi forum,
I Have to migrate an existing application to JEE (with EJB3).
In this application there are a lot of dataBase views.
I'm new in EJB3 and I don't know how to structure cleanly entity beans (Specially when it are created from dataBase views).
My problem is that for each query made from the same dataBase view I almost have to create a new entity bean with a different @Id.
What can I do to to avoid it?
I had thought to genrate a new @Id (attribute not persistent) but I think is not possible. It's right ?
Someone can help me ?
Thanks

"CMP provides you with database independence and less coding efforts."
BMP is not database dependent, unless you invoke database specific things in your SQL (something I do not do). CMP on the otherhand is inherently appserver specific (which was it's goal when BEA, IBM, et al. came up with it), and still limits your design possibilities. See this thread for an example:
http://forum.java.sun.com/thread.jsp?forum=13&thread=318785
As for less coding effort, that is a relative statment. Yes a simple CMP bean requires less coding to develop the first time. I personally view a few lines of SQL to load and store the data as being fairly trivial. But that needs to be offset with the problems inherent in using appserver specific CMP implementations.
As an example, try mapping WebSphere CMP to a pre-existing database without using IBM's IDE. It's an incredible pain in the ass since WebSphere does not come with a "meet-in-the-middle" solution. Any J2EE developer that has had the experience of working with different appservers (especially if they have had to port an app, as I have) can attest to the complications that arise with each implementation.
A BMP bean, written with non-DB-specific SQL, is the most portable, most flexible approach to EntityBeans. Yes, it requires the developer to be able to write some SQL, which should not take a significant amout of time. WRT queries, you have to write them, either SQL, EQL, or some appserver specific format.
As an aside, the use of code generators to simplify the creation of EJBs lends itself well to BMP. By using a (or writing your own) code generator, you can mitigate the annoying SQL bugs that creep up early in development.

JDeveloper 3.0 and Database-Views

I have started to test JDeveloper and have the first problem.
I make extensive use of database-saved views and started with the Package Wizard to create default business-objects (view objects and view links).
But when I select a database-view I get a save error: Entity creation failed no primary key attributes and does not support ROWID. oracle.jbo.dt.objects.Jbo.Exception
The connection was established with jdbc.oracle.thin, the database is NT-Version 7.3.4, JDeveloper is Version 3.0 (Build 532).
What have I done wrong or is it not possible to connect to views ???
I hope for your help.
Henry
null

Dan,
JDeveloper 3.0 provides an 'Add-In API' and 'Custom Addins' for
integration with third-party vendors. These were also in place
for 2.0.
If you go to the JDeveloper page, there are HTML documents
available under the 2.0 Technical Information section, and also
an online walkthrough available if you follow the 'Online Demo'
link.
Laura
Dan Schroeder (guest) wrote:
: Will JDeveloper 3.0 support source control integration? I am
: currently MKS Source Integerty 7.5. I haven't seen any
: references about source control within JDeveloper 3.0.482, but
: someone told me that it may be available as an add on.
null

Migrate custom folders over to database views

Similar Messages

Maybe you are looking for