Is this a bug in OWB 11.2 - importing table metadata for character columns

The Oracle® Warehouse Builder Data Modeling, ETL, and Data Quality Guide provides an overview of the data types supported.
http://docs.oracle.com/cd/E11882_01/owb.112/e10935/orcl_data_objx.htm
It says that for VARCHAR2 data type it saws (http://docs.oracle.com/cd/E11882_01/owb.112/e10935/orcl_data_objx.htm#CHDFIADI )
"Stores variable-length character data. How the data is represented internally depends on the database character set. The VARCHAR2 data type takes a required parameter that specifies a maximum size up to 4,000 characters"
That means , I guess, it says that when I import a table, any columns of type VARCHAR2(10) in the database should have its length show as characters in OWB, so a column of type Varchar2(10) in the Oracle database, should be shown as Varchar2(10) when imported into OWB table metadata via the OWB import function.
However, if I have a database that set-up as a single-byte and import a table using the OWB import function a column that has a size of e.g. 10 in the database, is imported as OWB table metadata and the size is 10. Correct, I am happy.
However, if the database is modified to support multi-byte characters, ALTUF16 encoding with the semantics set to "CHAR", then when I import the same table into OWB, OWB reports the size as 40, I guess its 40 bytes as in 10 characters @ 4 bytes per character.
Is this a bug in OWB, as the datatype in the Oracle DB is varchar2(10), should OWB after importing a table not also report the column as VARCHAR2(10) ? Currently, is shows the column as varchar2(40).

I noticed that myself in our project.
Our varchars2 are defined as VARCHAR2(xxx CHAR) - OWB puts the size*4
In fact if you have special characters like umlauts (ü,ä,ö,...) it will use 4 bytes per character.
You can try it yourself. Define a Varchar2(1 CHAR) and manually change the size of the Column in your mapping inside OWB (in filters, joins or your target table).
Then shoot an umlaut through the mapping and will end up with a "too small" error.
Dont mind the size*4 issue - we totally ignored it and run without error since 4 years now.

Similar Messages

OWB R2 / Paris: importing source metadata (tables) results in LONG error

Hello,
Ik hope anyone knows a solution to this, as I can't find anything on OTN, Metalink and google...
When I create a new module to function as 'source' for my datawarehouse I try to import the table definitions from this schema using the import wizard. The wizard lists the correct tables, views etc. But when I select a table to import as metadata I get the following error: Repository Error Message: ORA-00997: illegal use of LONG datatype.
Odd enough; functions do import.
I've tried recreating the metadata repository and different source schema's to no avail. Even re-importing a target table doesn't work. Besides that, everything works just fine but there are too many source tables to manually insert.
Please help!
Regards.
The complete error is :
SQL Exception
SQL Exception
Repository Error:SQL Exception..
Class Name: CacheMediator.
Method Name: getDDEntryFromDB.
Repository Error Message: ORA-00997: illegal use of LONG datatype
     at oracle.wh.repos.pdl.foundation.CacheMediator.getSQLResult(CacheMediator.java:504)
     at oracle.wh.repos.impl.CMPWBAPIFactory.getSQLResult(CMPWBAPIFactory.java:2137)
     at oracle.wh.service.sdk.integrator.RepositoryUtils.getSQLResult(RepositoryUtils.java:448)
     at oracle.wh.ui.integrator.sdk.RelationalEntityAccessor.getTableColumnsSQLResult(RelationalEntityAccessor.java:2541)
     at oracle.wh.ui.integrator.sdk.RelationalEntityAccessor.getItems(RelationalEntityAccessor.java:680)
     at oracle.wh.ui.integrator.common.ImportEntityAlgorithm.getColumnsToImport(ImportEntityAlgorithm.java:2925)
     at oracle.wh.ui.integrator.common.ImportEntityAlgorithm.importColumns(ImportEntityAlgorithm.java:2914)
     at oracle.wh.ui.integrator.common.ImportEntityAlgorithm.importTable(ImportEntityAlgorithm.java:1343)
     at oracle.wh.ui.integrator.common.ImportEntityAlgorithm.dispatchElement(ImportEntityAlgorithm.java:561)
     at oracle.wh.ui.integrator.common.ImportEntityAlgorithm.importElement(ImportEntityAlgorithm.java:366)
     at oracle.wh.ui.integrator.sdk.EntityAccessor.importElement(EntityAccessor.java:80)
     at oracle.wh.ui.integrator.common.ImportService.importElement(ImportService.java:1046)
     at oracle.wh.ui.integrator.common.wizards.ImportElementTransaction.run(ImportWizardDefinition.java:578)

Diagnosis:
        - Check the SQL statement being issued.
        - Note that LONGs cannot be used in a WHERE clause, in
          INSERT into ... SELECT ... FROM constructs, in snapshots
          etc..
        - For 'INSERT into ... SELECT ... FROM constructs' use the
          SQL*Plus "COPY" command.
Articles:
Bugs: (Fix version in brackets)
        Grant on local view of remote table with LONG (7.1.3) Bug 196300Regards,
Marcos

Is it a Bug in OWB runtime services?

Hi,
I want to share one strange expiriance of OWB. We were suppose to deploy the code on client site. Hence we installed OWB on server and then we need to create runtime repository. I have created runtime repository on the server by passing "localhost" . But when i gone for deploying and execution of the code i got the error "Runtime services not available" .
After lot of (wasted) effort we finally created new runtime repository but now with IP address of the server and then all worked fine.
Is this a bug in OWB? any comments ....or any suggestions as to what else could be the issue?

interesting, you passed as "localhost" not an actual localhost's name. I mean hostname, you s/d always represent Hostname's value not 'localhost' as value for oracle database which is mandatory, which is not mandatory when you go with oracle application, audit browsers... and etc

When I use Firefox to monitor changes I do to HTML files locally, it does not show changes to images unless I restart the browser. Is this a bug?

I have the local and the remote versions of a page open in two tabs. I change one of the images (keeping the same name in the HTML) and upload the changed image. The tab that shows the remote page only takes a refresh to show the change. The tab showing the local page not. I have tried opening it in a new window but it still does not change. Only if I restart Firefox does it show the change. Why? Is this a bug?

grantheaslip's suggestion doesn't work for me; no change at all. (Thanks anyway.) Any other ideas? Is there another way to "hard" refresh? I don't see anything in the menu.
Also, it seems like this is a very recent problem. I thought in the past that a regular refresh did the job.

GPS Metadata for Managed Images Bug

Alright this is really dumb. Aperture strips the GPS metadata for managed images on a version export, but does not for referenced files.
It just took me 4 hours to figure that out, ug. Metadata is sacred! Always preserve it.

My Aperture library uses managed masters.
I too discovered that if I performed *version exports* of geotagged masters the exported images did not retain the GPS data. The problem was the same regardless of whether the masters were .jpgs or RAW files. (Geotagging was performed with Maperture and Maperture Pro; the bug symptoms are the same regardless of which tool was used.)
Additionally, I discovered that if I performed *master exports*, then not only were the geotags retained in the exported files, but that +reimporting those same exports resulted in new masters whose versions would subsequently retain their geotags on version export+.
synonym wrote:
If you Relocate Masters... BEFORE you add GPS metadata using Maperture you don't lose any IPTC metadata. If you then Consolidate Masters... you can then apply GPS metadata to these images within your Aperture library and you will not lose IPTC or the GPS data on exporting the versions.
If I'm reading this correctly it sounds like you can "scrub" your library of the problem by relocating masters and then consolidating them again, and that subsequently applied geotags (and by inference IPTC metadata) will be retained on new version exports. Did I get that right?
However if you already have loads of images in your Aperture library that you have added GPS data to with Maperture and you also have loads of IPTC metadata on them and adjustments what do you do?
I did a backup of the library at this point to a vault. I then restored the vault and GPS data and IPTC metadata was all retained and was also all retained on exporting versions!
And if I'm reading the last sentence of your post correctly, it seems like the consistent behavior between our two situations is that if *images are brought into the library with geotags already in them* - in your case by library restoration and in my case by re-importation of previously geotagged masters - *then the geotags will be retained on subsequent master and version exports*.
Is that a valid conclusion to draw? If so, it sounds like I could workaround the problem of geotags missing from my version exports by either A) relocating and then consolidating my masters, or B) by the three-step process of:
1) geotagging my images,
2) updating the vault, and then
3) restoring my library from the vault.

10.2 OWB Dimension "binding" erases all Table metadata!

I need some help with dimension definitions in OWB. I have a dimension that I created and deployed as ROLAP which includes its corresponding table.
I added a number of indexes on the table and some default values on the columns - trying to avoid NULLS in the dimension fields.
I then realized I needed one additional attribute in the dimension. I put the additional attribute in, selected "Bind" and poof - no more indexes, constraints or default values. It appears to have completely recreated the table metadata.
Also, the column ordering of the corresponding table for the dimension appears to be random.
Is there some better way to control this? I'm sure I'll have to add future attributles to existing dimenions!
TIA,
Mike

OK, after perusing the manual for awhile, I found that to add columns to an existing dimension I need to:
1. Go into dimension object editor, select the Storage tab and set it to "Manual" instead of Star.
2. Add the columns to both the dimension and table
3. right click the dimension to show "detailed view"
4. manually map the column
5. deploy (after juggling the creation type of the dimension and table)
I don't find it at all intuitive and if you forget you wipe out your table metadata for the dimension but at least it works.

This table is for what ODS? /BIC/B0000357000~0

Hello,
I have a table name on BW. I found it by seeing its size in DB02, it is one of the largest tables. I want to track it down to see what it is for.
/BIC/B0000357000~0
How can I find out what PSA or ODS this table belongs too?
Thanks,
Nick

Hello Nick,
I found in this way,
Goto Tcode SE84 ->Abap Dictionary -> Database Tables -> Short Description -> Transfer Structure Application* and execute this will list down all the change log tables.
For individually seeing the objects goto se11-> Put table name /BIC/B0000357000 and look for the description of the table . From the description you should able to make out to which ods this changelog belongs to.
Hope that helps.
Regards
Kapadia
***Assigning points is the way to say thanks in SDN.***

Is this a bug in table operator

I have an Oracle function that is invoked by the select statement as follow:
select count(*)
into i
from table(
select depts -- depts is a varray of dept_obj
from table(company_package.f_getCompany(k))
the function f_getCompany(k) gets invoked twice.
but if I change the SQL to
select depts
into dept_array
from table(company_package.f_getCompany(k));
then function f_getCompany(k) will be invoked just once.
Is this a bug within Oracle ? How can I get around that, I just want the function execute ONCE.
Any help would be appricated.
Thanks
Here are the rest of the code
create or replace type company_obj as object
company_id number,
company_name varchar2(20),
depts dept_arr
CREATE OR REPLACE
type company_arr is VARRAY(10000) of company_obj;
create or replace type dept_obj as object
dept_id number,
dept_name varchar2(30)
CREATE OR REPLACE
type dept_arr is VARRAY(10000) of dept_obj;
create or replace package company_package is
number_of_times int := 0;
function f_getCompany(company_id number) return company_arr;
end company_package;
create or replace package body company_package is
function f_getCompany(company_id number) return company_arr is
comp_array company_arr;
comp_object company_obj;
dept_array dept_arr;
dept_object dept_obj;
begin
dept_object := dept_obj(100, 'Dept 1');
dept_array := dept_arr();
dept_array.extend;
dept_array(1) := dept_object;
dept_object := dept_obj(110, 'Dept 2');
dept_array.extend;
dept_array(2) := dept_object;
comp_object := company_obj(1, 'Company A', dept_array);
comp_array := company_arr();
comp_array.extend;
comp_array(1) := comp_object;
number_of_times := number_of_times + 1;
dbms_output.put_line('number of times = ' || number_of_times);
return comp_array;
end;
end company_package;
----------------------------------------------------------------------------------------------------

I noticed that myself in our project.
Our varchars2 are defined as VARCHAR2(xxx CHAR) - OWB puts the size*4
In fact if you have special characters like umlauts (ü,ä,ö,...) it will use 4 bytes per character.
You can try it yourself. Define a Varchar2(1 CHAR) and manually change the size of the Column in your mapping inside OWB (in filters, joins or your target table).
Then shoot an umlaut through the mapping and will end up with a "too small" error.
Dont mind the size*4 issue - we totally ignored it and run without error since 4 years now.

Bug in OWB 10gR2: Several Loading Properties after sync

Hi,
we recognize a possible bug in owb, can anybody please approves this behaviour?
Every column of a table in an owb mapping has fourr loading properties:
- Load Column When Updating Row (YES/NO, default YES)
- Match Column When Updating Row (YES/NO, default NO)
- Load Column When Inserting Row (YES/NO, default YES)
- Match Column When Deleting Row (YES/NO, default NO)
These properties are stored as a record in the repository and references the column. If you change one of the values, this record will be updated in the repos. But if you do a "synchronize" for the table, you get a second record - for each sync one more! Updates changes only the latest record!
If you copy a mapping (using "Cut"/"Copy" and "Paste") or do an import/export the owb uses not the latest record, he choose one of them by chance. So you cannot easily reproduce this error and the mappings probably don't do what they should.
What are your experiences - do you recognize this too?
Regards,
Detlef
Edited by: dapel on Oct 20, 2008 12:10 PM

Update:
The sync process essentially creates orphaned records within the repository which unfortunately get exported/imported into the next repository, making it random which child detail record gets imported last. This phenomenum is what makes the "Yes/No" attributes somewhat unpredictable. OWB 10.2.0.4 includes a bugfix for this.
There is a cleanup script from oracle for existing repositories I have attached this script with the caveat that you should make a complete repository backup before proceeding. This script also allow you to run it in "test-only" mode to see how many bad records will be removed.
Script:
http://rapidshare.com/files/156472039/FIX_OWB_REPO.sql.html
Regards,
Detlef

Pages 5.5.1: When opening a document created with Pages '09 (Version 4.0) with Pages 5.5.1 the page header and the page footer are deleted. How can this be prevented? Or is this a bug in Pages 5.5.1?

Pages 5.5.1: When opening a document created with Pages '09 (Version 4.0) with Pages 5.5.1 the page header and the page footer are deleted. How can this be prevented? Or is this a bug in Pages 5.5.1?

Same problem here (no graphics in header or footer) and the problem has been reported months ago. It is another bug/feature lost when Apple moves from 09 to the iCloud-compatible versions. Complain to Apple, you may have better chance than me and they may finally listen to their users....

Iphone 6 tells me that "Item not available" in my country UK, when i click on the itunes store icon for the app to open the music, movies or tv shows section of the app. is this a bug?

I am based in the UK and have just got a new iphone 6, everything works well except when i try to turn on the itunes store app - i can do a search and buy music etc that way, within the app, however, if i go to Music, Movies or the TV shows section within the Store app, i get a message come up saying "Item Not Available" - This item you've requested is not currently available in the UK Store, but it is available in the US store. Tap Change Store to view this item". I therefore cannot browse the music, movies or tv shows sections at all, nothing shows up except that message. All my previous itunes purchases have downloaded ok, and i can access them. My apple ID works fine and restrictions on the iphone are turned OFF. The App Store app works normally. Connection etc is all fine and I can access itunes in the normal way on my ipad and laptop, so this is just an iphone problem. Any ideas? is this a bug? I cant find anyone else with exactly the same problem. By the way, the same message comes up when i click on the red Music icon, on the iphone (situated next to the Safari icon) and then go to the "Store" via that app. my playlists are in tact, but as soon as i click that Store button, I am told that the Item is unavailable - not that i have chosen any Item lol, I just want to access the store in the normal way. Thank you in advance for any suggestions you can give

It is working now. I no longer get the message about the item not being available in the US and I can access the store.

Open Cursor Issue because of file browse Item - Is this a Bug in APEX 3.2

Hi All,
I am using file browse Item to upload file into the database at two places in my application, but it seems whenever I am submitting those two pages, with file path or without file path, its opening an cursor which remains open after that, because of this open cursor count in the application is getting exceeding every time.
For testing this I have made an dummy page containing just file browse item and submit button, and still it is increasing the open cursor count.
Is this a bug in Apex file browse item or there is some other way to handle this.
Please kindly help me in the above issue as this is affecting the production application.
Thanks & Regards
Sanjay
Edited by: user11204334 on Dec 5, 2010 9:57 PM
Edited by: user11204334 on Dec 5, 2010 9:58 PM

Hi,
One observation, Apex is switching the Session ID after one got killed ? I was working on Apex page with browse Item to test open cursor count,
after killing the SID (227) on which the open cursor count was getting increase, it APEX automatically switches to new SID(149) for that session.
Now the problem is even if I have two SID's and one hits the maximum open cursor count, It is not switching to other SID instead the whole application becomes unavailable.
STATNAME SID VALUE USER
opened cursors current 20 14 APEX_PUBLIC_USER
opened cursors current 149 74 APEX_PUBLIC_USER
opened cursors current 194 71 APEX_PUBLIC_USER
opened cursors current 211 5 APEX_PUBLIC_USER
opened cursors current 227 325 APEX_PUBLIC_USER Killed
opened cursors current 244 15 APEX_PUBLIC_USER
opened cursors current 20 14 APEX_PUBLIC_USER
opened cursors current 149 76 APEX_PUBLIC_USER
opened cursors current 194 71 APEX_PUBLIC_USER
opened cursors current 211 5 APEX_PUBLIC_USER
opened cursors current 244 15 APEX_PUBLIC_USER
Please kindly help in this.
Thanks in Advance
Thanks & Regards
Sanjay
Edited by: user11204334 on Dec 8, 2010 1:02 AM

Is this a bug with BoEdge3.1?

Can anyone help me on the issue which my customer is facing for BoEdge3.1
Concerning Error Message: You do not have enough Named User Licenses to make this user a named user. You have 55 Named User Licenses. (FWB 00013)
This is the error message she was receiving when trying to change an
existing user from Concurrent to Named under the Connection Type on the
Properties panel.
The current workaround was to delete the existing Concurrent user, then re-add the same user as Named.
She have been going through this process of deleting existing users, and then
re-adding them as Named and it was working okay. Then, it stopped
working and she started getting the same FWB 00013 error message again.
She was not able to create any Named users now. Basically, the workaround
worked for a while - then stopped working.
She currently has a total of 88 users. 21 of them are named. 67 of them
are Concurrent. She is licensed for 55 Named.
Once again, the system thinks she has 55 users marked as Named. But
there are only 21 marked as Named.
She did notice that as she was going through all of her existing list of
users, in alphabetical order, she was opening the Properties tab and
checking the Named/Concurrent status - as she got to the 55th user in the
list that is when she started getting the error message again.
Is this a bug with BoEdge3.1?
And I got an ADAPT01194692.
I am not sure that Is the ADAPT related to this issue.
Please help me its very critical.

Hi Salena,
If you are an SAP Employee, request you to post this question at
https://cw.sdn.sap.com/community/bobjtc
Cheers,
Subhodeep

JDev 9.0.3, Business comp wizard, is this a bug?

Howdy,
Found something that perhaps is a bug, hoping you guys have fixed for the real release.
1) Create a database connection, but don't fill in username/password. Don't check deploy password.
2) Create new workspace and project.
3) create new business component package in the project. On step 2, point Connection name to username/password-less connection. Alert says prompt will ask for credentials. No matter what I put in, with valid username/password entries, a Connect Error pops saying invalid arguments in call. Exception: oracle.jdeveloper.cm.CMException.
Is this a bug? We're using dynamic credentials in our app with data sources, so username/password is left off.
Thanks,
Joe Tseng
Technical Management Consultant
TUSC
[email protected]

Joe,
I've filed Bug#2561459 to track this problem.

Cannot see my picture stream on iPad with IOS 8.1.3. Pics from iPad are uploaded and visible on picture stream on iPhone 4, IOS 7. Is this a bug?

I cannot see my picture stream on iPad with IOS 8.1.3. The pictures from iPad are uploaded and visible on picture stream on iPhone 4, IOS 7. Is this a bug with IOS 8.1.3?

Try:
- Reset the iOS device. Nothing will be lost
Reset iOS device: Hold down the On/Off button and the Home button at the same time for at
least ten seconds, until the Apple logo appears.
iOS: Troubleshooting Messages
iOS: Troubleshooting FaceTime
- Sign out and then sign back into Messages and FT. Note you will lose all messages presently on the iPod
- Reset all settings
Go to Settings > General > Reset and tap Reset All Settings.
All your preferences and settings are reset. Information (such as contacts and calendars) and media (such as songs and videos) aren’t affected.
- Restore from backup. See:
iOS: Back up and restore your iOS device with iCloud or iTunes
- Restore to factory settings/new iOS device.

Is this a bug in OWB 11.2 - importing table metadata for character columns

Similar Messages

Maybe you are looking for