oracle: table names case sensitive problem

[Markus]

Hello, I am trying to run an older oracle database ( created with Vanatec OpenAccess 4.x ) with newest OpenAccess version. I imported a dumfile using oralce's imp utility and migrated the schema (VScheme.exe). When I try to start my application exception ORA-00942: Table or view does not exist. The problem is table "voa_key" is named in upper case after import from dumpfile. If I rename the table to lower case, table will be found, but there are other tables named in upper case and OpenAccess cannot found these.
With older OpenAccess version 4.3.20 it doesn't matter if table names are upper or lower case. So I think these behavior changed.
Another problem is with VSchema. Table and column names where double quoted, so if table or column name exists in upper case, object could not be found. I had to manually delete quotation marks for this objects to get sscript running.
I work with genericADO driver, odp not testd yet

Is it possible to get OpenAccess work case insensitive ?

Regards,

Markus

[Thomas]

Hello Markus,

please add one element into your backend configuration in the app.config file:
<useDelimitedSQL>false</useDelimitedSQL>

This element controls if quotation marks are used when the SQL is produced. If no quotation marks are used, the upper case form of the table name will be found by Oracle. 

Kind regards,
Thomas
the Telerik team

[Joshua Kent]

Does this work in the domain model (rlinq)? currently openaccess is quoting all of my columns and table names which is making them case sensitive and breaking some of our database tools.

Thanks,
   Josh

[Thomas]

Hi Josh,

yes, please go to the 'Show Model Settings' dialog on the DSL, select 'Database Names' and control the behavior with the 'Delimit identifiers' checkbox.

Greetings,
Thomas
the Telerik team

[Joshua Kent]

Thanks this works perfect for tables, however it looks like Triggers and Sequences are still being delimited. Is there anyway to change this behavior?

Thanks,
   Josh

[Joshua Kent]

It also appears that this configuration option breaks the checking for modifications when generating the ddl. Every time I tell it to update database from model it tries to appears to think none of my tables exist and tries to recreate them all.

[Damyan Bogoev]

Hello Joshua,

 
Firstly I want to apologize for the inconvenience caused.

1. It appears that the use delimiter is not respected when creation script for sequences and triggers is being generated. We will improve this behavior in the next major release of the product.
2. Actually this is expected behavior. The diff / merge algorithm that is used for checking the schema model differences is case-sensitive.

We will notify you in this thread once the release, containing the schema creation improvements, is available for download.

Greetings,
Damyan Bogoev
the Telerik team

[Luca Galbiati]

Hello Damyan,
any news about that? I'm using the 2013.2.701.2 version, but the problem on sequences and triggers still remains.
Thanks.
Luca,

[Doroteya]

Hi Luca,

Please accept our apologies for the inconvenience you are experiencing.

Indeed, version 2013.2.701.2 does not contain the improvement in question. It was applied in the OpenAccess ORM code, but by mistake it was not released. Next week the Q3 2013 version will be available for download and it will deliver the necessary fix.

Thank you for your patience and understanding. If you have additional questions, do not hesitate to get back to us.

 

Regards,
Doroteya
Telerik

[Allen]

Where does this key go in app.config?  Is there a way to set it in code-behind?

 

[Doroteya]

Hi Allan,

The key provided in the the post of Thomas is for the Classic API of Telerik Data Access and might no be suitable in your situation. To define a solution applicable for your case, I would kindly ask you to provide me with more details about the scenario and the behaviour you are experiencing.

I am looking forward to your feedback.

Regards,
Doroteya
Telerik