I am trying to migrate source code into different versions of Oracle depending on my requirement. Currently I know Oralce 12c version supports IDENTITY column, but when I am trying to convert a table with IDENTITY column, scratch editor is converting it into SEQUENCE and Triggers.
If any option is present using sdcli.exe also, please let me know.
I am not familiar with Scratch, but can't you just use a normal column with a sequence as default value?
That is basically what Oracle creates under the covers when you use an identity column.
Some background:
https://oracle-base.com/articles/12c/identity-columns-in-oracle-12cr1
Related
When I add an Oracle Database as a Data Source to my project, the automatically generated code in the TableAdapter section is uses System.Data.OracleDataAdapter which is obsolete.
Following the documentation explains that you should use the third party Oracle software (ODAC). Given that the code is autogenerated, I am confused.
Am I missing something here?
Possible Solutions:
I could use .NET 3.5 for my Data Access Layer
Ignore the warnings
Manually Create Table Adapters
Something I am missing?
The solution is to use Oracle Developer Tools for Visual Studios. Add the Data Source in the same way as before (except the form now looks different as a result of the newly installed add on).
I'm new to use Linq to SQL. I've read that the first step to use it - you need to create "The Object Model". It is simple for MS SQL databases. But I use SQLite and I havent't found tool to create it.
I've tried DbLinq and MONO but the first tool is too old and work only with .NET Framework 3.5 and the second tool doesn't have database add-in in windows version.
How to create this model in SQLite? Thanks.
LINQ to SQL does not work with SQLite out of the box, only SQL Server and SQL Server Compact. You must use DbLinq or maybe use Entity Framework instead (if you are brave - http://erikej.blogspot.dk/2014/11/using-sqlite-with-entity-framework-6.html )
Got following error: if table name contains # (like z#users) code completion does not work.
With normal table names all works fine.
Is there any workarounds? Renaming all tables is not possible.
Sqldeveloper version 4.0.0
Oracle 10.2.0.5
I just tried this with a table called ABC#DEF - with both table and column completion in the worksheet, and it worked with no issues.
I'm on 4.0.2 and 11.2.0.4, so an upgrade to our latest SQL Developer patch might help.
I have a package which I've just replaced and the method being called (ASP.NET client) is raising an error about an incorrect number of parameters.
If Oracle is still running the old version, how can I get it to discard the old "compiled" version?
Oracle 10g won't run the old version of a package after you've done a CREATE OR REPLACE (I'm assuming that you did that both for the package spec and the package body).
If you were using 11.2 and explicitly asked it do, Oracle can run two different versions of a piece of code in two different editions which would let some clients continue using the old edition and gradually migrate to the new edition. Edition-based redefinition is new in 11.2 though.
We are using an oracle database in a project. Most of the tables represents classes or objects in the application. The application currently doesn't have a substantial amount of documentation. I am using StarUML to make up some class diagrams and such for other developers on the project to increase their understanding of the overall project. Using the database tables as a starting guide, and then making modifications to the diagrams as needed would be the absolute easiest and quickest way to get these set up. Is there any free applications that could assist me in pulling the schema out of the Oracle database and create class diagrams from them? Currently, there are 98 "objects" or classes closely modeled in the database and to create these all in a modeling application from scratch would be very time consuming.
You don't say what your target language is.
You can use Hibernate to generate schemas from an object model and mapping.
Middlegen is a tool that can create Java classes from schemas. Maybe those will help.
A 1:1 object-to-table mapping isn't always the best way to do things. It's hardly object-oriented. I'd view it as a starting point only.
I looked (briefly) through the StarUML documentation and don't see any way to import a database definition, so I'm not sure how you do this, sorry.
If you can find a way to get the data into StarURL, you could use Oracle's free SQL Developer tool to get the table definitions out as DDL or XML.
I know that Microsoft's Visio tool (Pro & Enterprise editions) can read selected tables from your Oracle database and generate models from that, but it ain't cheap. I really like Allround Automation's PL/SQL Developer as a reasonably-priced IDE targeted at programmers (vs DBAs) and I know that will generate diagrams locally. But I'm not sure it can save the metadata in a form you could use in StarUML.
Good luck,
-- Stew
It's a shame you didn't bother to respond to replies to your own question. Or am I misunderstanding the standard practice here that people just give points and move on?