Re: Migrating an Application from SQL Server 2000 to Oracle 10g
From: DA Morgan <damorgan_at_psoug.org>
Date: Mon, 21 Jul 2008 18:04:14 -0700
Message-ID: <1216688650.59778@bubbleator.drizzle.com>
>>> serially reusable, or inline in the product.
>>
>> BTW Daniel there is one thing with your excellent comparison paper -
>> SQL Server 2005 of course doesn't have Hierarchical queries in form of
>> start with ..connect by, but recursive queries can be performed using
>> common table expressions (in oracle world these are called subquery
>> factoring or WITH clause). So in general recursive queries are
>> available also in SQL Server just with different mechanism.
>> And I've also looked a bit in SQL Server 2008. It has quite nice Merge
>> statement wich it seems works smarter than Oracle Merge (for example
>> does only delete and not at first update and then delete).
>>
>> Gints Plivna
>> http://www.gplivna.eu
Date: Mon, 21 Jul 2008 18:04:14 -0700
Message-ID: <1216688650.59778@bubbleator.drizzle.com>
Michael Austin wrote:
> Gints Plivna wrote:>>> again, I see nothing like autonomous transactions, exception_init,
>>> I just completed a review of a massive amount of SQL Server 2008
>>> docs and have an installed copy of the product that I am using
>>> to create a similar mapping page for the newest version. But,
>>> serially reusable, or inline in the product.
>>
>> BTW Daniel there is one thing with your excellent comparison paper -
>> SQL Server 2005 of course doesn't have Hierarchical queries in form of
>> start with ..connect by, but recursive queries can be performed using
>> common table expressions (in oracle world these are called subquery
>> factoring or WITH clause). So in general recursive queries are
>> available also in SQL Server just with different mechanism.
>> And I've also looked a bit in SQL Server 2008. It has quite nice Merge
>> statement wich it seems works smarter than Oracle Merge (for example
>> does only delete and not at first update and then delete).
>>
>> Gints Plivna
>> http://www.gplivna.eu
> > > One thing I ran into is having to create sequences and triggers for all > of those "auto-increment" data types in SQLServer and MySQL that a lot > of people like to use for the PK... (which IMO is a complete waste of > compute time if the data can be a self-unique PK - but that is just me) > > Some of the procedures also had to be replicated - some with more > success than others...
You don't have to create more than one sequence ... ever. More is because you chose to do it that way. Neither are triggers required.
INSERT INTO t
(rid)
VALUES
(seq.NEXTVAL);
works just fine.
Again what you did was your choice not required by Oracle.
-- Daniel A. Morgan Oracle Ace Director & Instructor University of Washington damorgan_at_x.washington.edu (replace x with u to respond) Puget Sound Oracle Users Group www.psoug.orgReceived on Mon Jul 21 2008 - 20:04:14 CDT