db-derby-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Trevor Squires <tre...@protocool.com>
Subject Re: identity column causes grief when populating table - suggestions?
Date Tue, 22 Feb 2005 19:57:45 GMT
Thanks Ali,

the docs I've found only show support for "generated always as" 
tinkering with variations on that syntax doesn't seem to work...

I agree that it would solve my problem - if it existed today.

Thanks,
Trevor

Suavi Ali Demir wrote:
> Will derby support "default" keyword as in:
> 
> create table a(
> id int not null generated by DEFAULT as identity,
> ...
> 
> or something like that? That would solve these
> problems as it would only generate values if we do not
> specify the value.
> 
> Regards,
> Ali
> 
> 
> 
> --- Trevor Squires <trevor@protocool.com> wrote:
> 
>>Hi,
>>
>>I've been googling all morning and can't find info
>>to solve this problem 
>>(using derby of course), I hope someone here can
>>help...
>>
>>As I write my application I have a script which
>>blats and recreates my 
>>database and then populates the tables with test
>>data.
>>
>>However, many of my tables have identity columns and
>>it's causing the 
>>following error when I insert the data:
>>
>>  11:06:26  [INSERT - 0 row(s), 0.016 secs]  [Error
>>Code: 30000, SQL 
>>State: 42Z23]  Attempt to modify an identity column
>>'ID'.
>>
>>Here's my table:
>>
>>create table sample (
>>	id int not null generated always as identity,
>>	description varchar(128) not null,
>>	quantity double not null,
>>         constraint product_pk primary key (id)
>>);
>>
>>Here's my insert statement
>>
>>insert into sample (ID, DESCRIPTION, QUANTITY)
>>values (1, 'blah', 1.1);
>>
>>I've tried to create the table without "generated
>>always as identity" 
>>and then altering the table after the insert(s).
>>
>>Unfortunately I can't get the syntax right and I
>>can't figure out if 
>>there's a way to turn off identity generation
>>temporarily.
>>
>>Does anyone have any suggestions like pointer to the
>>right way to do 
>>this or a relatively painless workaround?
>>
>>Thanks for listening,
>>Trevor
>>
>>
> 
> 


Mime
View raw message