Re: Some of my favourite 11g new features

From: DA Morgan <damorgan_at_psoug.org>
Date: Fri, 17 Aug 2007 10:13:58 -0700
Message-ID: <1187370837.593338@bubbleator.drizzle.com>

hjr.pythian_at_gmail.com wrote:
> On Aug 17, 5:39 pm, matthias.h..._at_gmail.com wrote:

>> - Allow Sequences in PL/SQL Expressions
>> - Browser-Based Enterprise Manager Integrated Interface for LogMiner
>> - Read-Only Tables (ALTER TABLE ... READ ONLY, works even for the
>> schema owner)
>> - RMAN Improved Backup Compression Performance
>>
>> Although I don't see us using 11g in the next 2 years or so, we use a
>> lot of third-party software and only finished the migration to 10g in
>> Q1 2007.
>>
>> I guess even 9i will be around for quite some time.
>>
>> Matthias

>
> Read-only tables that are read-only for the length of time it tales
> for someone to work out that 'alter table read write' is a valid piece
> of syntax really gets up my nostrils.
>
> If you declare something to be read-only, it had better actually be
> read-only, otherwise the Sarbanes-Oxley crew will be down on you like
> a tonne of hot bricks.
>
> The ONLY actual way of making a table read-only is to build it onto a
> piece of read-only media. End of story. Anything that makes you think
> its read-only when it's actually not is very, very bad news and not
> something to be welcomed at all.
>
> Just my tuppence-worth
> HJR
Calm down Howard. <g>

CREATE TABLE ro (
testcol VARCHAR2(20))

ALTER TABLE ro READ ONLY;

conn / as sysdba

CREATE OR REPLACE TRIGGER ro_enforcer
BEFORE ALTER
ON DATABASE
...

   IF ora_dict_obj_owner = ... THEN

Of course you are correct. But it will be read only for all practical purposes.

Of course there is also:

ALTER TABLE ro MODIFY CONSTRAINT pk_ro DISABLE VALIDATE;

That will take them a lot longer to learn how to disentangle.

-- 
Daniel A. Morgan
University of Washington
damorgan_at_x.washington.edu (replace x with u to respond)
Puget Sound Oracle Users Group
www.psoug.org

Received on Fri Aug 17 2007 - 12:13:58 CDT