Hello,
Sorry guys I made the unbelievable mistake of talking on the #mysql
channel today. Can you believe that they
actually feel that the fact that you can insert data outside the
boundaries of the data type (int for example) and
mySQL won't throw an exception is still valid ACID compliance.
There argument is that the application (user) should not have tried to
insert bad data. I was completely blown
away. The C in ACID explictly states that internal (data type boundaries
anyone) AND user defined constraints
can not be violated. I am just flabbergasted. I am going to come back to my safe PostgreSQL
world.
I need a hug.
J
Tom Lane wrote:
>"Joshua D. Drake" <jd@commandprompt.com> writes:
>
>
>> I just read a rather disturbing post....
>>
>>
>
>
>
>> PostgreSQL does not support read uncommited and repeatable read
>>isolation levels? If that is so... then PostgreSQL is NOT ACID compliant?
>>
>>
>
>Why do you find that disturbing? Read uncommitted is the very
>definition of "not ACID".
>
>
>
>> What is the real deal on this?
>>
>>
>
>We don't support read uncommitted because it's not ACID, and we don't
>support repeatable read because it doesn't map to any useful behavior in
>an MVCC model. It is legal to omit support for these isolation levels
>per spec. (I think the spec actually wants implementations to silently
>treat them as the next higher isolation level rather than complaining,
>but that seems to me like it'd just add confusion.)
>
>Anyone who tries to paint this as a big deal is just trolling.
>
> regards, tom lane
>
>
--
Command Prompt, Inc., home of Mammoth PostgreSQL - S/ODBC and S/JDBC
Postgresql support, programming shared hosting and dedicated hosting.
+1-503-222-2783 - jd@commandprompt.com - http://www.commandprompt.com
The most reliable support for the most reliable Open Source database.
