In article <439650F1.4050901@familyhealth.com.au>,
Christopher Kings-Lynne <chriskl@familyhealth.com.au> writes:
mysql> SELECT EXTRACT(MICROSECOND FROM '2003-01-02 10:30:01.00123');
>> +-------------------------------------------------------+
>> | EXTRACT(MICROSECOND FROM '2003-01-02 10:30:01.00123') |
>> +-------------------------------------------------------+
>> | 1230 |
>> +-------------------------------------------------------+
>> 1 row in set (0.00 sec)
>> Does contrary behavior from MySQL count as evidence that PostgreSQL's
>> behavior is correct? :-)
> No...I happen to think that their way is more consistent though. Pity
> it's not in the spec.
I'd say the comparison with MySQL is useless because MySQL is unable
to store microseconds in a DATETIME or TIMESTAMP column, although you
can extract microseconds from a date/time literal.