Thread: Datetime -> Timestamp conversion

Datetime -> Timestamp conversion

From
Christof Petig
Date:
Your name               :      Christof Petig
Your email address      : christof.petig@wtal.de

System Configuration
---------------------
  Architecture (example: Intel Pentium)         : dual Intel Pentium II

  Operating System (example: Linux 2.0.26 ELF)  : Linux 2.2.11 ELF

  PostgreSQL version (example: PostgreSQL-6.5.1):   PostgreSQL-6.6
(yesterday's cvs)

  Compiler used (example:  gcc 2.8.0)           : egcs-1.1.1

Please enter a FULL description of your problem:
------------------------------------------------
christof=> select timestamp(datetime 'now');
ERROR:  No such function 'datetime_stamp' with the specified attributes

christof=> select timestamp(abstime(datetime 'now'));
timestamp
-----------------------------
Wed Aug 25 11:54:23 1999 MEST
(1 row)

It looks like if the SQL parser needs 'datetime_timestamp' , it looks
for 'datetime_stamp' which doesn't exist!

christof=> select datetime_timestamp(datetime 'now');
datetime_timestamp
----------------------
1999-08-25 11:55:46+02
(1 row)

Also all the different output formats for date/time are more than
confusing.
I read that TIMESTAMP (SQL92) shall replace DATETIME (PostgreSQL) in the
future, has anybody signed this work, yet? Thomas Lockhart?



Re: [BUGS] Datetime -> Timestamp conversion

From
Tom Lane
Date:
Christof Petig <christof.petig@wtal.de> writes:
> christof=> select timestamp(datetime 'now');
> ERROR:  No such function 'datetime_stamp' with the specified attributes

> christof=> select timestamp(abstime(datetime 'now'));
> [ OK ]

> It looks like if the SQL parser needs 'datetime_timestamp' , it looks
> for 'datetime_stamp' which doesn't exist!

Yes, there is an error in OID 1370 of the pg_proc table :-(.  I have a
to-do note to fix it for 6.6, but AFAIK we can't easily repair it
short of an initdb, so there will be no fix in the 6.5.* series.
You could probably poke it by hand in your local database, though,
if it's getting in your way.

> I read that TIMESTAMP (SQL92) shall replace DATETIME (PostgreSQL) in the
> future, has anybody signed this work, yet? Thomas Lockhart?

Lockhart's been muttering about doing that, but I don't know when he
plans to do it.

            regards, tom lane