> Thomas, any status on this? If not, I should add it to the TODO list.
Well, sure, there is *always* status ;)
I started coding a couple of days ago. So far, no showstoppers.
There are two related issues:
1) I should recode TIME WITH TIME ZONE to conform to SQL99. I had done
it originally with a "persistant time zone" since the SQL9x definition
is weaker than I imagined at the time. afaict it is still a minimally
useful data type.
2) upgrading from 7.1.x to 7.2 will likely require a schema change,
since 7.1.x TIMESTAMP should become TIMESTAMP WITH TIME ZONE, but afaik
someone took out that feature during the 7.1.x series of releases. So a
7.1.x dump will give columns labeled as TIMESTAMP but with values
containing time zones. We *might* want to accept (and ignore?) time zone
fields in TIMESTAMP values for input for 7.2, but that would still leave
folks expecting a data type which recognizes time zones needing to
adjust their schemas during the upgrade.
In response to
pgsql-hackers by date
|Next:||From: Peter Harvey||Date: 2001-09-06 06:49:21|
|Subject: Inherited Table|
|Previous:||From: johan27||Date: 2001-09-06 05:22:07|
|Subject: I have a serious problem with hanging backends.:SELECT waiting|