Re: possible ODBC bug with '-infinity'
- From: Hiroshi Inoue <inoue(at)tpf(dot)co(dot)jp>
- To: Richard Broersma Jr <rabroersma(at)yahoo(dot)com>
- Cc: Greg Cocks <gcocks(at)stoller(dot)com>, Postgresql ODBC List <pgsql-odbc(at)postgresql(dot)org>
- Subject: Re: possible ODBC bug with '-infinity'
- Date: Sun, 02 Dec 2007 17:25:47 +0900
- Message-id: <47526C0B(dot)9060905(at)tpf(dot)co(dot)jp>
Richard Broersma Jr wrote:
--- On Fri, 11/30/07, Greg Cocks <gcocks(at)stoller(dot)com> wrote:
Using a MS product? Is this *maybe* related to the same
'issue' with
NULLs from PostgreSQL to MS products i.e., in the
<postgresql.conf>
file, turning 'transform_null_equals' to
'on'?
Likely not, but thought I would put it out there... :-)
.. DEFAULT '-infinity', ...
I should probably clarify what I am seeing.
I have a table that has a time stamp field. The timestamp fields have two special values in postgres:
'infinity' which implies '+infinity'
and '-infinity'
ODBC doesn't have such a concept.
when I create a recordset of records that have the value 'infinity' in the timestamp field all is well. The client side timestamp displays '9999-12-31 .... '.
Though I have no confidence, I'm doing so.
Do you think it is an appropriate handling ?
However, when I try to create a recordset that has field that contains the '-infinity' values I get an error message.
How do you think the driver should treat -infinity ?
regards,
Hiroshi Inoue
Home |
Main Index |
Thread Index