RE: Postgresql/Cold Fusion cfinsert date problems

Is this a formatting issue, or are you actually getting different dates from
the ones that you are entering?
If the issue is presentation/formatting, just use a function to get the look
you want:

http://postgresql.planetmirror.com/users-lounge/docs/7.1/postgres/functions-
formatting.html

If the dates are actually being altered, there are basically two
possibilities:
1. You have set up some constraint or trigger which is producing this
unwanted result as a side effect, or:
2. Your cold fusion app is doing it.

If you suspect 1 above, just post your table and/or trigger definitions and
someone will enlighten you.
If 2: someone who knows about cold fusion will need to answer.

> -----Original Message-----
> From: Paul Haddon [SMTP:paullist(at)mail(dot)hartingdale(dot)com(dot)au]
> Sent: Monday, July 02, 2001 9:43 PM
> To: pgsql-odbc(at)postgresql(dot)org
> Subject: [ODBC] Postgresql/Cold Fusion cfinsert date problems
>
> I'm evaluating Postgresql on Linux as a database back end
> to Cold Fusion on NT, and running into a problem with dates.
>
> The test setup is: Cold Fusion 5 on Win2000; the latest
> ODBC driver from odbc.postgresql.org; Postgresql 7 on Linux.
>
> Cold Fusion has two methods of inserting records into the
> database. The first is where you wrap a standard 'insert into...'
> SQL command in a <cfquery..> tag. This is working correctly.
>
> The second method takes defined form variables and inserts them
> into a specified table. Essentially it is building the SQL command for
> you.
> eg: <cfinsert datasource="postgresql" tablename="test_table"
> formfields="fielda, fieldb">
>
> The problem with this method is that the date that ends up in date
> fields in the table is always the servers date, regardless of what
> information is placed into the form field, and in what format.
>
> Is this a Cold Fusion oddity; is there some setting in Cold Fusion
> or the odbc driver I can modify; is there a way of formatting date
> variables to get around the problem? Or do I look at not using
> the cfinsert method altogether?
>
> Thanks
>
> Paul Haddon
> Technical Services Manager
> Hartingdale Internet
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 4: Don't 'kill -9' the postmaster