Re: [RFC] Date/Time standard for input/output

2002-08-09 Thread H.Merijn Brand
On Sun 04 Aug 2002 22:30, Tim Bunce [EMAIL PROTECTED] wrote: On Sun, Aug 04, 2002 at 04:22:11PM +0200, Daniel Ritz wrote: hi [please CC as i'm not on the list] DBI has a real problem: input/output of date/time/timestamp values. each driver handles it different. most of the

Re: [RFC] Date/Time standard for input/output

2002-08-09 Thread Tim Bunce
On Fri, Aug 09, 2002 at 05:02:09PM +0200, H.Merijn Brand wrote: DBD::Unify Probably, but how should we be dealing with Time Zones? If I fetch a Date/Time from the DB, I get a single number, and I would have to convert that to a struct. Do whatever unify would do in terms of converting

[RFC] Date/Time standard for input/output

2002-08-04 Thread Daniel Ritz
hi [please CC as i'm not on the list] DBI has a real problem: input/output of date/time/timestamp values. each driver handles it different. most of the drivers returns string, but all in a different format. this is where DBI should set a standard. in DBD-InterBase we now have a very

Re: [RFC] Date/Time standard for input/output

2002-08-04 Thread Matt Sergeant
ODBC has a built-in standard for this - I forget exactly what it is, but it's something like {d -MM-DD}, {t HH:MM:SS} and {ts -MM-DD HH:MM:SS} (for input, that is - I think output is still database specific, but I could be wrong). Alternatively, you could specify something like a