Re: R: Bad timestamp external representation '2001-05-11

(back on list)

> Ok. My initial hope was to port a sql server db to Postgresql with
> PGadminII. This tool has a feature that read from
> sql server natively, and transport data to Postgresql. I don't know if this
> dot notation is a standard! In my SQL server
> database ( I'm in Italy, but all seems configured with English locale )
> these are the datas. Probably is a bug of PGAdmin tool;
> those guys told me to talk with a Postgres expert to learn about that. So, I
> cannot change the query because the tool is not mine
> ( and because I can't compile in C, really ... ) . Probably I can change the
> data that goes into the db, but I think this 00.00.00 is
> quite a standard for sql server databases. What do you think about ?

I'm not sure where the dot notation comes from; I haven't run into it
before and I've been supporting the PostgreSQL date/time code for the
last six or seven years! It looks like it *could* be supported in the
PostgreSQL backend, but I'd like to have some idea that it is actually
something to be expected. Are we sure that the the original SQL Server
database represents dates that way, or are the dates actually stored as
text?

Do any of the pgadmin folks have some ideas? Does the tool manipulate
the format of date values?

- Thomas