> Tom Lane wrote:
> > FWIW, I don't see any obvious problem:
>
> > regression=# create table foo (f1 int, f2 date);
> > CREATE
> > regression=# insert into foo values (42, null);
> > INSERT 686666 1
> > regression=# select * from foo;
> > f1 | f2
> > ----+----
> > 42 |
> > (1 row)
> >
> >Can you be more specific about your problem?
>
> Sorry about the ambiguity.
> I am copying data from a tab delimited text file using the psql command
> test=> \copy collar from collar.txt;
> where the table "collar" contains a date field (not set to "NOT NULL") and
> the text file contains date fields for some records and nothing between
> two tabs for records without date data. The same table structure and load
> file were used successfully with 6.5.3 (although I must admit that I
> haven't switched back to the old PostgreSQL to double check).
>
> The table structure is as follows;
> DROP TABLE collar;
> CREATE TABLE collar (
> holeid varchar(10) PRIMARY KEY,
> easting float8 NOT NULL,
> northing float8 NOT NULL,
> rl float8 NOT NULL CHECK (rl > 0),
> depth float8 NOT NULL,
> basedepth float8,
> date date,
> starttime time,
> endtime time,
> location varchar(30),
> geologist varchar(20),
> driller varchar(20),
> rigid varchar(10),
> samplers varchar(30),
> company varchar(10),
> program varchar(5),
> loadbatch int2
> );
>
> Looking at it again now, I am also uncertain whether "date" is an
> allowable field name (PostrgreSQL didn't object).
>
> Cheers
> Colin Rothnie
>
>
PS Sorry about the direct mail Tom, I hit the reply button without
confirming the addressee