* Tom Lane (tgl(at)sss(dot)pgh(dot)pa(dot)us) wrote:
> Andrew Dunstan <andrew(at)dunslane(dot)net> writes:
> > ... Oh, and if we did allow the
> > quote char we should surely only allow it on input - just because other
> > programs produce absurd output there is not reason we should.
> Yeah. The *real* problem with the patch as proposed is that it allows a
> COPY OUT to emit a file that cannot be reloaded correctly, even given
> the same options used to prepare it. I think that the restrictions were
> put there more to prevent that scenario than to restrict COPY IN.
erp. My apologies, I hadn't ever intended for this to be used with COPY
OUT. For some reason I had thought my changes were isolated to the COPY
CSV IN path. I'd be happy to adjust the patch to only accept the
quote-in-null syntax when doing a COPY CSV IN.
In response to
pgsql-hackers by date
|Next:||From: Peter Eisentraut||Date: 2007-07-31 19:06:35|
|Subject: Re: stats_block_level|
|Previous:||From: Greg Smith||Date: 2007-07-31 18:51:03|
|Subject: Re: Machine available for community use|
pgsql-patches by date
|Next:||From: Gregory Stark||Date: 2007-07-31 19:09:56|
|Subject: Re: Export user visible function to make use of convert_to_scalar|
|Previous:||From: Tom Lane||Date: 2007-07-31 17:31:06|
|Subject: Re: Export user visible function to make use of convert_to_scalar |