From: | Dave Cramer <dave(at)fastcrypt(dot)com> |
---|---|
To: | Patrick Welche <prlw1(at)newn(dot)cam(dot)ac(dot)uk> |
Cc: | Pgsql Hackers <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: request for sql3 compliance for the update command |
Date: | 2003-02-19 20:18:01 |
Message-ID: | 1045685881.1168.249.camel@inspiron.cramers |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Patrick,
No, they support the syntax:
update table set (col1, col2, col3) = ( val1, val2, val3 )
I have a customer with a rather large application which uses this
syntax, because they were using informix. There is also a rather
interesting 4GL project called aubit which is on sourceforge. They would
also like to see this supported for the same reasons.
Dave
On Wed, 2003-02-19 at 15:02, Patrick Welche wrote:
> > I have a large customer who is converting from informix to postgres and
> > they have made extensive use of
> >
> > update table set (col...) = ( val...)
> >
> > as a first pass would it be possible to translate this in the parser to
> >
> > update table set col=val
> >
> > It would appear that this is SQL3 compliant
> >
> > <set clause> ::=
> > <update target> <equals operator> <row value designator>
> >
> > <update target> ::=
> > <object column>
> > | <left paren> <object column list> <right paren>
> >
> >
> > or can someone think of another way?
>
> I don't understand the original problem. What does informix give you? A
> text file full of "update table set ()=()" which you then try to feed
> into postgres? In that case, why not pass said text file through a sed or
> perl script first?
>
> Cheers,
>
> Patrick
--
Dave Cramer <dave(at)fastcrypt(dot)com>
Cramer Consulting
From | Date | Subject | |
---|---|---|---|
Next Message | Neil Conway | 2003-02-19 20:30:40 | psql regression in CVS HEAD |
Previous Message | Patrick Welche | 2003-02-19 20:02:47 | Re: request for sql3 compliance for the update command |