Re: [HACKERS] INSERT and parentheses

Marko Tiikkaja writes: > [ patch for a misleading error message ] I've committed the attached revised version of this patch. The main change is to only emit the hint if the insertion source can be shown to be a RowExpr with exactly the number of columns expected by the INSERT. This should avoid

Re: [HACKERS] INSERT and parentheses

On 2010-08-24 8:25 AM +0300, igor polishchuk wrote: Marko et al, This is my first ever attempt of a patch review just for learning the procedure. I'm not a postgres developer, so the review is partial and mostly from the usability prospective. That's all right. I'm sure any help is appreciated

[HACKERS] INSERT and parentheses

Marko et al, This is my first ever attempt of a patch review just for learning the procedure. I'm not a postgres developer, so the review is partial and mostly from the usability prospective. The original message id of the patch is 4bd58db3.4070...@cs.helsinki.fi

Re: [HACKERS] INSERT and parentheses

I have added this to the next commit-fest: https://commitfest.postgresql.org/action/commitfest_view?id=6 --- Marko Tiikkaja wrote: > Hi, > > This came up on IRC today and I recall several instances of this during

Re: [HACKERS] INSERT and parentheses

Stephen Frost writes: > Not to be a pain, but the hint really is kind of terrible.. It'd > probably be better if you included somewhere that the insert appears to > be a single column with a record-type rather than multiple columns of > non-composite type.. I don't much care for the test, either

Re: [HACKERS] INSERT and parentheses

* Robert Haas (robertmh...@gmail.com) wrote: > > The first version is a lot more common and as it turns out, is sometimes > > very hard to spot.  This patch attaches a HINT message to these two > > cases.  The message itself could probably be a lot better, but I can't > > think of anything. > > > >

Re: [HACKERS] INSERT and parentheses

On Mon, Apr 26, 2010 at 8:57 AM, Marko Tiikkaja wrote: > Hi, > > This came up on IRC today and I recall several instances of this during > the last two months or so, so I decided to send a patch.  The problem in > question occurs when you have extra parentheses in an INSERT list: > > INSERT INTO f

[HACKERS] INSERT and parentheses

Hi, This came up on IRC today and I recall several instances of this during the last two months or so, so I decided to send a patch. The problem in question occurs when you have extra parentheses in an INSERT list: INSERT INTO foo(a,b,c) SELECT (a,b,c) FROM ..; or INSERT INTO foo(a,b,c) VALUES((