Ian,
On 3/7/2016 4:17 PM, Ian Barwick wrote:
>
> FYI something similar has been proposed before:
>
> http://www.postgresql.org/message-id/53953EFB.8070701@2ndquadrant.com
>
> The linked thread might provide more insights into the issues surrounding
> this proposal.
It's funny how I've encountered the same issue and reached the same
conclusion as you did. The main difference is that I suggested
returning NULL values instead of throwing an error.
I read through the whole thread and it seems to me like there was quite
a bit of support for that feature, with Tom still unconvinced that this
feature is useful -- but quite a few others who see the benefit in it,
especially Java users who experience that problem first hand -- and
Rushabh complaining about white space in the patch?
I'm not sure why it was not accepted at the end?
Igal