On Wed, Oct 13, 2010 at 10:35 PM, Mladen Gogala
wrote:
> You will see that for most of the columns, the length of the histogram array
> corresponds to the value of the default_statistics_target parameter. For
> those that are smaller, the size is the total number of values in the column
> in the s
Ants Aasma wrote:
I hit an issue with window aggregate costing while experimenting with
providing a count of the full match along side a limited result set.
Seems that the window aggregate node doesn't take into account that it
has to consume the whole input before outputting the first row. When
Ants Aasma writes:
> Seems that the window aggregate node doesn't take into account that it
> has to consume the whole input before outputting the first row.
Well, the reason it doesn't assume that is it's not true ;-). In this
particular case it's true, but more generally you only have to read
Ants Aasma wrote:
I hit an issue with window aggregate costing while experimenting with
providing a count of the full match along side a limited result set.
Seems that the window aggregate node doesn't take into account that it
has to consume the whole input before outputting the first row. When
I hit an issue with window aggregate costing while experimenting with
providing a count of the full match along side a limited result set.
Seems that the window aggregate node doesn't take into account that it
has to consume the whole input before outputting the first row. When
this is combined wit
