[Discuss][C++][Statistics] Should arrow::Array::Equals() checks wheter two arrow::Array have the same arrow::ArrayStatistics

[andishgar]

The current implementation of arrow::Array::Equals() ignores the value of arrow::ArrayStatistics. Should this method take arrow::ArrayStatistics into account when comparing arrays?

Component(s)

C++

[kou]

I think that we should ignore statistics because metadata in data type is also ignored. (Array data is only compared.)

@pitrou @wgtmac @mapleFU @zanmato1984 @bkietz @westonpace @lidavidm Do you have any opinion for this?

[lidavidm]

+1 to ignore statistics

[wgtmac]

+1 to ignore it. statistics are totally optional and two arrays with identical values may use different lower and upper bounds.

[bkietz]

+1 to ignore stats. I think this is worth explicit documentation in the doccomment for Array::statistics(), with the example given by @wgtmac; max=3,min=1 is an accurate description of [1,2,1,3] but so is max=3,min=0,is_min_exact=false so statistics are not considered part of an Array's value

[pitrou]

+1, but we may also later add an option to take statistics into account if that can make tests easier to write

[zanmato1984]

+1 to ignore stats, due to its nature of being optional and varying among instances of arrays having the same data.

[kou]

Thanks.

Let's ignore statistics.

but we may also later add an option to take statistics into account if that can make tests easier to write

Good point. We will add AssertArraysSomething to arrow/testing/ for the case.

[andishgar]

+1 to ignore stats. I think this is worth explicit documentation in the doccomment for Array::statistics(), with the example given by @wgtmac; max=3,min=1 is an accurate description of [1,2,1,3] but so is max=3,min=0,is_min_exact=false so statistics are not considered part of an Array's value

@kou As @bkietz suggested, I also think it would be helpful to document this.

[kou]

OK. Let's document this.