Subject: Re: [office-comment] [office-formula] SUMIF and behavior of empty cells in reference ranges
|Thanks for the reply David.|
Your explanation is indeed better, though it doesn't cope with the reference-list parameter form (minor addition can solve this).
Blank (empty) cells in test_range are ignored (they never satisfy the condition).
But I say:
As another example, and more interesting, since it is a case that does >not< behave correctly in CALC at the moment: the criterion "<>y" should be treated as the formula 'R1<>"y"' and we would obtain the value TRUE if R1 referred to an empty cell (because it is a text comparison and so the value "" would be assumed which is, indeed, NOT equal to "y"). [At the moment, in Calc, such a cell is ignored and the corresponding cell in S is not added into the result.]
This was deliberate in my post.
How do I go about proposing we change this in Calc? When I raised it as an issue it was taken as a duplicate of another issue (treating empty cells in many other cases) which risks losing the particular issue here. I was also referred to the ODF, and mention was made that it was vaguely defined there.
I agree, and want to change the ODF to be explicit about what to do with empty cells in the test_range (sic) and not to ignore the corresponding cells in this case. Saying: "empty cells can never satisfy the criterion" is in fact not true, or deliberately specifying an inconsistency. This, in turn, will encourage Calc to 'do the right thing' since it wants to be consistent with the ODF.
Is this the right place for this discussion?