#44 - Group Rows And Combine Non-Null Values in Each of The Non-Grouping Columns

Problem description & analysis:

In the Excel table below, the 1^st column is the category; columns from the 2^nd to the 42^nd are parallel columns of data items (below only shows some of the columns), where there are two types of values – X and null. Occasionally, there are duplicate values in a column under the same category.

	A	B	C	D	E
1	ID	Criteria1	Criteria2	Criteria3	Criteria4
2	FirstValue	X
3	FirstValue		X
4	FirstValue			X
5	FirstValue				X
6	SecondValue	X
7	SecondValue			X
8	SecondValue				X
9	ThirdValue		X
10	ThirdValue		X
11	ThirdValue				X

Task: Group rows and combine values in each column in each group; only display one of the duplicate values if there are any.

	A	B	C	D	E
13	ID	Criteria1	Criteria2	Criteria3	Criteria4
14	FirstValue	X	X	X	X
15	SecondValue	X		X	X
16	ThirdValue		X		X

Solution:

Use SPL XLL to do this:

=spl("=?.group(~1).(g=~,(r1=~1).(g.(~(r1.#)).ifn()))",A2:E11)

As shown in the picture below:

Explanation:

group()function groups data and computes data in each group. ifn() function returns the first non-null member in the sequence, and it returns null if each member of the sequence is null. ~ is the current group; ~1 represents the 1^st row of the current group; and # is the ordinal number of the current member.