Using UNIQUE with non-adjecent columns on different sheets - excel

Question

I have two tables on two sheets - let's say tblFruits1 and tblFruits2.
Both have a column "Name".
Apple - for example - exists on both lists.
The lists might have a different number of rows
tblFruits1 on Sheet1
Name
Color
Apple
red
Peach
yellow
Ananas
yellow
tblFruits2 on Sheet2
Name
Color
Apple
red
Cherries
red
Banana
yellow
Melone
green
Now I would like to get - on a third sheet - a UNIQUE list of names of both tables.
expected result on Sheet3
Name
Apple
Peach
Ananas
Cherries
Banana
Melone
=UNION((tblFruits1[Name],tblFruits2[Name])) returns an error.
I tried variants with SEQUENCE and INDEX but didn't succeed.
So the question is:
How can I "construct" the matrix-parameter for UNIQUE from two column-ranges on two different sheets?
(What I am looking for is a non-VBA-solution - I know how to handle this in VBA.)

Answer 1

The VSTACK function makes the Union obsolete (only available to insiders at time of writing)
Since finding the Union of several ranges is a quite usefull function on its own, I use a LAMBDA to do that. The output of that can then be passed to UNIQUE
The Lambda, which I call, unimaginatively, UNION
=LAMBDA(tabl1, tabl2,
LET(rowindex, SEQUENCE(ROWS(tabl1)+ROWS(tabl2)),
colindex, SEQUENCE(1,COLUMNS(tabl1)),
IF(rowindex<=ROWS(tabl1),
INDEX(tabl1,rowindex,colindex),
INDEX(tabl2,rowindex-ROWS(tabl1),colindex)
)
)
)
Then
=UNIQUE(Union(tblFruits1[Name],tblFruits2[Name]))
gives the result you seek

Answer 2

Try:
=LET(X,CHOOSE({1,2},tblFruits1[Name],tblFruits2[Name]),Y,COUNTA(X),Z,MOD(SEQUENCE(Y)-1,Y/2)+1,A,INDEX(X,Z,CEILING(SEQUENCE(Y)/(Y/2),1)),UNIQUE(FILTER(A,NOT(ISNA(A)))))

Answer 3

This is a solution I created where you can replace a2# and c2# with any two arrays, dynamic arrays, etc. It also deduplicates and sorts it. This works on Excel for Mac (FILTERXML is not supported)
=LET(
firstArray, a2#,
secondArray, c2#,
totalCount, COUNTA(firstArray)+COUNTA(secondArray),
firstCount, COUNTA(firstArray),
SORT(UNIQUE(MAKEARRAY(totalCount,1,LAMBDA(r,c,IF(r<=firstCount,INDEX(firstArray,r),INDEX(secondArray,r-firstCount+1))))))
)

Answer 4

Can you try like this and make your Sheet1 data and Sheet2 data into Table
an in your Sheet3 cell A2 copy paste the formula below
=UNIQUE(FILTERXML(""&TEXTJOIN("",1,(IFNA(IF({0,1},Table1[Name],Table2[Name]),"")))&"","//b"),FALSE,FALSE)

Answer 5

There is a new function that simplifies this: VSTACK
For a unique (distinct) union (as per the original question), try this:
=UNIQUE((tblFruits1[Name],tblFruits2[Name]))
And to sort them:
=SORT(UNIQUE((tblFruits1[Name],tblFruits2[Name])))

Answer 6

I don't have VSTACK or HSTACK at this point (or LAMDA either), unless I resort to keeping workbooks on slow-and-clunky Excel Online. But say you have 3 dynamic arrays (named three, four and five), each with 2 columns and a variable number of rows. Then (to keep things readable) use a named formula combo defined as =SEQUENCE(ROWS(three)+ROWS(four)+ROWS(five)). Then this works:
=IFS(
combo<=ROWS(three), three,
combo<=ROWS(three)+ROWS(four), INDEX(four,combo-ROWS(three),{1,2}),
TRUE, INDEX(five,combo-ROWS(three)-ROWS(four),{1,2})
)
You can wrap UNIQUE and/or SORT around this, if you like. If the arrays have 3 columns rather than 2 then use {1,2,3} in the formula.
Obviously you can expand this to more than 3 arrays by building more conditions into the IFS formula.
Perhaps it's worth noting that three is the same thing as INDEX(three,combo,{1,2}). And using COLUMN(three) would make the syntax more generalisable, though it doesn't allow you to re-order the columns if you want something like {2,1,3} in your output. Also, TRUE is effectively how you say "ELSE" in an IFS formula--it's a bit more concise here than combo<=ROWS(three)+ROWS(four)+ROWS(five). Using these longer forms would make the formula more symmetrical but a lot wordier!
(It would be nice if VSTACK ever becomes available on the desktop.)