Unique rows in distributed table

New to Julia
1

Hi,

how do I get the unique rows of a Distributed Table in JuliaDB?

df = table(["a", "a", "a", "b", "b"], ["x", "x", "y", "x", "x"], names = (:X1, :X2), chunks = 2)

julia> df
Distributed Table with 5 rows in 2 chunks:
X1   X2
────────
"a"  "x"
"a"  "x"
"a"  "y"
"b"  "x"
"b"  "x"

Ideally, the output would look like this (i.e. be another Distributed Table):

Distributed Table with 3 rows in 2 chunks:
X1   X2
────────
"a"  "x"
"a"  "y"
"b"  "x"

Best
Jakob

2

Quick update, in case anyone else is look for this.
Not sure if this is the way you’re supposed to do it but using groupreduce with an auxiliary column works:

df = transform(df, :count => ones(length(df)))

Distributed Table with 5 rows in 2 chunks:
X1   X2   count
───────────────
"a"  "x"  1.0
"a"  "x"  1.0
"a"  "y"  1.0
"b"  "x"  1.0
"b"  "x"  1.0

df = groupreduce(+, t, (:X1, :X2), select = :count)

Distributed Table with 3 rows in 2 chunks:
X1   X2   +
─────────────
"a"  "x"  2.0
"a"  "y"  1.0
"b"  "x"  2.0