naming dynamically computed f.expression in by

[lmullany]

Is it possible to name the dynamically computed column in by()?

from datatable import dt, f, by
df = dt.Frame({"group1": ["A", "A", "B", "B", "A"],
               "group2": [1, 0, 1, 1, 1],
               "var1": [343, 345, 567, 345, 212]})
df[:, {"total":f.var1.sum()},by(f.group1=="A")]

This returns

	C0	total
▪	▪▪▪▪▪▪▪▪
0	0	912
1	1	900

i.e. rather than C0, can we name it as its being computed?.. Something similar to R, where we can do

df[, .("total" = sum(var1)), by=.(grpA = group1=="A")]

[samukweku]

There is a feature request for this; in one of the discussion threads, @st-pasha suggested the keyword as, similar to that in SQL. No implementation yet.

[samukweku]

.alias handles this now