Another post-rows-deletion-by-reference FR/suggestion

[iagogv3]

setUnique(dt) would remove duplicate rows by reference, so the same as

dt[which(duplicated(dt)), .ROW := NULL].

Maybe some "inside" operation, like dt[, setUnique(.SD)]???

Thanks!

[venom1204]

hi @tdhock @joshhwuu @iagogv3
i have a proposed fix for the issue

• R Layer: a setUnique wrapper using existing duplicated() logic to find indices.
• C Layer: an in-place "squeeze" in deleterows.c using memmove and R_resizeVector to shorten vectors without reallocating.

if this sounds good i will be happy to proceed with teh pr