uniqueN() fails for zero-length vectors in by

[mcol]

If in uniqueN a zero-length vector is passed to by, the function throws an error. I would expect it to return 0.

library(data.table)
DT <- data.table(idx=1:4, value="val")
uniqueN(DT, character(0))
# Error in forderv(x, by = by, retGrp = TRUE, na.last = if (!na.rm) FALSE else NA) :
# Internal error: DT has 2 columns but 'by' is either not integer or is length 0

This was unexpected, as instead the following works:

uniqueN(DT[, .SD, .SDcols=character(0)])
# [1] 0

Output of sessionInfo()

R version 3.6.3 (2020-02-29)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 18.04.4 LTS

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/blas/libblas.so.3.7.1
LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.7.1

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C
 [9] LC_ADDRESS=C               LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base

other attached packages:
[1] data.table_1.12.8

loaded via a namespace (and not attached):
[1] compiler_3.6.3

[MichaelChirico]

Note that your example is not really the relevant case for consistency -- .SDcols=character() is not the same as by=character()

I'm not sure uniqueN is well-defined in this case. Should it be nrow(DT)? That would match uniqueN(DT, by = NULL) at least.

Certainly you shouldn't be getting an Internal Error so there is something to fix.

[mcol]

I think you are right in your assessment, using by=character() should match by=NULL. Thanks for providing such a quick fix!