[upstream rENA 0.3.1] ena.rotation.h emits a data.table warning on every call

[HUDongpin]

This is an issue in rENA 0.3.1, not in ena-python. It is tracked here because it came out of porting rENA to Python. Reported as an observation, not an accusation — it may be fixed upstream or intended, in which case please correct me. Full write-up: docs/rena-upstream-issues.md

Summary

ena.rotation.h emits a data.table warning on every call. The rotation itself is correct.

Severity: trivial — console noise.

Setup (rENA 0.3.1, R 4.4.2)

library(rENA)
rows <- data.frame(
  unit  = rep(paste0("U", 1:6), each = 4),
  conv  = rep(rep(c("c1", "c2"), each = 2), times = 6),
  grp   = rep(c("g1", "g2"), each = 12),
  score = rep(c(10, 20, 30, 40, 50, 60), each = 4),
  A = c(1,0,1,0, 0,1,1,1, 1,1,0,1, 0,0,1,0, 1,0,0,1, 0,1,1,0),
  B = c(0,1,1,1, 1,1,0,0, 0,1,1,0, 1,0,0,1, 0,1,1,0, 1,0,0,1),
  C = c(1,1,0,0, 1,0,1,0, 1,0,0,1, 0,1,1,1, 1,1,0,0, 0,0,1,1),
  D = c(0,0,1,1, 0,1,0,1, 0,1,1,0, 1,1,0,0, 0,0,1,1, 1,1,0,0),
  stringsAsFactors = FALSE)
acc <- ena.accumulate.data(
  units = rows[, "unit", drop = FALSE], conversation = rows[, "conv", drop = FALSE],
  metadata = rows[, c("grp", "score")], codes = rows[, c("A","B","C","D")],
  window.size.back = 2)
set <- ena.make.set(acc, dimensions = 2)

Reproduce

ena.rotation.h(set, list(x_var = "grp"))
#> Warning in `[.data.table`(data, , ..value_vars) :
#>   Both 'value_vars' and '..value_vars' exist in calling scope.
#>   Please remove the '..value_vars' variable in calling scope for clarity.

Root cause

The data.table ..var prefix collides with a local named value_vars inside ena.rotation.h.

Suggested fix

Rename the local, or index with .SDcols.

What ena-python does

Nothing to reproduce — ena-python matches this rotation exactly, including with control_vars. Noted only for anyone comparing console output.

[HUDongpin]

Reported upstream: rENA #53 (2026-07-16).

The report there pins its source references to rENA main @ a09a4b20, verified byte-identical to the released 0.3.1 this was found against — so it applies to their current tree, not just the CRAN build.

This issue stays open here as ena-python's record of the divergence; it can close when the upstream one resolves.