complete fails when fill is a column name in a grouped df

[pwwang]

When there is a column called fill in a grouped data frame, complete always fails.

r$> library(tidyr)
 
    data <- tibble::tibble(
        x = factor(rep(c("A", "B"), 2), levels = c("A", "B", "C")),
        y = c(1, 3, 6, 4),
        fill = rep(c("F1", "F2"), each = 2)
    )
    data
# A tibble: 4 × 3
  x         y fill 
  <fct> <dbl> <chr>
1 A         1 F1   
2 B         3 F1   
3 A         6 F2   
4 B         4 F2   

r$> data |> complete(x)
# A tibble: 5 × 3
  x         y fill 
  <fct> <dbl> <chr>
1 A         1 F1   
2 A         6 F2   
3 B         3 F1   
4 B         4 F2   
5 C        NA NA   

r$> data |> dplyr::group_by(y) |> complete(x)
Error in `reframe()`:
ℹ In argument: `complete(data = pick(everything()), ..., fill = fill, explicit = explicit)`.
ℹ In group 1: `y = 1`.
Caused by error in `replace_na()`:
! `replace` must be a list, not a string.
Run `rlang::last_trace()` to see where the error occurred.

r$> data |> dplyr::group_by(fill) |> complete(x)
Error in `reframe()`:
ℹ In argument: `complete(data = pick(everything()), ..., fill = fill, explicit = explicit)`.
ℹ In group 1: `fill = "F1"`.
Caused by error in `replace_na()`:
! `replace` must be a list, not a character vector.
Run `rlang::last_trace()` to see where the error occurred.

r$> data$fill1 <- data$fill; data$fill <- NULL

r$> data |> dplyr::group_by(fill1) |> complete(x)
# A tibble: 6 × 3
# Groups:   fill1 [2]
  fill1 x         y
  <chr> <fct> <dbl>
1 F1    A         1
2 F1    B         3
3 F1    C        NA
4 F2    A         6
5 F2    B         4
6 F2    C        NA

r$> sessionInfo()
R version 4.4.1 (2024-06-14)
Platform: x86_64-conda-linux-gnu
Running under: Ubuntu 24.04 LTS

Matrix products: default
BLAS/LAPACK: /path/to/lib/libopenblasp-r0.3.27.so;  LAPACK version 3.12.0

locale:
 [1] LC_CTYPE=C.UTF-8       LC_NUMERIC=C           LC_TIME=C.UTF-8        LC_COLLATE=C.UTF-8     LC_MONETARY=C.UTF-8    LC_MESSAGES=C.UTF-8    LC_PAPER=C.UTF-8      
 [8] LC_NAME=C              LC_ADDRESS=C           LC_TELEPHONE=C         LC_MEASUREMENT=C.UTF-8 LC_IDENTIFICATION=C   

time zone: America/Los_Angeles
tzcode source: system (glibc)

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

other attached packages:
[1] tidyr_1.3.1

loaded via a namespace (and not attached):
 [1] utf8_1.2.4       R6_2.5.1         tidyselect_1.2.1 magrittr_2.0.3   glue_1.7.0       tibble_3.2.1     pkgconfig_2.0.3  dplyr_1.1.4      generics_0.1.3   lifecycle_1.0.4 
[11] cli_3.6.3        fansi_1.0.6      vctrs_0.6.5      withr_3.0.1      compiler_4.4.1   purrr_1.0.2      pillar_1.9.0     rlang_1.1.4     

[DavisVaughan]

Ah, I think tidyr:::complete.grouped_df() needs to do .env$fill and .env$explicit when passing them through since they go into a dplyr function (which will use column names first if they exist).

Probably also look at tidyr:::expand.grouped_df() and do the same with .name_repair