make fill argument in complete work similarly to value_fill in pivot: assume everything() if a single value is passed

[luispfonseca]

Is it possible to make the fill argument in complete work the same way as the value_fill argument in pivot_wider, i.e., if only a single value is passed, assume this applies to all variables? This would make it easier to deal with many variables, without having to specifically write a list with every single variable and the value I want to fill in with.

Apologies if I have missed a way to actually do this. In any case, consistency across the two functions could be useful.

library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union
library(tidyr) #1.1.3

# with values_fill in pivot_wider, if I provide a single value, default is to assume propagation to all variables
pivot_wider(fish_encounters, names_from = station, values_from = seen, values_fill = 2)
#> # A tibble: 19 x 12
#>    fish  Release I80_1 Lisbon  Rstr Base_TD   BCE   BCW  BCE2  BCW2   MAE   MAW
#>    <fct>   <int> <int>  <int> <int>   <int> <int> <int> <int> <int> <int> <int>
#>  1 4842        1     1      1     1       1     1     1     1     1     1     1
#>  2 4843        1     1      1     1       1     1     1     1     1     1     1
#>  3 4844        1     1      1     1       1     1     1     1     1     1     1
#>  4 4845        1     1      1     1       1     2     2     2     2     2     2
#>  5 4847        1     1      1     2       2     2     2     2     2     2     2
#>  6 4848        1     1      1     1       2     2     2     2     2     2     2
#>  7 4849        1     1      2     2       2     2     2     2     2     2     2
#>  8 4850        1     1      2     1       1     1     1     2     2     2     2
#>  9 4851        1     1      2     2       2     2     2     2     2     2     2
#> 10 4854        1     1      2     2       2     2     2     2     2     2     2
#> 11 4855        1     1      1     1       1     2     2     2     2     2     2
#> 12 4857        1     1      1     1       1     1     1     1     1     2     2
#> 13 4858        1     1      1     1       1     1     1     1     1     1     1
#> 14 4859        1     1      1     1       1     2     2     2     2     2     2
#> 15 4861        1     1      1     1       1     1     1     1     1     1     1
#> 16 4862        1     1      1     1       1     1     1     1     1     2     2
#> 17 4863        1     1      2     2       2     2     2     2     2     2     2
#> 18 4864        1     1      2     2       2     2     2     2     2     2     2
#> 19 4865        1     1      1     2       2     2     2     2     2     2     2

# fill argument in complete does not work the same way

# error
complete(fish_encounters, station, fish, fill = 2)
#> Error in replace_na.data.frame(full, replace = fill): is_list(replace) is not TRUE

# appears to work, but doesn't..
complete(fish_encounters, station, fish, fill = list(2)) %>% filter(seen == 2)
#> # A tibble: 0 x 3
#> # … with 3 variables: station <fct>, fish <fct>, seen <int>

# ..they're NAs
complete(fish_encounters, station, fish, fill = list(2)) %>% filter(is.na(seen))
#> # A tibble: 95 x 3
#>    station fish   seen
#>    <fct>   <fct> <int>
#>  1 Lisbon  4849     NA
#>  2 Lisbon  4850     NA
#>  3 Lisbon  4851     NA
#>  4 Lisbon  4854     NA
#>  5 Lisbon  4863     NA
#>  6 Lisbon  4864     NA
#>  7 Rstr    4847     NA
#>  8 Rstr    4849     NA
#>  9 Rstr    4851     NA
#> 10 Rstr    4854     NA
#> # … with 85 more rows

# works if i name the variable individually, but cumbersome when I have many variables
complete(fish_encounters, station, fish, fill = list(seen = 2)) %>% filter(seen == 2)
#> # A tibble: 95 x 3
#>    station fish   seen
#>    <fct>   <fct> <dbl>
#>  1 Lisbon  4849      2
#>  2 Lisbon  4850      2
#>  3 Lisbon  4851      2
#>  4 Lisbon  4854      2
#>  5 Lisbon  4863      2
#>  6 Lisbon  4864      2
#>  7 Rstr    4847      2
#>  8 Rstr    4849      2
#>  9 Rstr    4851      2
#> 10 Rstr    4854      2
#> # … with 85 more rows

^{Created on 2021-03-24 by the reprex package (v0.3.0)}

[hadley]

Duplicate of #1108

[hadley]

(since this is implemented via replace_na())