mutate_if does not behave as expected

[burkovae]

It seems mutate_if modifies all columns if .predicate returns an empty number of columns

sampleData <- tibble(numbers = 1:10, factors = factor(sample(letters[1:3], size = 10, replace = T)))

# does work
sampleData %>%
  mutate_if(is.factor, funs(length)) # whaterever function

# does work
sampleData %>%
  mutate_if(is.numeric, funs(mean)) # whaterever function

# does not work -> expect no changes to factor column
sampleData %>%
  select(- numbers) %>%
  mutate_if(is.numeric, funs(mean)) # whaterever function

# does not work -> expect no changes to numbers column
sampleData %>%
  select(- factors) %>%
  mutate_if(is.factor, funs(length)) # whaterever function

Using dplyr * 0.5.0 2016-06-24 CRAN (R 3.3.1)

[cderv]

Duplicates of #2181, #2009, #1989 and already fixed in dev version by #2011 .

library(dplyr) # dev version 0.5.0.9000

sampleData <- tibble::tibble(numbers = 1:10, factors = factor(sample(letters[1:3], size = 10, replace = T)))

# does work
sampleData %>%
  mutate_if(is.factor, funs(length)) 
#> # A tibble: 10 x 2
#>    numbers factors
#>      <int>   <int>
#> 1        1      10
#> 2        2      10
#> 3        3      10
#> 4        4      10
#> 5        5      10
#> 6        6      10
#> 7        7      10
#> 8        8      10
#> 9        9      10
#> 10      10      10

# does work
sampleData %>%
  mutate_if(is.numeric, funs(mean)) 
#> # A tibble: 10 x 2
#>    numbers factors
#>      <dbl>  <fctr>
#> 1      5.5       b
#> 2      5.5       b
#> 3      5.5       a
#> 4      5.5       b
#> 5      5.5       a
#> 6      5.5       a
#> 7      5.5       a
#> 8      5.5       a
#> 9      5.5       a
#> 10     5.5       c

# does work -> expect no changes to factor column
sampleData %>%
  select(- numbers) %>%
  mutate_if(is.numeric, funs(mean)) 
#> # A tibble: 10 x 1
#>    factors
#>     <fctr>
#> 1        b
#> 2        b
#> 3        a
#> 4        b
#> 5        a
#> 6        a
#> 7        a
#> 8        a
#> 9        a
#> 10       c

# does work -> expect no changes to numbers column
sampleData %>%
  select(- factors) %>%
  mutate_if(is.factor, funs(length)) 
#> # A tibble: 10 x 1
#>    numbers
#>      <int>
#> 1        1
#> 2        2
#> 3        3
#> 4        4
#> 5        5
#> 6        6
#> 7        7
#> 8        8
#> 9        9
#> 10      10

[krlmlr]

Works for me as advertised by @cderv in the dev version.

[hadley]

@cderv thanks for collecting the related issues together!