R dtplyr distinct.dtplyr_step 子集不同/唯一行

# S3 method for dtplyr_step
distinct(.data, ..., .keep_all = FALSE)

library(dplyr, warn.conflicts = FALSE)
df <- lazy_dt(data.frame(
  x = sample(10, 100, replace = TRUE),
  y = sample(10, 100, replace = TRUE)
))

df %>% distinct(x)
#> Source: local data table [10 x 1]
#> Call:   unique(`_DT7`[, .(x)])
#> 
#>       x
#>   <int>
#> 1    10
#> 2     4
#> 3     1
#> 4     5
#> 5     7
#> 6     2
#> # … with 4 more rows
#> 
#> # Use as.data.table()/as.data.frame()/as_tibble() to access results
df %>% distinct(x, y)
#> Source: local data table [64 x 2]
#> Call:   unique(`_DT7`)
#> 
#>       x     y
#>   <int> <int>
#> 1    10     8
#> 2     4     3
#> 3     1     8
#> 4     1     6
#> 5     5     7
#> 6     7    10
#> # … with 58 more rows
#> 
#> # Use as.data.table()/as.data.frame()/as_tibble() to access results
df %>% distinct(x, .keep_all = TRUE)
#> Source: local data table [10 x 2]
#> Call:   unique(`_DT7`, by = "x")
#> 
#>       x     y
#>   <int> <int>
#> 1    10     8
#> 2     4     3
#> 3     1     8
#> 4     5     7
#> 5     7    10
#> 6     2     7
#> # … with 4 more rows
#> 
#> # Use as.data.table()/as.data.frame()/as_tibble() to access results