0

I have up to 30 data frames with a common ID column. There are other colums in each df but im just showing the ID here.

Library      DF1         DF2      DF3

ID#          ID#         ID#      ....
1111         1111        1112     ....
2222         1111        3333     ....
3333         3333        3333     ....
4444         2222        4444     ....

I have to compare the ID# colum in each of theses tables to the library ID colum to make sure the Id number matches an ID number in the library.

Currently I use dplyr and do...

DF1 %>%
      anti_join(library, by = 'ID#')

and the same for each table. It would just return any ID numbers that are not in the library, I do this same command for every data table but would like to run it for all 30 of my tables. I put all my DFs in a list but I'm not really sure how to proceed, for loop? apply? any help would be appreciated, as this pushes my boundaries of R knowledge.

Numan Karim
  • 43
  • 1
  • 6

1 Answers1

4

You can iterate over a list of data.frame using purrr. Here is an example using 3 data.frame to extract ID that are not common with reference one.

You can use any map_* fonction taht suits you best any function you want inside a map_* function

see purrr website for more info


library(dplyr)
#> 
#> Attachement du package : 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union
set.seed(999)
df_library <- data_frame(ID = sort(sample(1:12, 10)))
df1 <- data_frame(ID = sort(sample(1:12, 10)))
df2 <- data_frame(ID = sort(sample(1:12, 10)))
df3 <- data_frame(ID = sort(sample(1:12, 10)))


library(purrr)
#> 
#> Attachement du package : 'purrr'
#> The following objects are masked from 'package:dplyr':
#> 
#>     contains, order_by

list(df1 = df1, df2 = df2, df3 = df3) %>%
  map_df(~ anti_join(.x, df_library, by = "ID"), .id = "df_name")
#> # A tibble: 4 x 2
#>   df_name    ID
#>     <chr> <int>
#> 1     df1    12
#> 2     df2    12
#> 3     df3     3
#> 4     df3    12
cderv
  • 6,272
  • 1
  • 21
  • 31