CODEWITHSUNDEEP
rregexdplyrtidyverse
littleworth
littleworth

Reputation: 5169

How to rename filnames stored as tibble/dataframe using dplyr

I have the following data frame which contains a list of files. 

library(tidyverse)
dat <- structure(list(source_file = structure(c("data/monroe_20180214/180131 WT PB d5/PB x10_01.tif", 
"data/monroe_20180214/180131 WT PB d5/PB x10_02.tif", "data/monroe_20180214/180131 WT PB d5/PB x10_03.tif", 
"data/monroe_20180214/180131 WT PB d5/PB x10_04.tif", "data/monroe_20180214/180131 WT PB d5/PB x10_05.tif", 
"data/monroe_20180214/180131 WT PB d5/PB x10_06.tif"), class = c("fs_path", 
"character"))), .Names = "source_file", row.names = c(NA, -6L
), class = c("tbl_df", "tbl", "data.frame"))


dat
#> # A tibble: 6 x 1
#>   source_file                                       
#>   <chr>                                             
#> 1 data/monroe_20180214/180131 WT PB d5/PB x10_01.tif
#> 2 data/monroe_20180214/180131 WT PB d5/PB x10_02.tif
#> 3 data/monroe_20180214/180131 WT PB d5/PB x10_03.tif
#> 4 data/monroe_20180214/180131 WT PB d5/PB x10_04.tif
#> 5 data/monroe_20180214/180131 WT PB d5/PB x10_05.tif
#> 6 data/monroe_20180214/180131 WT PB d5/PB x10_06.tif

What I want to do is to create second column new_filename by replacing the first two directory paths with new path pooled/ and replace whitespace with ., backslash with __. How can I achieve that?

The desired result is

  source_file                                         new_filename                                   
1 data/monroe_20180214/180131 WT PB d5/PB x10_01.tif  pooled/180131.WT.PB.d5__PB.x10_01.tif 
2 data/monroe_20180214/180131 WT PB d5/PB x10_02.tif  ...
3 data/monroe_20180214/180131 WT PB d5/PB x10_03.tif  .etc.
4 data/monroe_20180214/180131 WT PB d5/PB x10_04.tif  
5 data/monroe_20180214/180131 WT PB d5/PB x10_05.tif  
6 data/monroe_20180214/180131 WT PB d5/PB x10_06.tif

Upvotes: 0

Views: 76

Answers (2)

Ronak Shah
Ronak Shah

Reputation: 388982

One liner: 

paste0("pooled/",chartr(" /", "._",(sub("^(?:[^\\/]*\\/){2}","",dat$source_file))))


#[1] "pooled/180131.WT.PB.d5_PB.x10_01.tif"
#[2] "pooled/180131.WT.PB.d5_PB.x10_02.tif"
#[3] "pooled/180131.WT.PB.d5_PB.x10_03.tif"
#[4] "pooled/180131.WT.PB.d5_PB.x10_04.tif"
#[5] "pooled/180131.WT.PB.d5_PB.x10_05.tif"
#[6] "pooled/180131.WT.PB.d5_PB.x10_06.tif"

Here we first replace the part with first two occurrence of / with empty strings(""), then use chartr function from base R to replace spaces with dot (.) and forward slash (/) with underscore (_) and then paste the string with pooled/.

The regex for sub part has been taken from here.

Adding this in dplyr call : 

dat %>%
 mutate(new_filename =paste0("pooled/", chartr(" /", "._", 
                            (sub("^(?:[^\\/]*\\/){2}", "", source_file))))) %>%
 select(new_filename)


#new_filename                        
#  <chr>                               
#1 pooled/180131.WT.PB.d5_PB.x10_01.tif
#2 pooled/180131.WT.PB.d5_PB.x10_02.tif
#3 pooled/180131.WT.PB.d5_PB.x10_03.tif
#4 pooled/180131.WT.PB.d5_PB.x10_04.tif
#5 pooled/180131.WT.PB.d5_PB.x10_05.tif
#6 pooled/180131.WT.PB.d5_PB.x10_06.tif

Upvotes: 2

Edu
Edu

Reputation: 903

With gsub() from string you can do it as well

     dat %>% mutate(new_var = gsub("data/monroe_20180214", "pooled", source_file),
+                new_var = gsub(" ", ".", new_var), 
+                new_var = gsub("/", "_", new_var), 
+                new_var = gsub("pooled_", "pooled/", new_var))
# A tibble: 6 x 2
                                         source_file                              new_var
                                               <chr>                                <chr>
1 data/monroe_20180214/180131 WT PB d5/PB x10_01.tif pooled/180131.WT.PB.d5_PB.x10_01.tif
2 data/monroe_20180214/180131 WT PB d5/PB x10_02.tif pooled/180131.WT.PB.d5_PB.x10_02.tif
3 data/monroe_20180214/180131 WT PB d5/PB x10_03.tif pooled/180131.WT.PB.d5_PB.x10_03.tif
4 data/monroe_20180214/180131 WT PB d5/PB x10_04.tif pooled/180131.WT.PB.d5_PB.x10_04.tif
5 data/monroe_20180214/180131 WT PB d5/PB x10_05.tif pooled/180131.WT.PB.d5_PB.x10_05.tif
6 data/monroe_20180214/180131 WT PB d5/PB x10_06.tif pooled/180131.WT.PB.d5_PB.x10_06.tif

Upvotes: 2

Related Questions