Add columns recursively by string match occurrence R gsub regex-CodePudding

I have an R df where one column, Author, looks like this (the names themselves are replaced by 'Last' and 'First' in this post for anonymity):

id	Author
1	Last, First & Last, First
2	Last, First & Last, First & Last, First & Last, First
3	Last, First & Last, First & Last, First

I need to add a new column for the first and last name of every author. It should look like this:

id	First1	Last1	First2	Last2	First3	Last3	First4	Last4
1	First	Last	First	Last
2	First	Last	First	Last	First	Last	First	Last
3	First	Last	First	Last	First	Last

Of course, not all readings have the same number of authors, so I'm unable to append a specific number of columns for all rows.

** NOTE: I've done this for columns that only have one author using

data <- data %>% 
  mutate(FirstName=unlist(lapply(strsplit(Author,", "),function(x) x[2])),
         LastName=gsub(",.*","",Author))

How can I do this?

CodePudding user response：

One option can be:

df %>%
 mutate(map_dfr(.x = str_split(Author, ", | & "),
                ~ set_names(.x, ave(.x, .x, FUN = function(y) paste0(y, cumsum(duplicated(y))   1)))))

  id                                                Author Last1 First1 Last2 First2 Last3 First3 Last4 First4
1  1                             Last, First & Last, First  Last  First  Last  First  <NA>   <NA>  <NA>   <NA>
2  2 Last, First & Last, First & Last, First & Last, First  Last  First  Last  First  Last  First  Last  First
3  3               Last, First & Last, First & Last, First  Last  First  Last  First  Last  First  <NA>   <NA>

CodePudding user response：

Here I use a small helper function to return a list of last and first names

get_names <- function(x) {
  authors = strsplit(x, " & ")[[1]]
  lapply(authors, \(a) setNames(strsplit(a, ", ")[[1]], c("Last", "First")))
}

Then I apply it to each id, unnest, and pivot wide

df %>% 
  group_by(id) %>% 
  summarize(names=list(get_names(Author))) %>% 
  unnest(names) %>% 
  unnest_wider(names) %>% 
  group_by(id) %>% 
  mutate(rn=row_number()) %>% 
  pivot_wider(id, names_from =rn, names_sep="",values_from = c(First,Last),names_vary="slowest")

CodePudding user response：

Here's a version using tidyr::separate:

library(tidyr)
library(stringr)
n_auth = max(str_count(df$Author, pattern = "&"))   1
df %>%
  separate(
    Author,
    sep = " *[,&] ",
    into = c(outer(c("First", "Last"), 1:n_auth, FUN = paste0)),
    fill = "right"
)
#   id First1 Last1 First2 Last2 First3 Last3 First4 Last4
# 1  1   Last First   Last First   <NA>  <NA>   <NA>  <NA>
# 2  2   Last First   Last First   Last First   Last First
# 3  3   Last First   Last First   Last First   <NA>  <NA>

CodePudding user response：

In base R you will do:

(df1 <- read.table(text=gsub('[,&]', '', df$Author), fill=TRUE))

   V1    V2   V3    V4   V5    V6   V7    V8
1 Last First Last First                      
2 Last First Last First Last First Last First
3 Last First Last First Last First

You can then add names:

names(df1)<-paste0(c("Last", "First"), gl(ncol(df1), 2,ncol(df1)))
df1

  Last1 First1 Last2 First2 Last3 First3 Last4 First4
1  Last  First  Last  First                          
2  Last  First  Last  First  Last  First  Last  First
3  Last  First  Last  First  Last  First

of course you can rearrange the table to have first before last:

df1[order(sub("\\D ", '',names(df1)), sub("\\d ", '', names(df1)))]

  First1 Last1 First2 Last2 First3 Last3 First4 Last4
1  First  Last  First  Last                          
2  First  Last  First  Last  First  Last  First  Last
3  First  Last  First  Last  First  Last