How can I spread values from multiple rows across multiple columns in R

I feel like this is a very simple procedure and yet I haven't managed to do it yet. I have a df like this:

book_title  author
book1       bob
book1       jane
book2       alice
book2       fred
book2       greg
book3       louise
book3       sam

And I would like to transform it to look like this:

book_title  author_1  author_2  author_3
book1       bob       jane      NA
book2       alice     fred      greg
book3       louise    sam       NA

I know that spread(), pivot_wider() and separate() are not going to be of much use. All help is greatly appreciated!

I've included three ways to do this: a base R approach (i.e. using no packages), a tidyverse method and a data.table solution.

I wouldn't generally use base R for reshaping wide/long as I find the syntax a little confusing - although in this case it's not too bad. In my view, the tidyverse syntax is the most intuitive. However, with large datasets it can be much slower than data.table

base R approach

dt$author_num <- unlist(lapply(rle(dt$book_title)$lengths, seq))

reshape(
    dt,
    idvar = "book_title",
    timevar = "author_num",
    direction = "wide",
    sep = "_"
)

#   book_title author_1 author_2 author_3
# 1      book1      bob     jane     <NA>
# 3      book2    alice     fred     greg
# 6      book3   louise      sam     <NA>

tidyverse approach

library(dplyr)
library(tidyr)

dt  |>
  group_by(book_title)  |>
  mutate(author_num = paste0(
    "author_", row_number()
  ))  |>
  pivot_wider(
    id_cols = book_title,
    names_from = author_num,
    values_from = author
  )

# # A tibble: 3 × 4
# # Groups:   book_title [3]
#   book_title author_1 author_2 author_3
#   <chr>      <chr>    <chr>    <chr>   
# 1 book1      bob      jane     NA      
# 2 book2      alice    fred     greg    
# 3 book3      louise   sam      NA      

data.table approach

Here is a method using the data.table::dcast function:

library(data.table)
setDT(dt)

dt[,
    author_num := paste0("author_", seq_len(.N)),
    by = book_title
]

dcast(dt, book_title ~ author_num, value.var = "author")

#    book_title author_1 author_2 author_3
#        <char>   <char>   <char>   <char>
# 1:      book1      bob     jane     <NA>
# 2:      book2    alice     fred     greg
# 3:      book3   louise      sam     <NA>

Original data

dt  <- read.table(text = "book_title  author
book1       bob
book1       jane
book2       alice
book2       fred
book2       greg
book3       louise
book3       sam", h = T)