tidyr - Transforming a column of arrays to column of single values in R

Question

I have two columns of arrays that I would like to give individuals cells to. For example, my columns currently look like this:

NDVI	Dates
[0.1,0.5,0.7]	[1.54E12, 1.54E12, 1.54E12]

question from:https://stackoverflow.com/questions/65910362/transforming-a-column-of-arrays-to-column-of-single-values-in-r

Answer 1

We could extract the characters that are not a ] that succeeds the [ with str_extract_all, which returns a list, then unnest the columns, and split the column values at the , followed by one or more spaces (\s+) with separate_rows

library(dplyr)
library(stringr)
library(tidyr)
df  %>% 
   mutate(across(everything(), str_extract_all, "(?<=\[)[^]]+")) %>% 
   unnest(c(NDVIs, dates)) %>% 
   separate_rows(c(NDVIs, dates), sep=",\s+", convert = TRUE)

-output

# A tibble: 198 x 2
#      NDVIs         dates
#      <dbl>         <dbl>
# 1  0.00485 1527502095000
# 2  0.0319  1544955875000
# 3  0.0347  1544955875000
# 4  0.0569  1545126378000
# 5  0.0596  1545126378000
# 6  0.0500  1545394367000
# 7  0.0523  1545394367000
# 8  0.0525  1545561339000
# 9  0.0552  1545561339000
#10 -0.0102  1545820928000
# … with 188 more rows

---

Or with the [] is not in every element, then remove the already existing one with str_remove_all and use separate_rows

df %>%
   mutate(across(everything(), str_remove_all, "\[|\]")) %>% 
   separate_rows(c(NDVIs, dates), sep=",\s+", convert = TRUE)
# A tibble: 198 x 2
#      NDVIs         dates
#      <dbl>         <dbl>
# 1  0.00485 1527502095000
# 2  0.0319  1544955875000
# 3  0.0347  1544955875000
# 4  0.0569  1545126378000
# 5  0.0596  1545126378000
# 6  0.0500  1545394367000
# 7  0.0523  1545394367000
# 8  0.0525  1545561339000
# 9  0.0552  1545561339000
#10 -0.0102  1545820928000
# … with 188 more rows