R nested tibble map2 comparisons

Question

I'm trying to use map2 to compare nested tibble columns. Here is my data format:

> tbl
# A tibble: 3 x 3
  ID    data.x           data.y          
  <chr> <list>           <list>          
1 a     <tibble [2 x 2]> <tibble [2 x 2]>
2 b     <tibble [2 x 2]> <tibble [2 x 2]>
3 c     <tibble [2 x 2]> <tibble [2 x 2]>

tibbles in data.x and data.y are identical from column name perspective, values might be different. I would like to get maximum value from val column. I thought that this would work, but only returns max for data.x. I don't fully grasp idea how map2 works.

tbl %>%
  mutate(col1 = map2_dbl(data.x, data.y, ~ max(.$val)))

result should be:

# A tibble: 3 x 4
  ID    data.x           data.y            col1
  <chr> <list>           <list>           <dbl>
1 a     <tibble [2 x 2]> <tibble [2 x 2]>    7.
2 b     <tibble [2 x 2]> <tibble [2 x 2]>    8.
3 c     <tibble [2 x 2]> <tibble [2 x 2]>    8.

data:

> dput(tbl)
structure(list(ID = c("a", "b", "c"), data.x = list(structure(list(
    text = c("Y", "Y"), val = c(1, 1)), .Names = c("text", "val"
), row.names = c(NA, -2L), class = c("tbl_df", "tbl", "data.frame"
)), structure(list(text = c("N", "N"), val = c(2, 2)), .Names = c("text", 
"val"), row.names = c(NA, -2L), class = c("tbl_df", "tbl", "data.frame"
)), structure(list(text = c("Y", "Y"), val = c(3, 3)), .Names = c("text", 
"val"), row.names = c(NA, -2L), class = c("tbl_df", "tbl", "data.frame"
))), data.y = list(structure(list(text = c("Y", "Y"), val = c(6, 
7)), .Names = c("text", "val"), row.names = c(NA, -2L), class = c("tbl_df", 
"tbl", "data.frame")), structure(list(text = c("Y", "Y"), val = c(8, 
6)), .Names = c("text", "val"), row.names = c(NA, -2L), class = c("tbl_df", 
"tbl", "data.frame")), structure(list(text = c("N", "N"), val = c(7, 
8)), .Names = c("text", "val"), row.names = c(NA, -2L), class = c("tbl_df", 
"tbl", "data.frame")))), class = c("tbl_df", "tbl", "data.frame"
), row.names = c(NA, -3L), .Names = c("ID", "data.x", "data.y"
))

Answer 1

Based on the expected output, we are extracting the 'val' columns in the data.frame from both 'data.x' and 'data.y' lists, concatenate it together (c) and get the max value

tbl %>% 
    mutate(col1 = map2_dbl(data.x, data.y, ~ max(c(.x$val, .y$val))))
# A tibble: 3 x 4     
#    ID    data.x           data.y            col1
#   <chr> <list>           <list>           <dbl>
#1 a     <tibble [2 x 2]> <tibble [2 x 2]>  7.00
#2 b     <tibble [2 x 2]> <tibble [2 x 2]>  8.00
#3 c     <tibble [2 x 2]> <tibble [2 x 2]>  8.00

For multiple columns, of 'data', pmap can be used

tbl %>%
    mutate(col1 = pmap_dbl(.[-1], ~ max(c(..1$val, ..2$val))))