R dplyr: rowwise + mutate (+glue) - how to get/refer row content?

Question

The simple example of input data:

dataset <- data.frame("part1" = c("a", "b", "c"),
                       "part2" = c("x", "y", "z"),
                       "caption" = c("{part1} {part2}",
                                     "{part2} {part1}",
                                     "{part2} {part1} {part2}"),
                       stringsAsFactors = F)

Expected results:

# A tibble: 3 x 3
  part1 part2 caption
  <chr> <chr> <chr>  
1 a     x     a x    
2 b     y     y b    
3 c     z     z c z  

The below code doesn't work, because . refers to the whole dataset, instead of data of the whole row content:

dataset %>%
  rowwise() %>%
  mutate("caption" =
           glue::glue_data(., caption)
         )

Question: how to pass row (all) content to glue?

The code that works (row "content" declared explicitly) is not what I've been looking for, because there are more columns used in caption "pattern" in my data set, thus I would like to avoid to declare it manually, just pass the whole row content.

dataset %>%
  rowwise() %>%
  mutate("caption" =
           glue::glue_data(
             list("part1" =  part1,
                  "part2" = part2)
             , caption)
  )

Answer 1

It works simply as:

dataset %>% rowwise %>% mutate(r=as.character(glue(caption)))
#Source: local data frame [3 x 4]
#Groups: <by row>

## A tibble: 3 x 4
#  part1 part2 caption                 r    
#  <chr> <chr> <chr>                   <chr>
#1 a     x     {part1} {part2}         a x  
#2 b     y     {part2} {part1}         y b  
#3 c     z     {part2} {part1} {part2} z c z

NOTE : I added the as.characteronly to avoid warnings that seem to be an issue with rowwise(Vectorizing 'glue' elements may not preserve their attributes)