tibbles nested inside tibbles cause errors in repr_text

[michaelquinn32]

Hi team!

Here's a reprex.

library(tibble)
library(repr)

tbl_obj <- tibble::tibble(
  col1 = tibble::tibble(
    inner_col1 = 1
  )
)

repr::repr_text(tbl_obj)

Error in vapply(part, function(col) {: values must be length 1,
 but FUN(X[[1]]) result is length 4
Traceback:

1. repr::repr_text(tbl_obj)
2. repr_text.data.frame(tbl_obj)
3. ellip_limit_arr(obj, rows, cols)
4. arr_parts_format(parts)
5. structure(lapply(parts, arr_part_format), omit = attr(parts, 
 .     "omit"))
6. lapply(parts, arr_part_format)
7. FUN(X[[i]], ...)
8. vapply(part, function(col) {
 .     if (is.matrix(col)) 
 .         apply(apply(col, 2L, format), 1L, paste, collapse = ", ")
 .     else format(col)
 . }, character(nrow(part)))

The error is being thrown by this line of code: https://github.com/IRkernel/repr/blob/f4780e5ab58054d69ef77a0f61b239bebd106de5/R/repr_matrix_df.r#L68

IIUC, this is due to the fact that format(tbl()) can return a character vector that's longer than the number of rows in the data frame. For example:

str(format(tibble::tibble(a = 1)))
#> chr [1:4] "\033[38;5;246m# A tibble: 1 × 1\033[39m" "      a" ...

[MichaelChirico]

I think this is closed by #166 (just not marked as such by the PR)

[flying-sheep]

yup, printing this works fine now!