spread_first_col outside of as_flextable

[r2evans]

(I think this is a FR, but a bug.)

I have an existing flextable chain that produces a good table. From there, I have been manually (in Word) shifting the first column to a row-group header: add a row above, merge the cells, copy the value in, then eventually delete the first column.

I think this is what as_flextable(spread_first_col=TRUE) is supposed to enable. However, I have a regular flextable() on the data. I have no need to tabulate or summarize it, that's already been done, so I don't need (I think) summarizor() or tabulator(), therefore I cannot use as_flextable().

In the end, I'm hoping for something like this using spread_first_col=TRUE, and I can't figure out how to do it with a vanilla frame.

library(dplyr)
library(flextable)
relocate(mtcars, cyl, .before = everything()) |>
  arrange(cyl) |>
  flextable() |>
  save_as_docx(path="mytable.docx")

I think in the interim I could hand-jam using merge_h_range() to insert rows, merge the columns, etc. Is there a better way to do that?

[davidgohel]

Hello

The most straightforward way to achieve this today is with as_grouped_data(), which is to be called before the call to flextable():

library(dplyr)
library(flextable)

relocate(mtcars, cyl, .before = everything()) |>
  arrange(cyl) |>
  as_grouped_data(groups = "cyl") |>
  as_flextable()

This pulls cyl out of the columns and inserts group-label rows that span the full table width (this is what spread_first_col does).

I understand this means replacing flextable() with as_grouped_data() |> as_flextable() in your chain but there are chances everything work the same way ! (since the only structural change is that the grouped column becomes row labels instead of a regular column).

If it's not too late and this doesn't work for your use case, feel free to share a reproducible example and I'll try to help.

[r2evans]

Yep, that will work for my use-case, thank you so much!