[HELP] tar_map and write_xl

[oTheAnalyst]

Hello,

I this snippet of a targets script down below. I've try various method but even when the pipeline is successfully build I can't actually create excel files from tar_map.

  tar_target(
    sub_office_results,
    tar_map(
      suboffice_list,
       tar_target(
        results2,
        map_staff_list(data = transformed_data, soffice = .x)
      ),
       tar_target(
        adding_summary_tables,
        addstable(dlist = results2)
      ),
       tar_target(
        save_file,
        write_xlsx_file(data, path)

if you have any ideas please let me know!

[ginolhac]

Without a reprex it is difficult to help, but here a sketch of a static pipeline (tar_map()) using {writexl}.

_targets.R:

library(targets)
library(tarchetypes)

values <- data.frame(
  val = 1:3,
  names = c("A", "B", "C")
)

tar_map(
  values = values,
  tar_target(results2, paste(val, "bib")),
  tar_target(summary, data.frame(x = paste(results2, "sum"))),
  tar_target(save_file, writexl::write_xlsx(summary, paste0(tar_name(), ".xlsx")))
)

DAG after tar_make():

The folder content to show that the 3 expected excel were written:

> fs::dir_tree(recurse = 0)
.
├── _targets/
├── _targets.R
├── save_file_1_A.xlsx
├── save_file_2_B.xlsx
└── save_file_3_C.xlsx

and content of one of the excel to show it worked out

[oTheAnalyst]

awesome, I'm gonna test out this framework. I was also curious is there a way to use functions from {purrr} in a targets list? or we mostly limited to using tar_map() to iteratively apply function to a list or dataframe?

[ginolhac]

What do you have mind? The name map is the same for iteration but don't use purrr for running targets machinery.
William has foreseen most if not all needs so i doubt you will miss anything

[oTheAnalyst]

What do you have mind? The name map is the same for iteration but don't use purrr for running targets machinery. William has foreseen most if not all needs so i doubt you will miss anything

ah okay. That's fair, I think I will try and figure out the tar_map() function and seeing what I can with that.

Earlier you said you need the reprex? what is that exactly?

[ginolhac]

It is a minimal reproducible example. Not dependent on external datasets, private objects etc. Details here https://reprex.tidyverse.org/