Covariate layer

[stemangiola]

No description provided.

[Copilot]

Pull Request Overview

This PR introduces a covariate layer in the header formatting, adds tests and example scripts, updates documentation, and extends utility functions for custom formatting.

• Adds a format_covariate_header function and integrates it into the tbl_format_header.SE_print_abstraction method.
• Introduces new unit tests under tests/testthat/ and a standalone debug script.
• Updates the README examples and session info, and extends pillar utilities and NAMESPACE imports.

Reviewed Changes

Copilot reviewed 8 out of 9 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
vignettes/Introduction.R	Deleted outdated introduction vignette
tests/testthat/test-header-formatting.R	Added snapshot tests for format_covariate_header
test_header_formatting.R	Added standalone debug/example script (should be reviewed)
README.md	Updated example output rows and sessionInfo details
R/tidyprint_1_utlis.R	Added format_covariate_header and extended formatting logic
R/print_methods.R	Adjusted separator row computation and attribute packing
R/pillar_utlis.R	Extended pillar___format_comment with strip.spaces
NAMESPACE	Added new importFrom entries for dplyr, purrr, etc.

Comments suppressed due to low confidence (3)

R/pillar_utlis.R:3

• This function now calls get_extent, but get_extent isn't imported in NAMESPACE; add importFrom(pillar, get_extent) so the function resolves correctly at runtime.

pillar___format_comment <- function (x, width, strip.spaces = TRUE)

NAMESPACE:19

• Missing imports for newly used functions pillar::style_subtle and pillar::get_extent; please add importFrom(pillar, style_subtle, get_extent) to NAMESPACE.

importFrom(cli,col_br_black)

[Copilot]

This standalone debug/example script appears to be for local testing; consider removing it or moving it under tests/ so it does not clutter the root directory.

[Copilot]

Avoid using unexplained magic numbers like + 3; consider computing the extra spacing based on actual separator characters or document the reason for this offset.

Suggested change

	total_covariate_width <- sum(covariate_widths) + length(covariate_widths) + 3 # To compensate the white spaces of the tibble
	# Compute extra spacing dynamically based on separator characters
	separator_spacing <- nchar("|") * 2 + 1 # Account for '|' at both ends and one space
	total_covariate_width <- sum(covariate_widths) + length(covariate_widths) + separator_spacing

[Copilot]

Calculating row-ID spacing using nchar(number_of_total_rows) - 3 is brittle; consider deriving the padding from the actual row ID width rather than hard-coding - 3.

Suggested change

	header_row <- c(paste(rep(" ", number_of_total_rows |> nchar() - 3), collapse = ""), header_row)
	row_id_width <- nchar(total_rows)
	header_row <- c(paste(rep(" ", row_id_width), collapse = ""), header_row)

[Copilot]

Remove leftover commented-out code snippets to keep the implementation clean and maintainable.

Suggested change

# , after = 1)

[stemangiola]

@zhanchen this PR I dont think works for arbitrary datasets. I tagged into an issue about the non existance of a variable that counts the table size in its renfered form.

This was not meant to be merged I think. Unless you solved that issue and tested extensively on many datasets.

[zhanchen]

@stemangiola
Yes, I fixed the covariate-layer print issue and committed to the branch before the merge.
Now the covariate-layer can align with the covariate column under varying widths and shall be removed when the width is too small and no covariate column was actually printed.



I will test it with datasets in the next couple days.