Skip to contents

Reshape a parallel frame to match a snaked table

Source: R/gt_snake_align.R
gt_snake_align.Rd

Lays a data frame out in side-by-side blocks the same way gt_snake() lays out the table, so a frame you computed against the long data still lines up after the snake. It is the escape hatch for a precomputed grid you do not want to rebuild as a condition: a logical mask for gt_highlight_cells(), a frame of per-cell colors or text, anything the same shape as the snaked body.

Usage

gt_snake_align(x, n_cols = 2, rows_per_col = NULL, fill = NA)

Arguments

x

A data frame or matrix with one row per row of the un-snaked table.

n_cols

Integer. The number of blocks, matching the gt_snake() call. Defaults to 2.

rows_per_col

Integer. Rows in each block. Given this, n_cols is worked out from the data instead, exactly as gt_snake() does it. Defaults to NULL.

fill

What to put in the trailing cells when the rows do not divide evenly. Defaults to NA, which gt_highlight_cells() reads as no match.

Value

Returns a data frame of the reshaped blocks.

Details

Most of the time you do not need this. Body-cell styling applied before gt_snake() is carried through the reshape on its own, so coloring the long grid and then snaking already lands the fills in the right place. Reach for this only when the grid lives outside the table's styles.

Pass the same n_cols (or rows_per_col) you passed to gt_snake(). Each block's columns are suffixed with its block number, _1, _2, and so on, to match the names gt_snake() gives the table. The result has ceiling(nrow(x) / n_cols) rows and ncol(x) * n_cols columns, so it is the shape gt_highlight_cells() expects for the block of snaked columns.

See also

gt_snake(), which reshapes the table itself and carries body styling through.

Examples

if (FALSE) { # \dontrun{
library(gt)

wide <- head(mtcars[c("mpg", "hp", "wt", "qsec")], 8)

# a logical grid the same shape as the data, built before the snake
mask <- wide > 20

# after the snake, each column carries its block number as a suffix
snaked <- unlist(lapply(1:2, function(i) paste0(names(wide), "_", i)))

wide %>%
  gt() %>%
  gt_snake(n_cols = 2) %>%
  gt_highlight_cells(tidyselect::all_of(snaked),
                     condition = gt_snake_align(mask, n_cols = 2),
                     fill = "#EDBD68")
} # }