Skip to contents

Column-header cell specification

Source: R/col_header.R
col_cell.Rd

Convenience constructor for a single cell in a column-header row passed to rtftable() (via col_header =) or rtf_col_header().

Usage

col_cell(
  pos,
  label = "",
  align = NULL,
  bold = FALSE,
  italic = FALSE,
  underline = FALSE,
  border = NULL
)

Arguments

pos

Cell position, one of:

  • a numeric of length 1 (single column) or length 2 (c(start, end), inclusive) – start <= end required, values >= 1;

  • a character of length 1 (single column name) or length 2 (c(start_name, end_name)) referring to data columns by name; or

  • a selector function of the data column names – see col_key() – returning a logical, integer or character vector. It must match at least one column, and the matched columns must be adjacent. Name and selector resolution (and the start <= end check for named ranges) happens when the header is attached to a table.

label

Character scalar. Cell text; may be "".

align

Optional "left", "center", or "right". NULL (default) inherits the leftmost covered column's header_align.

bold, italic, underline

Logical. Default FALSE.

border

Optional rtf_border() applied to this header cell only, overriding the zone border and the automatic group-underline. This is how you fine-tune individual rules in a multi-row header – for example removing the bottom line under one spanning cell with border = rtf_border(bottom = "none"), or adding a thicker rule under one column. NULL (default) inherits the normal zone borders.

Value

A list of class "rtf_col_cell".

Details

Use pos = 1 for a single-column cell and pos = c(start, end) for a cell that spans several data columns. Positions are always relative to the underlying data columns, not to the previous header row.

pos may instead be column name(s) (character), resolved against the data columns when the header is attached to a table. This makes a spanning cell robust to column reordering and errors out on an unknown name – for example col_cell(c("g1", "g3"), "Drug A") spans from the column named g1 to the one named g3, and col_cell("total", "Total") targets a single named column.

pos may finally be a selector function – see col_key(). It is called with the data column names when the header is attached to a table and answers which columns the cell covers, so a spanning cell can be written in terms of what it labels rather than of column numbers:

col_cell(col_key("Placebo"), "Placebo")            # value before "____"
col_cell(function(nm) grepl("^Placebo____", nm), "Placebo")

See also

col_key() for selecting columns by a column-name segment.

Examples

col_cell(1, "Item")
#> <col_cell pos=1 label="Item">
col_cell(c(2, 5), "Treatment", align = "center", underline = TRUE)
#> <col_cell pos=2..5 label="Treatment", align=center [u]>

# Remove the group underline under just this spanning cell:
col_cell(c(2, 3), "Drug A",
         border = rtf_border(bottom = "none"))
#> <col_cell pos=2..3 label="Drug A">

# Select the columns to span by what they hold, not by where they sit:
col_cell(col_key("Drug A"), "Drug A")
#> <col_cell pos=col_key("Drug A") label="Drug A">