Formatting of numbers, when they are to appear in messages. Displays only significant digits in a "nice way" and adds commas to separate thousands. It does much less than the format function, but also a bit more though.
fsignif(x, s = 2, r = 0, commas = TRUE)
signif_plusAn object of class function of length 1.
A numeric vector.
The number of significant digits to be displayed. Defaults to 2. All digits not in the decimal are always shown.
For large values, the number of digits after the decimals to be displayed (beyond the number of significant digits). Defaults to 0. It is useful to suggest that a number is not an integer.
Whether or not to add commas to separate thousands. Defaults to TRUE.
It returns a character vector of the same length as the input.
x = rnorm(1e5)
x[sample(1e5, 1e4, TRUE)] = NA
# Dumb function telling the number of NA values
tell_na = function(x) message("x contains ", fsignif(sum(is.na(x))), " NA values.")
tell_na(x)
#> x contains 9,473 NA values.
# Some differences with signif:
show_diff = function(x, d = 2) cat("signif(x, ", d, ") -> ", signif(x, d),
" vs fsignif(x, ", d, ") -> ",
fsignif(x, d), "\n", sep = "")
# Main difference is for large numbers
show_diff(95123.125)
#> signif(x, 2) -> 95000 vs fsignif(x, 2) -> 95,123
show_diff(95123.125, 7)
#> signif(x, 7) -> 95123.12 vs fsignif(x, 7) -> 95,123.12
# Identical for small numbers
show_diff(pi / 500)
#> signif(x, 2) -> 0.0063 vs fsignif(x, 2) -> 0.0063