Can pillar_shaft.POSIXt() respect the digits.secs option? about pillar

[DavisVaughan]

@krlmlr, thanks for taking a look at this. Do you need to shift the width parameter too?

# with dev pillar and tibble
library(pillar)
library(rlang)

from <- as.POSIXct("14:03:55", format="%H:%M:%OS", tz="UTC")
to   <- as.POSIXct("14:04:00", format="%H:%M:%OS", tz="UTC")

options(digits.secs = 4)

ex <- tibble::tibble(datetime = seq(from, to, by = 0.01))

ex$col <- 100000

ex
#> # A tibble: 501 x 2
#>    datetime               col
#>    <dttm>               <dbl>
#>  1 2018-01-11 14:03:55.0000 100000
#>  2 2018-01-11 14:03:55.0099 100000
#>  3 2018-01-11 14:03:55.0199 100000
#>  4 2018-01-11 14:03:55.0299 100000
#>  5 2018-01-11 14:03:55.0399 100000
#>  6 2018-01-11 14:03:55.0499 100000
#>  7 2018-01-11 14:03:55.0599 100000
#>  8 2018-01-11 14:03:55.0699 100000
#>  9 2018-01-11 14:03:55.0799 100000
#> 10 2018-01-11 14:03:55.0899 100000
#> # ... with 491 more rows

# This should really be for POSIXt and would need the %||% from rlang
pillar_shaft.POSIXct <- function (x, ...) 
{
  width <- 19
  digits.secs <- getOption("digits.secs") %||% 0
  if (digits.secs) {
    width <- width + digits.secs + 1L
  } 
  
  date <- format(x, format = "%Y-%m-%d")
  time <- format(x, format = "%H:%M:%OS")
  datetime <- paste0(date, " ", style_subtle(time))
  datetime[is.na(x)] <- NA
  new_pillar_shaft_simple(datetime, width = width, align = "left")
}

ex
#> # A tibble: 501 x 2
#>    datetime                    col
#>    <dttm>                    <dbl>
#>  1 2018-01-11 14:03:55.0000 100000
#>  2 2018-01-11 14:03:55.0099 100000
#>  3 2018-01-11 14:03:55.0199 100000
#>  4 2018-01-11 14:03:55.0299 100000
#>  5 2018-01-11 14:03:55.0399 100000
#>  6 2018-01-11 14:03:55.0499 100000
#>  7 2018-01-11 14:03:55.0599 100000
#>  8 2018-01-11 14:03:55.0699 100000
#>  9 2018-01-11 14:03:55.0799 100000
#> 10 2018-01-11 14:03:55.0899 100000
#> # ... with 491 more rows

Created on 2018-01-11 by the reprex package (v0.1.1.9000).

Rationale for my solution is that it solves three cases:

• User hasn't set the option, should default to NULL which is changed to 0
• User has set the option to 0
• User has set the option to >0

[krlmlr]

@DavisVaughan: Would you mind opening a new issue?

[github-actions]

This old thread has been automatically locked. If you think you have found something related to this, please open a new issue and link to this old issue if necessary.