[R] Plots using POSIX

[Prof Brian Ripley]

On Thu, 26 Jun 2003, Duncan Murdoch wrote:

> On Thu, 26 Jun 2003 07:59:00 -0500, Shawn Way <sway at tanox.com> wrote :
> 
> >
> >Is there a reason that the bottom axis changes color when POSIX data is used
> >in plot function?
> 
> It's the old problem of too much of ... being passed onwards.  Here's
> the current definition:
> 
> plot.POSIXct <- function (x, y, xlab = "", xaxt = par("xaxt"), ...) 
> {
>     axisInt <- function(x, main, sub, xlab, ylab, ...) axis.POSIXct(1,
>         x, ...)
>     plot.default(x, y, xaxt = "n", xlab = xlab, ...)
>     if (xaxt != "n") 
>         axisInt(x, ...)
> }
> 
> The "col" argument is being passed to axisInt, but it should have been
> intercepted.  Here's one way to intercept it:
> 
> plot.POSIXct <- function (x, y, xlab = "", xaxt = par("xaxt"), col =
> par("col"), ...) 
> {
>     axisInt <- function(x, main, sub, xlab, ylab, ...) axis.POSIXct(1,
>         x, ...)
>     plot.default(x, y, xaxt = "n", xlab = xlab, col = col, ...)
>     if (xaxt != "n") 
>         axisInt(x, ...)
> }
> 
> However, this would still mess up if "lty" or "lwd" were specified;
> are there others?

Just add those that should not be passed on to the defn of axisInt,
rather than clutter the argument list.

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595