Perfect, thanks very much. I couldn't see the xlim/ylim solution which is really neat, and clearly better than my suggestion.
Cheers, Mike. On Wed, Apr 17, 2013 at 6:15 PM, Prof Brian Ripley <rip...@stats.ox.ac.uk>wrote: > On 12/04/2013 12:27, Michael Sumner wrote: > >> Hello, I would like to suggest the following change to image.default in >> src\library\graphics\R\image.**R: >> >> 98c98 >> < plot(NA, NA, xlim = xlim, ylim = ylim, type = "n", xaxs = xaxs, >> --- >> >>> plot(x[1], y[1], xlim = xlim, ylim = ylim, type = "n", xaxs = >>> xaxs, >>> >> >> > I think xlim, ylim is a better idea: this relies on range() preserving the > class. > > Done now (in R-devel pro tem). > > > This provides all the support of axis.POSIXt that plot.default gives, >> currently the default new plot in image.default uses NAs and the axis >> class >> is ignored. This is nice for plotting time series data stored in a matrix. >> >> Here's a dummy example that shows usage with my change: >> >> data(volcano) >> x <- list(x = Sys.time() + seq(1, 1e6, length = nrow(volcano)), y = >> 1:ncol(volcano), z = volcano) >> >> ## date-time formatting on the x-axis >> image(x) >> >> ## date-time formatting on the y-axis >> image(x$y, x$x, t(x$z)) >> >> Without the change we get the very large underlying numeric values for the >> times on the axis. >> >> We can still override the default axis to do a workaround: : >> image(x, axes = FALSE) >> axis.POSIXct(x$x, side = 1) >> >> Cheers, Mike. >> >> >> >> > > -- > Brian D. Ripley, rip...@stats.ox.ac.uk > Professor of Applied Statistics, > http://www.stats.ox.ac.uk/~**ripley/<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 > -- Michael Sumner Hobart, Australia e-mail: mdsum...@gmail.com [[alternative HTML version deleted]] ______________________________________________ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
