[R] how to control the overall shape of a figure?

Oliver fwang2 at gmail.com
Tue Feb 17 19:10:18 CET 2009


Thanks very much, exactly what I need.

Oliver

On Feb 16, 10:36 pm, Marc Schwartz <marc_schwa... at comcast.net> wrote:
> on 02/16/2009 07:51 PM Oliver wrote:
>
>
>
> > hi,
>
> > I am a R beginner. One thing I notice is that when do graphing is,
>
> > if I want to draw two figures in a row such as this:
>
> > par(mfrow(1, 2))
> > plot(...)
> > plot(...)
>
> > Each figure inside will be rectangle instead of the familiar square
> > shape.
> > Though you can drag the edge the window to resize it. I would have
> > prefer this can be done automatically ... also when I do the pdf
> > export for example. Is this possible?
>
> > The thing is if you do par (mfrow(2,2)), then you got all 4 figures in
> > perfect square shape ... but one row two figures are pretty common to
> > me.
>
> > thanks for help
>
> > Oliver
>
> The overall shape of the plot region is controlled by par("pty"), which
> is set to "m" by default, for 'maximal' plotting region.
>
> Set this to "s" to create a square plot region. For example:
>
> par(mfrow = c(1, 2), pty = "s")
> plot(1)
> plot(1)
>
> In order to do this with a PDF file, ensure that the overall plot
> dimensions are in the proper relative aspect ratio. For example:
>
> pdf(file = "SquarePlots.pdf", height = 4, width = 8)
> par(mfrow = c(1, 2), pty= "s")
> plot(1)
> plot(1)
> dev.off()
>
> This will create 2 square plots, side-by-side, within an overall plot
> size of 4x8. You can further adjust the plot margins and other
> characteristics as may be required.
>
> See ?par and ?Devices for more information regarding these
> customizations and options to set size arguments for specific devices.
>
> HTH,
>
> Marc Schwartz
>
> ______________________________________________
> R-h... at r-project.org mailing listhttps://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guidehttp://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.




More information about the R-help mailing list