maechler at stat.math.ethz.ch
Mon Aug 8 10:20:16 CEST 2011
please excuse a short "top-reply":
As I see you are using an operating system (instead of Win..),
can you try in the shell
echo 'X11.options()$type' | R --vanilla --slave
and does that really *not* report
(and BTW: Your subject had tolower("LL") instead of "11" )
>>>>> Rolf Turner <r.turner at auckland.ac.nz>
>>>>> on Mon, 8 Aug 2011 12:49:44 +1200 writes:
> This question seemed to me to be more appropriate for
> r-devel than for r-help. My apologies if this is not the
> Recently I installed ``cairo'' on my lap-top so that I
> could make use of the (newish) polypath() function, with
> on-screen graphics. (The polypath() function does not
> work with X11(type="Xlib").)
> The installation went smoothly, X11(type="cairo") works
> just fine, and polypath() works just fine.
> However the default "type" for X11() remains "Xlib" rather
> than "cairo".
> The help for X11.options() says (in respect of "type"):
> Default "cairo" where available and reliable,
> otherwise "Xlib".
> Now "cairo" is definitely available:
> cairo TRUE
> and moreover it works (just fine!) when I do
> X11(type="cairo") explicitly.
> I know that I can set the default to be "cairo" in my
> .Rprofile, but I am curious as to why "cairo" does not
> become the default automatically.
> Is it the case that "cairo" is ``not reliable'' under my
> system? It ***seems*** to be reliable.
> R version 2.13.1 Patched (2011-07-17 r56404) Platform:
> i686-pc-linux-gnu (32-bit)
> locale:  LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C 
> LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8 
> LC_MONETARY=C LC_MESSAGES=en_US.UTF-8 
> LC_PAPER=en_US.UTF-8 LC_NAME=C  LC_ADDRESS=C
> LC_TELEPHONE=C  LC_MEASUREMENT=en_US.UTF-8
> attached base packages:  datasets utils stats graphics
> grDevices methods base
> other attached packages:  misc_0.0-13 gtools_2.6.2
> spatstat_1.23-1 deldir_0.0-15  mgcv_1.7-6
> fortunes_1.4-2 MASS_7.3-13
> loaded via a namespace (and not attached):  grid_2.13.1
> lattice_0.19-30 Matrix_0.999375-50 nlme_3.1-101
> Rolf Turner
> R-devel at r-project.org mailing list
More information about the R-devel