[R] Sweave and X11 on OSX 10.3
Friedrich.Leisch@ci.tuwien.ac.at
Friedrich.Leisch at ci.tuwien.ac.at
Wed Jan 7 16:52:02 CET 2004
>>>>> On Tue, 6 Jan 2004 23:44:42 -0700 (MST),
>>>>> Kieran Healy (KH) wrote:
> Hi -
> I'm running R 1.8.1 (compiled from source) on Mac OS X 10.3 (Panther). I
> find that, if Apple's X11 application is not running, Sweave gives an
> error when it wants to create a pdf or eps figure. E.g., in the package's
> own example-1.Snw file a boxplot is created at chunk 2:
> <<fig=TRUE,echo=FALSE>>=
> boxplot(Ozone ~ Month, data = airquality)
> @
> Normally this will create a pdf and an eps version of the boxplot. I get:
>> Sweave("example-1.Snw")
> Writing to file example-1.tex
> Processing code chunks ...
> 1 : echo term verbatim
> 2 : term verbatim eps pdf
> Warning message:
> unable to open connection to X11 display`'
> Error: chunk 2
> Error in X11(display, width, height, pointsize, gamma, colortype,
> maxcubesize, :
> unable to start device X11
> If I start the X11 application, Sweave runs as normal but displays any
> figures on an x11() device in addition to creating the eps / pdf files.
> I don't remember this happening before and experience teaches me to assume
> the issue is with my own setup rather than Sewave. Maybe there's a
> problem with the DISPLAY variable, or something that I did while upgrading
> to Panther -- but I can't see what it is. Has anyone else had this
> issue?
Your R session has obviously x11() as the default plotting device
... e.g.,
plot(1:10)
at the prompt should also give an error, if no graphics device is
open. I don't know too much about MacOS X, but having the DISPLAY
variable set without having an X11 server surely means asking for
trouble on Linux, and I guess MacOS X is not different ...
--
-------------------------------------------------------------------
Friedrich Leisch
Institut für Statistik Tel: (+43 1) 58801 10715
Technische Universität Wien Fax: (+43 1) 58801 10798
Wiedner Hauptstraße 8-10/1071 Friedrich.Leisch at ci.tuwien.ac.at
A-1040 Wien, Austria http://www.ci.tuwien.ac.at/~leisch
More information about the R-help
mailing list