[R] Cairo SVG backend

Prof Brian Ripley ripley at stats.ox.ac.uk
Sat Oct 27 08:32:44 CEST 2007


You need recent enough cairo libraries in your OS.  Your OS appears to 
date from mid 2006, so likely you need a later cairo: my system is using 
cairo 1.4.10 and found all the backends.

Your kernel is quite old: does your OS have kernel and cairo updates you 
have not applied?

On Fri, 26 Oct 2007, Sam McClatchie wrote:

> System: 2.6.0
> Linux kernel 2.6.15 Ubuntu dapper
> R version 2.5.1
> ESS 5.2.11 on Emacs 21.4.1
>
> Colleagues
>
> I am still struggling to produce SVG file output in R.
>
> I initially started with RSvgDevice package. I produced a simple graphic
> from an example in the documentation and it imported into Inkscape fine
> and could be edited. However, my application involves plotting
> bathymetry and coastlines with the PBSmapping, and the plots contain a
> lot of data. Although I can still produce a file, and the screen graphic
> looks fine, Inkscape will not load the 1.8 MB file.
>
> I wasn't sure if the problem was with Inkscape or with the SVG file. I
> thought I'd try the Cairo package to see if RSvgDevice was failing
> because of the file size.  I've install Cairo and used update.packages()
> to get the 1.3-5 version which has the SVG device driver. However, the
> Cairo.capabilities() function returns true for png, jpeg, and x11, but
> false for
> svg, ps, pdf and tiff. The error message returned from the call to
> Cairo_svg() is "Cairo compiled without svg device". So my question is:
> How do you to compile Cairo with the svg device? Do you have to do this
> from source with some --enabled switch? The install.packages() and
> update.packages() does not seem to do it?
>
> Best fishes
>
> Sam
>

-- 
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



More information about the R-help mailing list