[Rd] PGF Device

Prof Brian Ripley ripley at stats.ox.ac.uk
Wed Jan 31 14:49:22 CET 2007


On Wed, 31 Jan 2007, Duncan Murdoch wrote:

> On 1/30/2007 8:50 PM, jtxx000 wrote:
>> Greetings all,
>> PGF is a package for LaTeX which works with both ps
>> and pdf output without any nasty hacks like pictex.
>> Is there any technical reason why there could not be a
>> PGF graphic device for R?  If not, I'm going to try to
>> throw one together.  Any tips for getting started?
>
> The existing drivers are in the grDevices package.  I would start with
> one of those (the picTeX driver would probably be most similar) and go
> from there.  Alternatively, start from devNULL.c and fill in the blanks.

Please don't start from pictex (it is only half implemented, including no 
real font metrics).  If you need Type1 fonts, I suggest you start from the 
postscript/PDF driver.

> Someone else may be able to offer advice about whether you should be
> targetting this to be your own package, or for inclusion in the R
> source.  In the past it's been impossible to create a driver in a
> package, and there's been talk about adding enough to the API to allow
> this, but I don't know whether that's in place yet.

There are several devices in packages, and this has been possible since at 
least R 1.4.0 (from 2001).  There used to be gtkDevice (still on CRAN, but 
old Gtk), GNOMEdevice (never finished) and an SVG driver 
(http://www.darkridge.com/~jake/RSvg/), and there are now cairoDevice on 
CRAN, the quartz device in the R.app front-end, and others.

Paul Murrell used to have a list of known graphics devices, but I was 
unable to locate it quickly.

I've written several devices, including pdf(), xfig() and versions of 
png() etc.  The hard part is getting text and font metrics working: also 
nowadays you need to be prepared to handle UTF-8 text and CJK locales, 
which many graphic systems are not up to speed on (including postscript, 
pdf and latex).

-- 
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-devel mailing list