On Sep 7, 2009, at 4:37 AM, Prof Brian Ripley wrote:
My guess is quartz() and that this is nothing to do with R but with copying in Mac OS X from a quartz() window.
Saving to a file from the quartz window also produces a PDF that is viewable but not printable by Preview, and Acrobat too. I suppose it could be the printer driver, though that means that R is guilty of putting out some pretty exotic Postscript. I'm hoping someone else sees this or doesn't so I can pin it down. Paul
par(mfrow=c(2,2)) y<-1:20 plot(y, pch="+") plot(y, pch="+") plot(y, pch="+") plot(y, pch="+") produces a pdf that is viewable in Preview (copied from the clipboard) but fails to print. pch default or ="." or =3 prints. Unfortunately pch=3 doesn't fix the mfrow=c(5,6) graph that I'm actually trying to print, so the bug lies somewhere deeper.
Why not use pdf()? That works for me (but then I don't have your printer drivers).
OS X 10.4.11 R 2.9.2 GUI 1.29 Tiger build 32-bit (5464) Laserjet 5MP Paul McGuire
-- 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