There was some discussion a while back of an R pgf driver for latex. Is anyone working on that?
On Mon, Mar 3, 2008 at 2:09 PM, Greg Snow <[EMAIL PROTECTED]> wrote: > The gp.plot function has a type option that can either be 'p' for points > or 'l' for lines, the function is not that complicated, you could easily > add additional options. > > I don't know why your code is not creating the pdf file, you could try > plotting to the screen first (don't change the output and terminal), > make sure that the plot is there, then change the output and terminal > and do a replot to see if that works. > > Given that, I agree with the others that gnuplot is probably not the > best tool for this. Others have mentioned how to use cmr and other > fonts within R, or I mentioned the tool for converting a .eps file to a > .pgf file that when input into LaTeX will use the current fonts. > > Gnuplot is a fine program for what it was designed for, but as you want > more and more sophisticated statistical plots it will become more and > more frustrating to do them using gnuplot. Gnuplot was not really > designed as a statistical tool (more a mathematical one). > > I once heard that the difference between a mathematician and a > statistician is that one knows the difference between a variable and a > constant and the other one doesn't. I think the difference between > mathematicians and statisticians can be seen in the response to the mean > value theorem: the mathematician sees it and says "cool, I can compute > averages using integrals", the statistician sees it and says "cool, I > can compute integrals using averages". > > Hope this helps, > > -- > Gregory (Greg) L. Snow Ph.D. > Statistical Data Center > Intermountain Healthcare > [EMAIL PROTECTED] > (801) 408-8111 > > > > > -----Original Message----- > > From: Louise Hoffman [mailto:[EMAIL PROTECTED] > > Sent: Sunday, March 02, 2008 8:48 AM > > To: Greg Snow > > Cc: Gabor Csardi; r-help@r-project.org > > Subject: Re: [R] Make plots with GNUplot. Have anyone tried that? > > > > > > There is a very basic interface between R and gnuplot in the > > > TeachingDemos package. Look at the help for gp.plot. > > [snip] > > > > This looks mighty interesting =) > > > > Is it possible to plot "with lines" (gnuplot syntex) so all > > the data points are connected? > > > > Also is it possible to make it write the output to a pdf > > file? (I using the gnuplot cvs which have pdfcairo support) > > > > When I try this, it doesn't write the pdf file. it works with > > gnuplot cvs. > > > > library(TeachingDemos) > > fuelData<-read.table('fuel.csv',header=TRUE, sep=',') > > attach(fuelData) > > gp.open(where='/usr/local/bin/gnuplot') > > gp.send('set terminal pdfcairo font 'cmr10' size 8cm,4.6cm') > > gp.send('set output '../figures/q1-raw-data-gp.pdf'') > > gp.send('unset key') gp.send('set xlabel 'rtime'') > > gp.send('set ylabel 'FPI'') gp.send('set xrange [1979:2005]') > > gp.plot(rtime,fpi) > > gp.send('unset output') > > gp.close() > > > > ______________________________________________ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
