[R] How to make nice tiny sized figures on graphic devices producing scalable vector output?

[Mikhail Titov]

Hello!

 

Usually whenever I want a tiny plot, I just create it as is (or even large) and 
then downscale it in the end application like LaTeX of MS Word. However, all 
these graphic devices like postscript, pdf, win.metafile retain physical sizes, 
so it would be natural if I can just insert graphics as is provided those have 
proper physical sizes embedded.

 

The question is what is the best method to create plots in R with their final 
physical sizes? I would like to create a good-looking figure by starting with 
let’s say win.metafile(“some.emf”, 3.35, 2) .

 

Of course all defaults will produce something unreadable, so I have to scale 
down everything with cex at least, and probably with lwd changes. I’ve tried 
something like below. But dashed (and all other) lines are still too thick, 
plotting symbols are undistinguishable etc. And in general it looks like a mess 
to override all possible values.

 

Is there a better way to downscale whatever is being plotted on a device?

 

windowsFonts(Arial=windowsFont("TT Arial"))

cex <- .3

win.metafile("some.emf", 3.35, 2)

    data <- data.frame(y=rep(c("a","b","c"), each=10), x=runif(30))

    bwplot(y~x, data,

           par.settings = modifyList(

           simpleTheme(cex=cex, lwd=cex),

           c(

#               theme.nopadding,

               axis.line=list(lwd=cex))),

           panel=function(...) {

               panel.bwplot(..., pch="|")

#               panel.mean(..., pch=16)

           },

           cex=cex,

           ylab=list(label=expression(bold("my x label")), cex=cex),

           xlab=list(label=expression(bold("my y label")), cex=cex),

           scales=list(fontfamily="Arial", cex=cex),

           horizontal=TRUE)

dev.off()

 

Mikhail

 


        [[alternative HTML version deleted]]


______________________________________________
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.