This is horribly hacky and probably belies my limited UNIX shell fu, but it works for me on a Fedora 17 box with the pdfjam package installed (not an R package, but from the YUM repos)
pdf("pdf1.pdf")
plot(1:10)
dev.off()
pdf("| pdfjoin --outfile \"pdf2.pdf\" && pdfjoin pdf1.pdf pdf2.pdf --outfile pdf1.pdf && rm pdf2.pdf")
plot(10:1)
dev.off()
The output in R is:
> pdf("| pdfjoin --outfile \"pdf2.pdf\" && pdfjoin pdf1.pdf pdf2.pdf --outfile pdf1.pdf && rm pdf2.pdf")## && pdfunite joined.pdf tmp.pdf joined.pdf && rm tmp.pdf")
> plot(10:1)
> dev.off()
----
pdfjam: This is pdfjam version 2.08.
pdfjam: Reading any site-wide or user-specific defaults...
(none found)
pdfjam: No PDF/JPG/PNG source specified: input is from stdin.
pdfjam: Effective call for this run of pdfjam:
/usr/bin/pdfjam --fitpaper 'true' --rotateoversize 'true' --suffix joined --outfile pdf2.pdf -- /dev/stdin -
pdfjam: Calling pdflatex...
pdfjam: Finished. Output was to 'pdf2.pdf'.
----
pdfjam: This is pdfjam version 2.08.
pdfjam: Reading any site-wide or user-specific defaults...
(none found)
pdfjam: Effective call for this run of pdfjam:
/usr/bin/pdfjam --fitpaper 'true' --rotateoversize 'true' --suffix joined --outfile pdf1.pdf -- pdf1.pdf - pdf2.pdf -
pdfjam: Calling pdflatex...
pdfjam: Finished. Output was to 'pdf1.pdf'.
null device
1
Basically, pdfjoin
will take input from stdin
if it is the only input file so I pipe the output from pdf()
to the pdfjoin
program and specify the output file using the --outfile
argument. Then using &&
is join the original pdf1.pdf
with the pdf2.pdf
just created, specifying that the output PDF is pdf1.pdf
, the name of the original PDF.