Full_Name: Wolfgang Huber Version: R version 2.4.0 Under development (unstable) (2006-07-09 r38523) OS: i686-pc-linux-gnu Submission from: (NULL) (62.253.128.15) There appears to be unintentional behaviour in the size of the bounding box for PDF graphics produced with the current R2.4, compared to 2.3.1. I posted two PDF files, both produced with the same R script below, but different versions of R: http://www.ebi.ac.uk/~huber/pub The code is here: pdf(file=sprintf("test-%s.pdf", version$"svn rev"), width=4, height=4) par(mai=par("mai")[rep(1:2,2)]) x = seq(0, 10*pi, length=3000) plot(x*cbind(sin(x),cos(x)), type="l", col="mistyrose", lwd=3) sink(textConnection("s", "w")) options(width=40) print(sessionInfo(), width=30) sink() text(-30, 30-4*seq(along=s), s, adj=c(0,0.5), cex=0.7) dev.off() and one of the differences between the resulting PDF files is: test-38323.pdf: /MediaBox [0 0 288 288] test-38523.pdf: /MediaBox [0 0 595 841] The former corresponds to 4x4 inches, the latter to 21cm x 29.67cm.
Unexpected difference in Bounding Box of PDF graphics between 2.3.1p and 2.4 (PR#9070)
5 messages · Wolfgang Huber, Hin-Tak Leung, Brian Ripley
huber at ebi.ac.uk wrote:
Full_Name: Wolfgang Huber Version: R version 2.4.0 Under development (unstable) (2006-07-09 r38523) OS: i686-pc-linux-gnu Submission from: (NULL) (62.253.128.15) There appears to be unintentional behaviour in the size of the bounding box for PDF graphics produced with the current R2.4, compared to 2.3.1. I posted two PDF files, both produced with the same R script below, but different versions of R: http://www.ebi.ac.uk/~huber/pub The code is here: pdf(file=sprintf("test-%s.pdf", version$"svn rev"), width=4, height=4) par(mai=par("mai")[rep(1:2,2)]) x = seq(0, 10*pi, length=3000) plot(x*cbind(sin(x),cos(x)), type="l", col="mistyrose", lwd=3) sink(textConnection("s", "w")) options(width=40) print(sessionInfo(), width=30) sink() text(-30, 30-4*seq(along=s), s, adj=c(0,0.5), cex=0.7) dev.off() and one of the differences between the resulting PDF files is: test-38323.pdf: /MediaBox [0 0 288 288] test-38523.pdf: /MediaBox [0 0 595 841] The former corresponds to 4x4 inches, the latter to 21cm x 29.67cm.
[0 0 595 842] is A4 size. Probably a default taken from something.
This is already fixed. There is no such thing as '2.4', nor does PDF have a bounding box! The issue was that for a short while the paper size of pdf() was defaulting to "default", the same as postscript() (and this was intentional). It is now the previous default of "special", and setting paper="special" always worked.
On Mon, 10 Jul 2006, Hin-Tak Leung wrote:
huber at ebi.ac.uk wrote:
Full_Name: Wolfgang Huber Version: R version 2.4.0 Under development (unstable) (2006-07-09 r38523) OS: i686-pc-linux-gnu Submission from: (NULL) (62.253.128.15) There appears to be unintentional behaviour in the size of the bounding box for PDF graphics produced with the current R2.4, compared to 2.3.1. I posted two PDF files, both produced with the same R script below, but different versions of R: http://www.ebi.ac.uk/~huber/pub The code is here: pdf(file=sprintf("test-%s.pdf", version$"svn rev"), width=4, height=4) par(mai=par("mai")[rep(1:2,2)]) x = seq(0, 10*pi, length=3000) plot(x*cbind(sin(x),cos(x)), type="l", col="mistyrose", lwd=3) sink(textConnection("s", "w")) options(width=40) print(sessionInfo(), width=30) sink() text(-30, 30-4*seq(along=s), s, adj=c(0,0.5), cex=0.7) dev.off() and one of the differences between the resulting PDF files is: test-38323.pdf: /MediaBox [0 0 288 288] test-38523.pdf: /MediaBox [0 0 595 841] The former corresponds to 4x4 inches, the latter to 21cm x 29.67cm.
[0 0 595 842] is A4 size. Probably a default taken from something.
______________________________________________ R-devel at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
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
Dear Prof Ripley,
There is no such thing as '2.4',
Can we forward this information to the author of "sessionInfo()", "version", and R's welcome message, this is where I took that from. nor does PDF have a bounding box! Yup, sorry for my imprecise language, I don't understand much of PDF, just noted the different appearance of the plots. Thanks for the quick fix (and all the other contributions), this really helps. Best wishes Wolfgang Huber ------------------------------------------------------------------ Wolfgang Huber EBI/EMBL Cambridge UK http://www.ebi.ac.uk/huber
On Mon, 10 Jul 2006, Wolfgang Huber wrote:
Dear Prof Ripley,
There is no such thing as '2.4',
Can we forward this information to the author of "sessionInfo()", "version", and R's welcome message, this is where I took that from.
I get
R version 2.4.0 Under development (unstable) (2006-07-09 r38523)
^^^^^^^^^^^^^^^^^^^^^^^
and the R posting guide makes the same point: it is 2.3.0, not 2.3.
nor does PDF have a bounding box! Yup, sorry for my imprecise language, I don't understand much of PDF, just noted the different appearance of the plots. Thanks for the quick fix (and all the other contributions), this really helps. Best wishes Wolfgang Huber ------------------------------------------------------------------ Wolfgang Huber EBI/EMBL Cambridge UK http://www.ebi.ac.uk/huber
______________________________________________ R-devel at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
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