Re: [R] resolution (dpi) problem

From: Ted Harding <>
Date: Tue, 29 Apr 2008 13:56:01 +0100 (BST)

On 29-Apr-08 12:09:39, Marc Schwartz wrote:
> Chris Walker wrote:

>> To all those who helped with this problem (Ted, Jon, Marc and

> Prof. Ripley).
>> When I submitted the postscript files to the journal, I had
>> given them the extension ".ps" even though they were (of course)
>> encapsulated postscript. I decided to simply rename the files
>> with the .eps extension and resubmit.
>> This time the images were accepted without question. It would
>> appear to be a case of someone assuming that I had submitted
>> postscript without bothering to check that it was encapsulated.

> It looks like the default extension on the 'Save as postscript'
> menu is '.ps'. That's both from the 'File' menu and the right
> click menu over the graphic device.
> I wonder if it might make sense to change that to '.eps', since
> the output is encapsulated postscript. Might help to avoid this
> in the future.
> Marc

I think it does make sense. Generally, someone will save a plot to PostScript for the purpose of including it in a document, and ".eps" will certainly help hapless editors to recognise it for what it is.

?postscript tells me:

postscript(file = ifelse(onefile, "", ""), [etc.]

so even if you invoke postscript() from the R command line you will get a default ".ps" extension. But here I think the danger is less, since I think most people will give an explicit (informative) filename when they save a plot, and can then choose to give it ".eps".

Perhaps the defaults here should read

postscript(file = ifelse(onefile, "", "Rplot%03d.eps"), [etc]

Certainly for Windows users, the suggested change could be wise, since Windows (for reasons best known to someone who should have known better) conceals the extension from the user, unless the user submits to cruel and unusual requirements such as (a) right-clicking on the file icon; (b) clicking on "Properties"!

Possibly, on some Windows systems (I'm not a Windows user), hovering the mouse over the file pops up a little box which says "PS file" if the extension is ".ps", and "EPS file" of the extension is ".eps"; and conceivably something like this may be the source of Chris's publisher's misapprehension.

Certainly, the filename should not matter when it comes to actually importing the EPS into a document. For example, in 'groff' I can import as EPS a file which has any name and any extension whatever, just so long as it has a valid "%%BoundingBox:" line in it.

But maybe, again, some software will only look at the extension in the first instance, and sulk if that is not ".eps".

Just some thoughts ...
(And felicitations to Chris for getting his article through the door).


E-Mail: (Ted Harding) <> Fax-to-email: +44 (0)870 094 0861
Date: 29-Apr-08                                       Time: 13:55:58
------------------------------ XFMail ------------------------------

______________________________________________ mailing list PLEASE do read the posting guide and provide commented, minimal, self-contained, reproducible code. Received on Tue 29 Apr 2008 - 13:07:59 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Thu 01 May 2008 - 14:30:34 GMT.

Mailing list information is available at Please read the posting guide before posting to the list.

list of date sections of archive