Here's an example.  I usually receive daily updates, so I will post  
this directly to the list.
Thanks much: cannot test much more today, as I must prepare the  
poster and the talk
and finish the report by Saturday PM (a common story, no doubt).  I  
will try to imitate
the reply format.

I will mail a PNG of the example directly to Steve if possible.

Notes from a Word 2004 session which I have captured into a PNG.
Printing gives similar results.  As I mentioned, Word is useful in  
capturing
exploratory sessions which are not suitable for finished code and a  
source(file.R)

 > quartz()
 > plot(c(1,2,3,4,5),(c(1,2,3,4,5))^2)
Copy/Paste to Preview, then Copy/Paste to Word (2004)
Also: Copy/Paste to Preview, then Copy/Paste to GraphicConverter

{ALL BLACK FIELD}


Insert Picture to Word (2005) from PDF saved from Preview:
{CORRECT DISPLAY }


 > library(fields)
 > quartz()
 > x<- 1:10; y<- 1:15; z<- outer( x,y,"+")
 > image.plot(x,y,z)
 > points( 5:12, 5:12, pch="X", cex=3)    # example from image.plot page
Copy/Paste to Preview, then Copy/Paste to Word (2004)
Similar results: Copy/Paste to Preview, then Copy/Paste to  
GraphicConverter v 5.5.2

{ALL BLACK O    UTSIDE AXES, CORRECT INSIDE. X's show}

Similar results with image(x,y,z) (I know itÂ’s not same usage).


 > Thanks, Steve; this works. albeit in a new way.  Some notes for the
 > > user follow.
 > >
 > > Simple-documentation copy-paste:
 > > I must start the device this way to avoid difficulties (see  
quartz())
 > > quartz(bg="white")
 > >
 > > Then I may copy and open up the file in Preview;
 > > from there, I may copy again and paste into Word or into
 > > Graphicconverter.  (These probably accepted picts earlier, from
 > > what I could infer from the FAQ).
 > >
 > > If one does not set the background, then the Preview version looks
 > > fine, but the image pasted elsewhere has black lettering (usually)
 > > on a black background.
 >
 > That is likely a bug in whatever program you're pasting it in as it
 > doesn't handle transparency properly (by default the plot is painted
 > on a transparent background so it's easier to adapt it to your style
 > in Keynote etc.). I was checking with Adobe CS2 and it works without
 > problems, so I suspect your application is simply too old to handle
 > that. You don't need to use Preview, you can paste straight from
 > Quartz without problems.


 > Hence quartz(bg="white")
 >
 > Higher quality pdf's:
 > I assume I should do a quartz.save to get a high quality image
 > if I think that the image may head to publication.


 > > Hence quartz(bg="white")
 > >
 > > Higher quality pdf's:
 > > I assume I should do a quartz.save to get a high quality image
 > > if I think that the image may head to publication.
 >
 > There is no such thing as "higher quality PDFs". All PDFs contain the
 > plot as-is (whether you use copy/paste, quartz PDF output or pdf
 > device). What you probably mean is high quality PNG (bitmap) images.
 > Those can be created with quartz.save by specifying the target
 > resolution such as:
 > quartz.save("foo.png",dpi=300)
 > However, that is not what you want to use for publications.
 >
 > Note that MS products cannot handle PDF (definitely not on the  
Windows
 > side), so you cannot use PDF there (which is really stupid, but  
that's
 > another story). That's why you may prefer higher-resolution PNG if  
the
 > document is to be shared with Windows people.

I am referring to Sections 7.2 and 7.3 of the "R for Mac OS X FAQ"
which implies a difference in the quality of the PDF's presented.
Presumably the same fundamentals are allowed, but different graphics
engines perform with different skill.  For example, I find that
gs frequently gives less-desirable images.

Microsoft Word 2004 for Mac seems to import PDF's OK, and so does  
PowerPoint 2004. I just tested these.
They seem to give identical results, although I believe
that Word is inclined to translate to Windows Bitmap Format at the
slightest provocation (usually a bad idea).  I have had some
difficulty with PowerPoint and pdf's in previous years.

Thanks,  Bob C
        [[alternative HTML version deleted]]

_______________________________________________
R-SIG-Mac mailing list
[email protected]
https://stat.ethz.ch/mailman/listinfo/r-sig-mac

Reply via email to