[R] [OT] 'gv' and fractional points

[(Ted Harding)]

On 15-Jun-07 20:41:36, Paul Murrell wrote:
> Hi
> 
> If I understand correctly, this is something that the 'grImport'
> package might be very useful for. You can import the PostScript
> image into R, which means that you can draw the image, but you
> also have the locations of everything that is drawn as numeric
> values so you should be able (probably after a bit of transformation)
> to extract the values to very good accuracy.  If you can provide
> me with an example file, I'd be happy to play around and see if
> I could get this to work.
> 
> Paul

As an example of this, try Figure 2 on page 3 of

INphoRM 4: Presenting performance indicators: alternative approaches

  http://www.erpho.org.uk/download.aspx?urlid=7518&urlt=1

This comes back as an 8-page PDF file

  "INphoRM 4 final amended.pdf"


I used Acrobat Reader to print Page 3 to a PS file, which I will
send you separately, Paul.

(Note that this one has no scale in the X-axis, so there's no
knowing what the absolute numbers are; but the relative positions
in "points" across the page were of interest anyway).

In fact, in this case I was able to extract the axact coordinates
in "points" by grepping through the PS file (having sussed out
the procedural definitions, then testing for which were points on
the graph and which were "furniture"), but that is really tedious!

So I'd be very interested to see how you get on.

If that's too heavy for illustrative use of your method, then
try Figure 1 on Page 2, in which case "your task" is to get
the Y-coordinates of the tops of the bars, of the upper and
lower ends of the rror-vars, and the Y-levels of the two
confidence limits.

Since I know the numbers as embedded in the PS file, I can
check your accuracy!

Thanks, and best wishes,
Ted.

--------------------------------------------------------------------
E-Mail: (Ted Harding) <ted.harding at nessie.mcc.ac.uk>
Fax-to-email: +44 (0)870 094 0861
Date: 16-Jun-07                                       Time: 00:25:00
------------------------------ XFMail ------------------------------