[R-sig-Geo] interacting with plots

[Greg Snow]

Determining if a point is in a polygon is not as simple a task as
might be assumed (I can see that it is in there, why can't the
computer?).  So the computer has to do this check for all (or maybe
just some) 12,438 features, so it is not surprising if it is a little
slow.  The people who created and maintain the sp package would be
better able to address issues of how to speed it up (I am not one of
those who get that credit).  For this task it may be better to use a
tool outside of R, qgis has been mentioned and there may be others.

On Thu, Aug 16, 2012 at 12:47 PM, Frazier, Tyler James
<tyler.j.frazier at tu-berlin.de> wrote:
> Hi Greg,
>
> The example works fine, but running the
>
> tmp <- SpatialPoints(locator())
>
> takes a really long time to execute on my map of 12,438 features each with 3
> fields.
>
> Is there anyway to speed it up?
>
> BTW -- I'm running on MacOS Lion with R 2.14.0.
>
> Also, the output plots to Quartz, which doesn't seem to have a pan function
> and the zoom function is very very limited.
>
> Thanks!
> Ty
>
> On Aug 16, 2012, at 7:30 PM, Greg Snow wrote:
>
> My example used locator(1) to get one click, but if you just use
> locator() (without the 1) then you can click in as many places as you
> want, then it should show you the info for all of them).  If you want
> to click and see the results, then click again and see the new
> results, etc. then you could put my code into a while loop.  Panning
> and zooming using the mouse would be a bit more difficult, but you
> might get a similar effect by setting the limits when plotting (and
> you can get limits to set using the locator function again).
>
> On Thu, Aug 16, 2012 at 11:09 AM, Frazier, Tyler James
> <tyler.j.frazier at tu-berlin.de> wrote:
>
> Hi Greg,
>
>
> That works but I can only chose one polygon and have its attributes
>
> identified.  I want to plot my map, then be able to pan and zoom, and chose
>
> several polygons to have their attributes identified (one at a time would be
>
> fine).
>
>
> Should I be exploring maptools?
>
>
> Thanks,
>
> Ty
>
>
> On Aug 16, 2012, at 6:33 PM, Greg Snow wrote:
>
>
> You could try something like:
>
>
> library(rgdal)
>
> library(sp)
>
> scot_BNG <- readOGR(dsn=dsn, layer="scot_BNG")
>
> plot(scot_BNG)
>
> tmp <- SpatialPoints(locator(1))
>
> proj4string(tmp) <- proj4string(scot_BNG)
>
> over(tmp, scot_BNG)
>
> # or
>
> tmp2 <- over(scot_BNG, tmp)
>
> plot( scot_BNG[!is.na(tmp2),], add=TRUE, col='green')
>
>
> Does that do what you want?
>
>
> On Thu, Aug 16, 2012 at 9:05 AM, Frazier, Tyler James
>
> <tyler.j.frazier at tu-berlin.de> wrote:
>
>
> Hello,
>
>
>
> I am new to this group and am still working to figure out many things.  I
>
> have a kind of basic question, regarding using R and interacting with plots.
>
> I have been exploring using the spplot in the sp package as a means to
>
> interact with plots, but haven't figured out how to use the function to
>
> identify specific attributes associated with an individual polygons.  My
>
> code is really basic, I simply
>
>
>
> ea <- readOGR(".","ghana_eas_ft")
>
>
> plot(ea)
>
>
>
> is it possible to plot the map in a way where I can use the cursor in R to
>
> chose and identify attributes associated with a polygon.  Also here is the
>
> output of str(ea) which follows with a lot of information after the
>
> @polygons which I assume is part of the sp class structure.
>
>
>
> Formal class 'SpatialPolygonsDataFrame' [package "sp"] with 5 slots
>
>
> ..@ data       :'data.frame': 12438 obs. of  3 variables:
>
>
> .. ..$ EA_NOS   : Factor w/ 1133 levels "001-005, 340-342",..: 1110 986 985
>
> 976 982 983 978 741 974 975 ...
>
>
> .. ..$ REG_CODE : Factor w/ 10 levels "01","02","03",..: 1 1 1 1 1 1 1 1 1
>
> 1 ...
>
>
> .. ..$ DIST_CODE: Factor w/ 18 levels "01","02","03",..: 9 9 9 9 9 9 9 9 9
>
> 9 ...
>
>
> ..@ polygons   :List of 12438
>
>
>
> I am wondering if there are newer packages which provide the identify
>
> function for polygons.  Currently I am referencing the Applied Spatial Data
>
> Analysis with R text.
>
>
>
> Thanks!
>
>
> Ty
>
>
>
>
>
>
>
>
>
>
>
>
> --
> Gregory (Greg) L. Snow Ph.D.
> 538280 at gmail.com
>
>
> ---------------------------------------------------
> Dr. Tyler Frazier, AICP
> Senior Research Scientist
> Department of Transportation System Planning and Telematics
> Technische Universität Berlin
> Sekretariat SG 12
> Salzufer 17-19
> 10587 Berlin
> fon:  +49 30 314 22418
> fax:  +49 30 314 26269
> www.vsp.tu-berlin.de
> www.matsim.org
>
>
>
>
>



-- 
Gregory (Greg) L. Snow Ph.D.
538280 at gmail.com