Python point neighbor processing ideas?

[ClintonCooper1]

I know a bit about python scripting, but am having some trouble figuring out where to get started on creating a script.

I need to loop down through a dataset (about 10,000 records), and select all points within a certain distance of the current point in the loop (it doesn't have to be a selection, but it must be within a spatial distance of the current point I just figured a selection would be good because it would limit the number of points that are queried).  After I create the selection of points,I need to count the number of points that satisfy 2 criteria:  1) have a value in another field 1 that is the same as the value in of the current point's field 1; 2) have values in field 2 that are greater than the current point's field 2.  I also have to do anther count that has the same above criteria of 1), but instead of greater than, I need to count those that are less than the current point's value in field 2.   I then need to print those counts (both greater than and less than) as well as the current point's  id  so I can go back and put them in another table for analysis.

I guess I would kind of be looking for an outline or some code that might be able to help get me started writing it.  Thanks!

(Curtis Price edited title [originally: "Python Script help"] (4/29/2015))

[curtvprice]

You may want to look at the Spatial Analyst tool Point Density.

[curtvprice]

I would try the Point Distance tool and then do processing on the output table. This can make a very big table but if you limit it with a search distance that keeps it from getting too big.

I've found Python sets and dictionaries to be very helpful for this kind of processing -- so you may run Point Distance, Add Join, load data into python objects using an arcpy.da.SearchCursor loop.

I 'm thinking Select by Location sounds straighforward -- but it would be a prohibitively slow approach if you have a significant number of points to process.

[DarrenWiens2 · Accepted solution]

You do most of your task using Spatial Join on the point file itself, as both Target and Join Features, ONE_TO_MANY, WITHIN your distance. In the output, select and delete points where TARGET_FID = JOIN_FID (that is the point joined to itself). You can then select the matching rows which meet your criteria, and count using Summary Statistics, grouping on TARGET_FID.

[ClintonCooper1]

I like this solution a lot, thank you!!!