How to make a loop for select by attribute?

Question

I am trying to pull out a sample out of a geodatabase file. I have 100800 sample and I need to select 28 records, basically every 3600, I am going to start with FID=0 and go from there. Unfortunately I haven't been able to make it work. this is what I have done so far. anyone can give me a hint, it's gonna be a huge help.

import arcpy, sys, os
Input=sys.argv[1]
field=sys.argv[2]
counts=sys.argv[3]
output=sys.argv[4]
def deleteMe(fc):
 if arcpy.Exists(fc):
 arcpy.Delete_management(fc)
 return fc
cursor =arcpy.da.SearchCursor(Input,field)
for row in cursor:
 a=row[0]
 while row:
 x=a+3600
 arcpy.MakeFeatureLayer_management(Input, deleteMe("lyr"))
 arcpy.SelectLayerByAttribute_management ("lyr", "NEW_SELECTION",a)
#arcpy.SelectLayerByAttribute_management ("lyr", "ADD_TO_SELECTION",x)
arcpy.CopyFeatures_management ("lyr", deleteMe(output))

• Michael Stimson: Skip through with an incrementing variable, when you hit your target value keep that OID (OIDlist.append(str(row[0])) and reset the counter to 0. Then use the query string "OBJECTID IN (" + ",".join(OIDlist) + ")".. it's that or (if your OIDs are unique and ascending) OBECTID MOD 3600 = 0
• mr.adam: yes, just what I was thinking. OP mentioned a geodatabase so I wrote my answer for non-consecutive OBJECTIDs. I didn't think of using MOD in the query though, if it was a shapefile that would be great.
• Michael Stimson: I know @mr.adam, the mod(FID,3600) would do it for sure in a shapefile as FIDs are compressed on save. Note the change of syntax for the mod operator in shapefiles. If you think that's fun try to select out a random 10% of features... now that was fun!

Answer 1

How about just making a list of the OBJECTID's you want and making it into a query?

## start with list of all objectids
all_oids = [row[0] for row in arcpy.da.SearchCursor(Input,"OBJECTID")]
## make list of every 3600th objectid
oids = []
for index, oid in enumerate(range(len(all_oids))):
 if index % 3600 == 0:
 oids.append(oid)
## make sql query
qry = '"OBJECTID" IN ({0})'.format(".join(oids))
## deal with the datasets
arcpy.MakeFeatureLayer_management(Input, deleteMe("lyr"))
arcpy.SelectLayerByAttribute_management ("lyr", "NEW_SELECTION", qry)
arcpy.CopyFeatures_management ("lyr", deleteMe(output))

Enumerate is a super useful function. It iterates a list and also returns the index of each item. You're using it here to run through the list of OBJECTIDs and if the index is a factor of 3600 the oid appended to a list. Then use that list to create a query that you can apply to the feature layer.

• Michael Stimson: Where does i come into it? That's basically what I said but you need to set i=0 to start and then i+=1 in the loop.
• mr.adam: oops, was from an earlier version of the answer, will update...