Getting RunTime Error: Object: CreateObject cannot create geometry from inputs when trying to add a polyline using an insert cursor

Question

The objective for this code is to take a csv that contains rhino observations and create a polyline shapefile for the data. The shapefile needs to have the paths the rhinos take based on observations (which is an x and y field in the csv) and the rhinos name must be stored in another field.

I have no idea what is going on, can't figure out the source of this error. I am not sure if the problem is in logic I used to create the dictionary, or if it with the code that follows the creation of the insert cursor.

It seems that the logic for the dictionary is working fine, I added print statements to make sure the rhino data is added properly, it follows the csv file as it should. Once you print the completed dictionary however, the rhino arrays are messed up. You can tell because some rhinos have more points than they should and some have less.

here is the code:

import arcpy
import csv
from arcpy import env
env.overwriteOutput = True

rhino_obsv = open(r"C:\Users\cvric\Desktop\GEOG485\project_4\RhinoObservations.csv")    # open csv containing the rhino observations for reading
sr = arcpy.SpatialReference(4326) # create a spatial reference object that is WGS1984
rhino_obsv_sf = arcpy.CreateFeatureclass_management(r"C:\Users\cvric\Desktop\GEOG485\project_4", "rhino_obsv.shp", "POLYLINE", spatial_reference = sr) # create empty polyline shapefile to store rhino data
arcpy.AddField_management(rhino_obsv_sf,"Name",field_type = "TEXT")

csv_reader = csv.reader(rhino_obsv) # create csv reader, delimiter is comma
header = csv_reader.next() # move reader to the header
name_index = header.index("Rhino") # find index pos. for rhino names
lat_index = header.index("X")   # find index pos. for latitude coordinate
lon_index = header.index("Y")   # index pos. for longitude coordinate

rhino_dict = {}
for row in csv_reader:
    lat = row[lat_index]
    lon = row[lon_index]
    name = row[name_index]
    if name in rhino_dict:
        sighting = arcpy.Point(lon,lat)
        rhino_track.add(sighting)
        #print("sighting added to " + name)

    else:
        #print("array created for " + name)
        rhino_track = arcpy.Array()
        sighting = arcpy.Point(lon,lat)
        rhino_track.add(sighting)
        rhino_dict[name] = [rhino_track]


with arcpy.da.InsertCursor(rhino_obsv_sf, ("SHAPEP@","Name")) as cursor: # use insert cursor to add rhink data to blank shapefile
    for key in rhino_dict:  # iterate through keys in dicitionary
        polyline = arcpy.Polyline(rhino_dict[key],sr) # use the array(value) that corresponds to each rhino(key) to create a polyline
        name = key # name of rhino is the key in the dictionary
        cursor.insertRow((polyline,name)) # use insertRow method to add polyline and name to the SHAPE and Name fields
del cursor

Answer 1

When the rhino track is already in the dictionary you need to pull out its array before trying to update. Also don't add the array to the dictionary as a list.

if name in rhino_dict:
    sighting = arcpy.Point(lon,lat)
    rhino_track = rhino_dict [name] #### get correct track
    rhino_track.add(sighting)
    #rhino_dict[name] = [rhino_track] ###adding as a list - incorrect
    rhino_dict[name] = rhino_track ### adding as array

else:
    rhino_track = arcpy.Array()
    sighting = arcpy.Point(lon,lat)
    rhino_track.add(sighting)
    rhino_dict[name] = rhino_track #not as list