delete_dups

'''
This script deletes identical rows in the mls point file.

Currently tested with 'apn' column which seemingly corresponds to unique listings.
The column 'apn' was chosen because it provided both a handle on unique mls listing
while also being a integer field making sorting easy. Sorting was necessary to identify duplicates

Additionally, many of the duplicate records contained reference to the same parcel but were found 7-10 meters apart.
This made the 'Delete Identical' tool (ArcGIS Advanced License and up ) inadequate here.

Once duplicates were removed, the MLS layer could be efficiently loaded in ArcMap.

'''
import arcpy
import time
workspace = arcpy.env.workspace = r'C:\Users\noah\Desktop\gundog\chandler\data.gdb'

# city variable
city = 'chandler'

def deleteDuplicates():
    fc = 'mls'
    fields = ['apn']

    counter = 0
    matchCount = 0
    sql_clause = (None, 'ORDER BY apn ASC')

    with arcpy.da.UpdateCursor(fc,'apn','','','', sql_clause) as cursor:

        for row in cursor:

            n = row[0]
            if counter > 0 and oldN == n:
                expression = "\"apn\" = {0}" .format(n)
                if expression:
                    #print 'We have a match!'
                    cursor.deleteRow()
                    matchCount = matchCount + 1

            oldN = n
            counter = counter + 1

    print matchCount


deleteDuplicates()