Hfrhyu

I'm trying to update a column based on the value in the next row, but cursor.next() is giving unexpected behavior. When I identify the next_row using cursor.next(), it is resetting my row in cursor! The result is my loop goes over every other row, not every row. What am I missing? (Code below is just to show cursor behavior; code for row update for column has been stripped.)

Code

import arcpy
temp1 = r'mypathmypointfile.shp'
with arcpy.da.UpdateCursor(temp1, ['OID@']) as cursor:
 for row in cursor:
 try:
 row_next = cursor.next()
 print 'current_oid=', row[0], 'next_oid=', row_next[0]
 except StopIteration:
 print 'no rows'
 except:
 print 'something else went wrong'

Actual Prints

current_oid= 0 next_oid= 1
current_oid= 2 next_oid= 3
current_oid= 4 next_oid= 5
current_oid= 6 next_oid= 7
current_oid= 8 next_oid= 9

Expected Prints

current_oid= 0 next_oid= 1
current_oid= 1 next_oid= 2
current_oid= 2 next_oid= 3
current_oid= 3 next_oid= 4
current_oid= 4 next_oid= 5

UPDATE

Purpose of this is to only keep rows that are the first or last point in the group.

Example attributes:

oid group id
0 1 1
1 1 2
2 1 3
3 2 1
4 2 2
5 3 1
6 3 2
7 3 3
8 3 4
9 3 5

Desired attributes:

group id
1 1
1 3
2 1
2 2
3 1
3 5

Original code, that didn't work because cursor.next() skips rows:

with arcpy.da.UpdateCursor(temp1, ['OID@', 'group', 'id']) as cursor:
 for row in cursor:
 group_current = row[1]
 id_current = row[2]
 if id_current == 1:
 check = "keep"
 else:
 try:
 row_next = cursor.next()
 group_next = row_next[1]
 id_next = row_next[2]
 if id_next == 1:
 check = "keep"
 else:
 check = "remove"
 except StopIteration:
 check = "keep"
 except:
 print 'something else went wrong'
 if check != "keep":
 cursor.deleteRow()

Yes, this is expected as it's how python generators work, once a value is consumed, it's gone. If your feature class isn't toooo big, pull all the rows into a list, then use enumerate and indexing to access the next row. E.g.

cursor = ((i,) for i in range(10)) # Fake a cursor as I'm not at a PC with ArcGIS
rows = list(cursor) # Exhaust the generator
for i, row in enumerate(rows):
 try:
 row_next = rows[i+1]
 print(row[0],row_next[0])
 except IndexError:
 print('no rows')

Output:

0 1
1 2
2 3
3 4
4 5
5 6
6 7
7 8
8 9
no rows

Or... turn it around, grab the first row before the loop then refer to the previous row instead of the next row:

prev_row = next(cursor)
for row in cursor:
 print(prev_row[0], row[0])
 prev_row = row

Output:

0 1
1 2
2 3
3 4
4 5
5 6
6 7
7 8
8 9

Thanks to Vince's comments and user2856's answer/comments, the following worked. Elapsed time = 30.17 seconds for 52,000 row, which seems slow.

temp1 = r'mypathmypointfile.shp'
import datetime
now = datetime.datetime.now()

cursor = arcpy.da.SearchCursor(temp1, ['OID@', 'group', 'id'])
rows = list(cursor)
dict_ids = 
for i, row in enumerate(rows):
 oid = row[0]
 group_current = row[1]
 id_current = row[2]
 if id_current == 1:
 check = "keep"
 else:
 try:
 row_next = rows[i + 1]
 id_next = row_next[2]
 if vid_next == 1:
 check = "keep"
 else:
 check = "remove"
 except IndexError:
 check = "keep"
 except:
 print 'something else went wrong'
 dict_ids.update([(oid, check)])
del i, row, check
with arcpy.da.UpdateCursor(temp1, ['OID@']) as cursor:
 for row in cursor:
 oid = row[0]
 if oid in dict_ids:
 check = dict_ids.get(oid)
 else:
 print 'error with dictionary'
 sys.exit()
 if check != "keep":
 cursor.deleteRow()

later = datetime.datetime.now()
elapsed = later - now
print elapsed