Using Update Cursor in a double for loop in arcpy

Question

I have some problems using a double for loop and getting results written into a geodatabase.

I have got a shapefile with polygons (param_polygone) and I want to add data to it, which is stored in a table (table_datei). In the polygon shapefile I have got features with an unique ID, called "FID". In the table I have got fields, which store lenghts ("LENGTH") belonging to a polygon (src_FID). So often I there are several lengths stored for one polygon. For example there is one FID in the polygon shapefile, which is 0 and several entries or rows in the table, where the src_FID is 0 and where unique lengths are stored in the same row. I want to sum up all lengths of one src_FID and store it in the field "Nbr_Laenge", which belongs to the FID that is the same value as the src_FID.

Somehow only the first value gets sumed up in "Nbr_Laenge". All the others have the value 0 in the end of the execution.

Here is what I got so far:

table_cursor = arcpy.da.UpdateCursor(table_datei, ["src_FID", "LENGTH"])
geb_cursor = arcpy.da.UpdateCursor(param_polygone, ["FID", "Nbr_Laenge"])
 for geb_row in geb_cursor:
 for table_row in table_cursor:
 if table_row[0] == geb_row[0]:
 geb_row[1]=geb_row[1]+table_row[1]
 geb_cursor.updateRow(geb_row)

Something with the for-loops isn't working properly. I am new to python and working with ArcGIS.

• Bera: Or you could run Summary Statistics on table with ID as case field and SUM lenght. Then join this to the polygon shapefile and calculate
• Vince: You don’t need two UpdateCursor objects if one is query-only -- that's what SearchCursor objects are for.
• Michel Jerome: But I still don't know, why my code isn't working

Answer 1

You don't need a nested loop (and it's written in a wrong way because you will scan your table only once)

So. First of all read all the data from the table and sum it up.

data = {}
with arcpy.da.SearchCursor(table_datei, ["src_FID", "LENGTH"]) as sc:
 for row in sc:
 try:
 data[row[0]] = data[row[0]] + row[1]
 except KeyError:
 data[row[0]] = row[1]

This will give you a dictionary {data} with keys as your FID and values as a sum of all lenghts linked to this FID.

Next, start an update cursor and write it to your FC:

with arcpy.da.UpdateCursor(param_polygone, ["FID", "Nbr_Laenge"]) as uc:
 for row in uc:
 row[1] = data[row[0]]
 uc.updateRow(row)

That's it.

• Michel Jerome: that was very useful, thank you so much. I just had to add another try-except block into the second for-loop, because some FIDs are not covered in the table before. So again: thank you very much :)
• Serge Norin: You can use data.get(row[0], 0). This will give you a corresponding value if a key exists or 0 if not. This should work a little bit faster then try-except.