Using ArcPy cursor to concatenate row values?

Question

I am ArcGIS 10 user and I want to create ArcPy script for simple concatenate row values like this tool Concatenate Row Value that works with ArcGIS 10.

I have three fields in my shapefile id,field1,field2.

Here the image from my shapefile .dbf after sort by id.

If you see the image you can see the id not have unique values.

I want to create the layer with unique id values but I don't want to loose some values from field2 (is text). For field2 I need to use for next conditions.

In Excel if I use some condition like this:

=IF(AND(A3=A2;C3=C2);D2;IF(AND(A3=A2;C3<>C2);D2;IF(AND(A3=A2;C3<>C2);D2&","&C3;C3)))

Then I take new field to like this:

That works in Excel, I want to do with Python or ArcPy or some tool from ArcGIS.

In QGIS, it is easy for me to do with this plugin that does dissolve with stats.

I try to follow this code:

updateFields = ["id", "field2", "field3"]

with arcpy.UpdateCursor("featurelayer", updateFields) as cursor:
    for row in cursor:
        if row[0]+1 == row[0] and row[2]+1==row[2]:
            row[3] = row[2]
        elif row[0]+1 == row[0] and row[2]+1!=row[2]:
            row[3] = row[2]
        elif row[0]+1 == row[0] and row[2]+1!=row[2]:
            row[3] = row[2]+','+row[2]+1
        cursor.updateRow(row)

That looks correct but any time to try to run it I take this error:

 Cursor AttributeError: __exit__

How do I fix the problem?

Answer 1

If I understand your question correctly, you want to concatenate three fields into a new field. You can use the field calculator say you create a new field called "conc" for your concatenated strings.

eg. field1=dog field 2=cat, field3=moose

Right click on the "conc" field and go to field calculator, change to python:

Then write:

!field1! + !field2! + !field3!

To concatenate all three fields into your new conc field. Giving you:

dogcatmoose

See the ESRI blog for more concatenation.

Based on the new edits:

There's a tool already called "Concatenate Row Values" that looks to do exactly what you want - which takes identical row values (like your id field) and concatenates the values in a given row into a new one.

Answer 2

You'll want to iterate with a cursor and then make use of a dictionary to store past values found.

#empty dictionary to hold past values
di = {}

#cursor to iterate rows
with arcpy.da.UpdateCursor ("featurelayer", ["id", "field2", "final_field"]) as curs:
    for Id, val, finalVal in curs:
        #update dictionary
        try: di [Id] += [val]
        except: di [Id] = [val]

        #get all values and create string
        vals = di [Id]
        valStr = ",".join (map (str, vals))

        #update row
        row = (Id, val, valStr)
        curs.updateRow (row)

For 10.0 and before:

#empty dictionary to hold past values
di = {}

#cursor to iterate rows
curs = arcpy.UpdateCursor ("featurelayer")
for row in curs:
    #update dictionary
    Id = row.getValue ("id")
    val = row.getValue ("field2")
    try: di [Id] += [val]
    except: di [Id] = [val]

    #get all values and create string
    vals = di [Id]
    valStr = ",".join (map (str, vals))

    #update row
    row.setValue ("final_field", valStr)
    curs.updateRow (row)

del curs