arcpy.da.UpdateCursor to concatenate two strings

[ClintonCooper1]

I am trying to figure out how to concatenate two strings using the arcpy.da.UpdateCursor method (newbie to python scripting).  I have tried many trial and error attempts in the basic code from resource center, and have yet to get one that will work.  Here is the code I am working with:

import arcpy  fc = " C:\Users\cc1\Desktop\Metro_Bus_Stops.shp" fields = ('City','TRIAL')   with arcpy.da.UpdateCursor(fc, fields) as cursor:  in cursor:         row[0] = row[1]         cursor.updateRow(row)  

This code works fine for updating a new field with the current data form another field.  But I cannot get it to concatenate two fields together for the new third field.  Also, I have seen some literature with the ''.join() method, that I am interested in.  I am working with over 7 million records, and every once of performance I can gain is needed.

[MathewCoyle · Accepted solution]

Here's a quick example. Takes field0 and field1 and writes them to field2.

fields = ['field0', 'field1', 'field2'] cursor = arcpy.da.UpdateCursor(fc, fields) for row in cursor:     row[2] = '{0}{1}'.format(row[0], row[1])     cursor.updateRow(row) del cursor

[ClintonCooper1]

do I change format to join?  Thank you for the quick reply!!

[MathewCoyle]

That depends on what you want your output to be. What reason would you have for using join in this context?

[ClintonCooper1]

I am trying to concatenate two fields, and the literature I have found says that .join is the solution to concatenate the two fields.  When I run the script you provided with my information, I am getting this error:

Runtime error
Traceback (most recent call last):
  File "<string>", line 8, in <module>
IndexError: list assignment index out of range

for this code:

import arcpy

fc = " C:\Users\cc1\Desktop\Metro_Bus_Stops.shp"

fields = ['City', 'County', 'TRIAL']
cursor = arcpy.da.UpdateCursor(fc, fields)
for row in cursor:
    row[3] = '{0}{1}'.format(row[0], row[1])
    cursor.updateRow(row)
del cursor

Clinton

[MathewCoyle]

Ah yes I mixed up the index for the 3rd field. Should be this.

    row[2] = '{0}{1}'.format(row[0], row[1])

And what literature are you referring to?

[ClintonCooper1]

It seems that the .join is an older method.  In your estimation, will this be the fastest method to concatenate strings (aka faster than field calculator)?   Or is  there another method that will be much faster?  As I have said in previous posts, I need to do this calc on 7+ million records...through field calculator this takes 24 hours....  Thanks again for all your help!!!

[MathewCoyle]

Assuming I/O isn't your issue this method should take less than 2 minutes for 7 million records (takes around 10 seconds for 500k records for me). The only faster way I can think of is it do this through SQL directly in a database.

[ClintonCooper1]

ran through it last night, and it took around 10 to 15 minutes for a shapefile.  I then ran it again on a table (hoping to skip the geocoding process), and it took around 30 to 45 minutes.  Can't complain as this is about 10 to 20 times faster than field calculator.  I may eventually look into doing it through sql, but for now this will suffice!  Thanks again!!!

[MathewCoyle]

It really shouldn't even be taking that long, unless your strings are really long or you are dealing with a slow computer. Did you try it in a file geodatabase?