Python Programming for ArcGIS: Part II

Python Programming for Arcgis 2
Daniel Sheehan
dsheehan@mit.edu, gishelp@mit.edu
9:30AM-12:30PM
January 15, 2015
This class was originally developed by David
Quinn and taught by David and Daniel in IAP
2010 and 2011.

Goals for the workshop (from yesterday)
• Learning enough Python to
– Access Arcgis commands through Python
– Access individual records from attribute table
– Acccess individual geometries for use in
geoprocessing
• Develop the ability to record and document
your geoprocessing

Slides available at …
• http://web.mit.edu/dsheehan/www/
dataForPythonIAP2015.zip (same data as
yesterday)
• http://web.mit.edu/dsheehan/www/
PythonProgrammingforArcgis2.pdf

Outline
• More data types
• Python functions
• Accessing Attribute Tables
• Select by attributes and location

Lists = []
An ordered set of elements enclosed in square brackets.
Zero based (first element is accessed by typing list[0]).
# This is a list []
animals = [ 'dog ' , ' cat ' , 'horse ' , ' seal ' ]
print animals[3]
animals.append(‘owl’)
Lists are mutable

Tuples = ()
Zero base (first element of a non-empty tuple is
always tuple(0)
# This is a tuple
Countries = (‘Costa Rica’, ‘India’, ‘Abu Dhabi’)
# You cannot change a tuple (immutable)

String processing – slicing a string
# slicing a string
Name = ‘Massachusetts’
#Fenceposts – starting at index 1 and ending
# before index 2
Name[1:2]
# find the first 4 characters of Name
Name[:4]

Example: using length
len(Name) shows length of string
• Find the length of a string variable called
name with a value of ‘Massachusetts’
• Print the last 4 characters in the name string

Example Function 1
# Python uses dynamic typing (type is set when
a variable is assigned or argument is passed)
# Save in file called test.py
def print_value(argument):
# Print the passed argument
print argument
# default return is none for a function

Import ‘test.py’ - DIY
# Importing a script(that is stored in the same
folder, otherwise use full path)
import test
# print number from function
Answer = test.print_value(15)

Example Function 2
# Function will give an error if argument is not a
number
# Save this file as example.py
def change_value(number):
# Add documentation here
number += 8
return number

Import ‘example.py’ - DIY
# Importing a script (that is stored in the same
folder, otherwise use full path)
import example
# get new number from function
Answer = example.change_value(15)
print Answer

Exercise 1: Function to return shapefile
name
• Take in full file path string:
“C:Usersdsheehandesktopinterstatehigh
ways.shp”
• Return shapefile name:
“InterstateHighways.shp”
Should work for any full file path string, not one
case. You should have a second .py file to call the
function.
Examine python documentation http://doc.python.org;
in particular string.split(), len()

Warning: beware when copying from
slides
• Double quotes aren’t the same in PP as in IDLE
• Maintain indents so logic of the program
remains as intended
• And don’t open shapefiles with a cursor that
are open in Arcmap

Accessing the Attribute Table
def sumUpLength(shapefile):
# returns the length of all line segments in an attribute
table
import arcpy, os
sumLength = 0.0
try:
with arcpy.da.SearchCursor(shapefile, (“LENGTH”))
as cursor:
for row in cursor:
sumLength += row[0]
del cursor
except:
return -9999
return sumLength

Exercise 2
Call the function in the previous slide (called
accessAT.py) from a .py file. You should submit
a full path to the shapefile
(example -
“C:dsheehandesktopInterstateHighways.shp”)
as the argument to the function. Once this works,
add a where clause to the searchCursor function
to find only records with a length of 40 KM or
greater. Search help for Search Cursor and look for
AddFieldDelimiters.

Exercise 3
• Working with cursors
• Select by attributes
• Select by location

Exercise 3 - cursor
We are working with a shapefile, stepping
through the atttribute table of a FEATURE
CLASS with the cursor:
FC =
"C:UsersdsheehanDesktopkenyaDistrict2.shp
"
with arcpy.da.SearchCursor(FC, ("DISTRICT")) as cursor:
for row in cursor:
print row[0]
del cursor

Exercise 3 – selecting using attributes
Selecting by layer requires converting from a FEATURE CLASS to
LAYER but uses the same shapefile but opened a second time.
before loop
arcpy.MakeFeatureLayer_management("C:Usersdsheehan
DesktopkenyaDistricts.shp", "lyr")
Inside loop
arcpy.SelectLayerByAttribute_management("lyr",
"NEW_SELECTION", ' "DISTRICT" = '' + row[0] + ''')
Confused by mix of quotes, single and double?

Exercise 3 – select by location
Using the Layer, not the feature class, all inside
the loop:
arcpy.SelectLayerByLocation_management
("lyr", "BOUNDARY_TOUCHES", "lyr")
And checking the number selected:
theCount = arcpy.GetCount_management("lyr")
print theCount

Exercise 3 - result
import arcpy, os
try:
with
arcpy.da.SearchCursor("C:UsersdsheehanDesktopPython2015currentkenyaDistricts.
shp", ("DISTRICT")) as cursor:
arcpy.MakeFeatureLayer_management("C:UsersdsheehanDesktopPython2015current
kenyaDistricts2.shp", "lyr")
for row in cursor:
arcpy.SelectLayerByAttribute_management("lyr", "NEW_SELECTION", ' "DISTRICT" = '' +
row[0] + ''')
arcpy.SelectLayerByLocation_management ("lyr", "BOUNDARY_TOUCHES", "lyr")
print row[0]
print theCount
arcpy.SelectLayerByLocation_management ("lyr", "BOUNDARY_TOUCHES", "lyr")
print theCount
del cursor
except:
print arcpy.GetMessages()
raise

Try this at home:
• Add code to test whether any of the nearby
polygons contain a specified district
• If not, select additional neighbors and test
• Count how many iterations before you find
the specified district
• You should do this inside a while loop

Python Programming for ArcGIS: Part II

More Related Content

Viewers also liked (14)

Similar to Python Programming for ArcGIS: Part II (20)

More from DUSPviz (20)

Recently uploaded (20)

Python Programming for ArcGIS: Part II