CODEWITHSUNDEEP
pythonpolygonarcgisshapefilearcpy
user2841098
user2841098

Reputation: 323

Python: how to create a point shape-file from a text file

I'm writing a python code to read the points in a polygon shape-file and save them in a point shape file. So first I made a text file and stored the points' (x,y) in that .txt file. then I tried to make a point shape-file from the text file but it gave an error.
here is the code (just the last part):

creat point shape-file from text file 
import fileinput
import string
import os
env.overwriteOutput=True
outpath="C:/roadpl"
newfc="newpoint.shp" 
arcpy.CreateFeatureclass_management(outpath, newfc, "Point")
infile="C:/roadpl/roadL5.txt"
cursor=arcpy.da.InsertCursor(newfc, ["SHAPE@"])
array=arcpy.Array()
for line in fileinput.input(infile):
    X, Y=string.split(line, " ")
    array.add(arcpy.Point(X,Y))
cursor.insertRow([arcpy.Point(array)])
fileinput.close()
del cursor

Here is the error:

Traceback (most recent call last):
  File "C:\Lab5\P_Code_L5", line 49, in <module>
    point.X, point.Y  = line.split()
  File "C:\Program Files (x86)\ArcGIS\Desktop10.1\arcpy\arcpy\arcobjects\_base.py", line 87, in _set
    return setattr(self._arc_object, attr_name, cval(val))
RuntimeError: Point: Input value is not numeric

Upvotes: 7

Views: 7557

Answers (2)

Casivio
Casivio

Reputation: 363

Here is an alternative solution. It reads in a tab-delineated .ascii file (which is very similar to a .txt file) which it assumes has headers of Longitude and Latitude. I came across this question and then ended up with this solution so I figured that it can probably help someone.

# ------------------------------------------------------
# IMPORTS
# ------------------------------------------------------

import os
import pandas as pd
from shapely.geometry import Point, mapping
from fiona import collection

# ------------------------------------------------------
# INPUTS
# ------------------------------------------------------

# Define path
path = os.path.abspath(os.path.dirname(__file__))

# Set working directory
os.chdir(path)  

# Define file to convert
file = 'points.ascii'

# Define shp file schema
schema = { 'geometry': 'Point', 'properties': { 'LocationID': 'str', 'Latitude': 'float', 'Longitude': 'float' } }

# Read in data
data = pd.read_csv(file, sep='\t') 

# Define shp file to write to
shpOut = 'points.shp'

# Create shp file
with collection(shpOut, "w", "ESRI Shapefile", schema) as output:
    # Loop through dataframe and populate shp file
    for index, row in data.iterrows():
        
        # Define point
        point = Point(row['Longitude'], row['Latitude'])
        # Write output
        output.write({
            'properties': {'LocationID': row['LocationID'], 'Latitude': row['Latitude'], 'Longitude': row['Longitude'] }, 
            'geometry': mapping(point)
        })

Upvotes: 1

bcollins
bcollins

Reputation: 3459

have you tried calling float(X), float(Y) as maybe its doesn't like strings?

If you can get your input into a numpy array, you can convert that to a feature class in one step:

http://arcpy.wordpress.com/2012/09/14/building-feature-classes-from-numpy-arrays/

Upvotes: 1

Related Questions