Adding elevation, name and description in GeoJSON when converting DataFrame with Python

Question

I have this working code, data2kml() function works ok, but data2geojson() is incomplete. How can I add the elevation, name and the description in the GeoJSON? I can't understand how FeatureCollection works.

Obs: I can't use GeoPandas because of some dependency conflicts.

import json
import geojson
import simplekml
import pandas as pd
def data2kml(df):
 kml = simplekml.Kml()
 df.apply(lambda X: kml.newpoint(
 name=X["name"], 
 description=unicode(X["description"].decode('utf8')), 
 coords=[( X["long"], X["lat"], X["elev"])]
 )
 , axis=1)
 kml.save(path='map.kml')
def data2geojson(df):
 points = []
 df.apply(lambda X: points.append( (float(X["long"]), 
 float(X["lat"]))
 )
 , axis=1)
 with open('map.geojson', 'w') as fp:
 geojson.dump(geojson.MultiPoint(points), fp, sort_keys=True)
col = ['lat','long','elev','name','description']
data = [[-29.9953,-70.5867,760,'A','Place a'],
 [-30.1217,-70.4933,1250,'B','Place b'],
 [-30.0953,-70.5008,1185,'C','Place c']]
df = pd.DataFrame(data, columns=col)
data2kml(df)
data2geojson(df)

Answer 1

Using what i learned from the answer of @gene this solution avoids the use of iterrows because iterrows have performance issues.

Valid for Python 3.X

import pandas as pd
import geojson
def data2geojson(df):
 features = []
 insert_features = lambda X: features.append(
 geojson.Feature(geometry=geojson.Point((X["long"],
 X["lat"],
 X["elev"])),
 properties=dict(name=X["name"],
 description=X["description"])))
 df.apply(insert_features, axis=1)
 with open('map1.geojson', 'w', encoding='utf8') as fp:
 geojson.dump(geojson.FeatureCollection(features), fp, sort_keys=True, ensure_ascii=False)
col = ['lat','long','elev','name','description']
data = [[-29.9953,-70.5867,760,'A','Place ñ'],
 [-30.1217,-70.4933,1250,'B','Place b'],
 [-30.0953,-70.5008,1185,'C','Place c']]
df = pd.DataFrame(data, columns=col)
data2geojson(df)

Valid for Python 2.X

import pandas as pd
import geojson
def data2geojson(df):
 features = []
 df.apply(lambda X: features.append( 
 geojson.Feature(geometry=geojson.Point((X["long"], 
 X["lat"], 
 X["elev"])), 
 properties=dict(name=X["name"], 
 description=unicode(X["description"].decode('utf8'))))
 )
 , axis=1)
 with open('map.geojson', 'w') as fp:
 geojson.dump(geojson.FeatureCollection(features), fp, sort_keys=True)
col = ['lat','long','elev','name','description']
data = [[-29.9953,-70.5867,760,'A','Place a'],
 [-30.1217,-70.4933,1250,'B','Place b'],
 [-30.0953,-70.5008,1185,'C','Place c']]
df = pd.DataFrame(data, columns=col)
data2geojson(df)

Answer 2

Geoff Boeing provides a solution in Exporting Python Data to GeoJSON and Convert a pandas dataframe to geojson for web-mapping (Jupyter notebook) for 2D coordinates and you can adapt his script for 3D coordinates

def df_to_geojson(df, properties, lat='lat', lon='long', z='elev'):
 geojson = {'type':'FeatureCollection', 'features':[]}
 for _, row in df.iterrows():
 feature = {'type':'Feature',
 'properties':{},
 'geometry':{'type':'Point','coordinates':[]}}
 feature['geometry']['coordinates'] = [row[lon],row[lat],row[z]]
 for prop in properties:
 feature['properties'][prop] = row[prop]
 geojson['features'].append(feature)
 return geojson
cols = ['name', 'description']
df_to_geojson(df, cols)
{'type': 'FeatureCollection', 'features': [{'geometry': {'type': 'Point', 'coordinates': [-70.5867, -29.9953, 760]}, 'type': 'Feature', 'properties': {'name': 'A', 'description': 'Place a'}}, {'geometry': {'type': 'Point', 'coordinates': [-70.4933, -30.1217, 1250]}, 'type': 'Feature', 'properties': {'name': 'B', 'description': 'Place b'}}, {'geometry': {'type': 'Point', 'coordinates': [-70.5008, -30.0953, 1185]}, 'type': 'Feature', 'properties': {'name': 'C', 'description': 'Place c'}}]}

To explain the process with single features

for i, row in df.iterrows():
 print i,
 feature = {'type':'Feature','properties':{},'geometry':{'type':'Point','coordinates':[]}}
 feature['geometry']['coordinates'] = [row.long,row.lat,row.elev]
 for prop in cols:
 feature['properties'][prop] = row[prop]
 print feature
0 {'geometry': {'type': 'Point', 'coordinates': [-70.5867, -29.9953, 760]}, 'type': 'Feature', 'properties': {'name': 'A', 'description': 'Place a'}}
1 {'geometry': {'type': 'Point', 'coordinates': [-70.4933, -30.1217, 1250]}, 'type': 'Feature', 'properties': {'name': 'B', 'description': 'Place b'}}
2 {'geometry': {'type': 'Point', 'coordinates': [-70.5008, -30.0953, 1185]}, 'type': 'Feature', 'properties': {'name': 'C', 'description': 'Place c'}}

• jrovegno: Thanks for the answer, I finally learned the internal structure of the geojson format. With this knowledge i'm going to publish a solution that avoid the use of iterrows because this.

Answer 3

For Python 3.6 users, this works for me:

where df = yourdataframe, properties are the usefull_columns, lat='latitude', lng='longitude'

The usefull_columns come from the original dataframe

I've added some testing points

# create the function
def df_to_geojson(df, properties, lat='latitude', lng='longitude'):
 """
 Turn a dataframe containing point data into a geojson formatted python dictionary
 df : the dataframe to convert to geojson
 properties : a list of columns in the dataframe to turn into geojson feature properties
 lat : the name of the column in the dataframe that contains latitude data
 lng : the name of the column in the dataframe that contains longitude data
 """
 # create a new python dict to contain our geojson data, using geojson format
 geojson = {'type':'FeatureCollection', 'features':[]}
 # loop through each row in the dataframe and convert each row to geojson format
 # x is the equivalent of the row, df.iterrows converts the dataframe in to a pd.series object
 # the x is a counter and has no influence
 for x, row in df.iterrows():
 feature = {'type':'Feature',
 'properties':{},
 'geometry':{'type':'Point',
 'coordinates':[]}}
 # fill in the coordinates
 feature['geometry']['coordinates'] = [float(row.lng),float(row.lat)]
 # be aware that the dataframe is a pd.series
 #print('rowitem converts to ndarray(numpy) :\n ', row)
 # convert the array to a pandas.serie
 geo_props = pd.Series(row)
 # for each column, get the value and add it as a new feature property
 # prop determines the list from the properties
 for prop in properties:
 #loop over the items to convert to string elements
 #convert to string
 if type(geo_props[prop]) == float:
 #print('ok')
 geo_props[prop] = str(int(geo_props[prop]))
 # now create a json format, here we have to make the dict properties
 feature['properties'][prop] = geo_props[prop]
 # add this feature (aka, converted dataframe row) to the list of features inside our dict
 geojson['features'].append(feature)
 return geojson

• Vince: Please remember to indent your code so it is formatted correctly. The {} button does this for you.

Answer 4

I have used this code:

def return_geojson(df, latlng):
 d = dict(type="FeatureCollection",features=[])
 for ind,row in df.fillna('').iterrows():
 p = row.drop(latlng).to_dict()
 g = dict(type="Point",coordinates=list(row[latlng])[::-1])
 f = dict(type="Feature", properties=p, geometry=g)
 d['features'].append(f)
 return d
with open('map.geojson','w') as f:
 json.dump(return_geojson(df,['lat','lon']),f)