def select_upload_data(dataset_to_update, dataframe):
try:
# initialize an empty list to hold the features that we have already uploaded to mapbox
existing = []
# initialize API session
datasets = Datasets()
os.environ['MAPBOX_ACCESS_TOKEN'] = 'sk.eyJ1IjoiZmlyZS1yaXNrLXNmIiwiYSI6ImNqZGw1dTlwdDA1aXMzM3FrbDZpZnpmczMifQ.JbUPsvx8384WHOAfA9Vy9w'
# get features of the mapbox dataset by calling the API
features_collection = datasets.list_features(dataset_to_update).json()
# loop through the features of the mapbox dataset and save them to the list "existing"
for i in range(0,len(features_collection['features'])):
existing.append([
features_collection['features'][i]["id"],
features_collection['features'][i]["properties"]["name"],
])
# convert "existing" to a dataframe
existing = pd.DataFrame(existing,columns=["id","address"])
print("succesfully accessed the existing mapbox dataset")
except:
print("Error: could not access the mapbox dataset. Check that you are accessing your token correctly")
try:
# get rows in dataframe that are not yet uploaded
new_rows = pd.merge(left = dataframe,
right = existing,
on = "address",
how = "left")
new_rows = new_rows[new_rows.id.isnull() == True]
new_rows = new_rows.set_index(np.arange(0,len(new_rows)))
# create a "start point" for new row ids
# note: this was not a very smart way to creat ids, but I'm stuck with it for now
start = max(existing["id"].astype("int64"))
# create new random ids for these new uploads
new_ids = random.sample(range(start,start*5),len(new_rows))
print("succesfully selected new upload data")
except:
print("there was a problem uploading the data")
# make sure these new random ids are not already in the dataset
try:
[new_id not in existing["id"] for new_id in new_ids]
except:
"Error! Duplicate ids created"
# return the new rows and new ids to upload
return new_rows, new_ids
def check_upload_result(upload, dataset_to_update):
try:
# check that the dataset upload was successful
datasets = Datasets()
os.environ['MAPBOX_ACCESS_TOKEN'] = 'sk.eyJ1IjoiZmlyZS1yaXNrLXNmIiwiYSI6ImNqZGw1dTlwdDA1aXMzM3FrbDZpZnpmczMifQ.JbUPsvx8384WHOAfA9Vy9w'
resp = datasets.list_features(dataset_to_update).json()
assert len(resp['features']) == len(upload), "Only " + str(len(resp['features'])) + " features are included in the dataset"
except:
print("Error: unable to check the number of uploaded records. To manually check the execution: Mapbox > Studio > Datasets > Menu > View details")
import os
os.environ[
'MAPBOX_ACCESS_TOKEN'] = "pk.eyJ1IjoidmliaHUxOTk0NiIsImEiOiJjam01MjB1YjkwMDZrM3BsMHI4d21reXduIn0.j1yPdjaTrJ1i4Uc_dcT62g"
import pandas as pd
import numpy as np
from mapbox import Datasets
import matplotlib.pyplot as plt
import json
import sys
import geopandas as gpd
import descartes
from pandas.io.json import json_normalize
import pyvisgraph as vg
datasets = Datasets()
col = datasets.list_features("ck5gcnyor0rcw2nmdwaeaiko9").json()
#attrs = datasets.read_dataset("ck6ub6amu0h6g2nlhtl42fguy").json()
df = pd.read_json(json.dumps(col['features']))
#df.head()
#nycc = json_normalize(df['properties'])
nyc = json_normalize(df['geometry'])
#len(nycc.index) #== len(nyc.index)
#nyc.head()
pp = []
for i in nyc.index:
for ff in nyc['coordinates'][i]:
pp.append(Polygon(ff))
f = pd.DataFrame(pp)
if len(sys.argv) != 3:
print(
'Please supply a mode <create|publish> and tile set <plaetze|strassen|viertel|denkmaeler|gebaeude>.'
)
sys.exit(1)
mode = sys.argv[1]
tileset = sys.argv[2]
url = 'https://api.mapbox.com/tilesets/v1/'
access_string = '?access_token=' + os.getenv("tileset_token")
if 'source' in mode:
dataset_id = os.getenv(tileset + '_dataset')
datasets = Datasets(access_token=os.getenv("dataset_token"))
collection = datasets.list_features(dataset_id).json()
features = collection['features']
ld_features = '\n'.join(json.dumps(feature) for feature in features)
files = {'file': (tileset + '.geojson', ld_features)}
sources_url = url + 'sources/strassenlaerm/' + tileset + access_string
if mode == 'create_source':
r = requests.post(sources_url, files=files)
elif mode == 'replace_source':
r = requests.put(sources_url, files=files)
else:
print('Please use a valid mode.')
sys.exit(1)
elif mode == 'create_tileset':
with open('recipe-' + tileset + '.json', 'r') as recipe_file:
r = requests.post(url + 'strassenlaerm.' + tileset + access_string,
json=json.loads(recipe_file.read()))
def get_features(dataset_id):
''' Takes the ID of a dataset and returns a dict with all
the features in the dataset
'''
datasets = Datasets()
return datasets.list_features(dataset_id).json()
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
import sys
import os
import json
from mapbox import Datasets
from dotenv import load_dotenv
load_dotenv()
datasets = Datasets(access_token=os.getenv("dataset_token"))
streets_dataset = os.getenv("strassen_dataset")
squares_dataset = os.getenv("plaetze_dataset")
existing_streets = datasets.list_features(streets_dataset).json()['features']
existing_squares = datasets.list_features(squares_dataset).json()['features']
existing_objects = existing_streets + existing_squares
# print(existing_objects)
def generate_geometry(street_square, parts, msg):
def reverse_part(part_to_reverse):
part_to_reverse['properties']['endet_bei_'], part_to_reverse['properties']['beginnt_be'] = \
part_to_reverse['properties']['beginnt_be'], part_to_reverse['properties']['endet_bei_']
part_to_reverse['geometry']['coordinates'][0].reverse()
if street_square['properties']['type'] == 'street':
# print(street_square['properties']['name'])
parts_multi = []
# go through all parts belonging to an object to simplify geometry
# for p in parts:
# print(p['properties']['beginnt_be'] + ' -> ' + p['properties']['endet_bei_'] + ': ' + str(p['geometry']['coordinates']))