geomon/gschliefgraben_glasfaser/update_daily_cron.py
2022-03-21 17:10:56 +01:00

178 lines
6.9 KiB
Python

'''
Tutorial link: https://realpython.com/flask-connexion-rest-api-part-2/
https://github.com/realpython/materials/blob/master/flask-connexion-rest-part-2/version_1/people.py
Sqlalchemy version: 1.2.15
Python version: 3.10
'''
import os
import json
import uuid
from datetime import datetime
from dotenv import load_dotenv, find_dotenv
from sqlalchemy.orm import session
from sqlalchemy import func, asc, desc
# from db.pg_models import Platform
from db.models import (
ObservationSchema, Observation, create_pg_session,
Dataset, Procedure, Phenomenon, Platform)
from gschliefgraben_glasfaser.my_api import MyApi
def main():
''' main method '''
pg_session: session = create_pg_session()
platform_sta_identifier = "gschliefgraben_glasfaser"
# sensor_list = ["inclino1_14", "inclino1_02"]
#sensor_list = os.environ.get("GLASFASER_GSCHLIEFGRABEN_SENSORS")
sensor_list = json.loads(os.environ['GLASFASER_GSCHLIEFGRABEN_SENSORS'])
# this will print elements along with their index value
for sensor in sensor_list:
pg_query = pg_session.query(Dataset) \
.join(Procedure) \
.join(Phenomenon) \
.filter(Procedure.sta_identifier == sensor.lower())
slope_dataset: Dataset = pg_query.filter(
Phenomenon.sta_identifier == "Slope").first()
if not slope_dataset:
print("Sensor " + sensor + " ist noch nicht angelegt!")
# exit()
continue
if not slope_dataset.is_published:
slope_dataset.is_published = 1
slope_dataset.is_hidden = 0
slope_dataset.dataset_type = "timeseries"
slope_dataset.observation_type = "simple"
slope_dataset.value_type = "quantity"
pg_session.commit()
# platform_exists: bool = pg_session.query(Platform.id).filter_by(
# sta_identifier=platform_sta_identifier).scalar() is not None
# if platform_exists:
# sensor_platform = pg_session.query(Platform.id) \
# .filter(Platform.sta_identifier == platform_sta_identifier) \
# .first()
# slope_dataset.fk_platform_id = sensor_platform.id
# create all the observation for the given sensor names
create_observations(sensor, slope_dataset)
first_slope_observation = pg_session.query(Observation) \
.filter(Observation.fk_dataset_id == slope_dataset.id) \
.order_by(asc('sampling_time_start')) \
.first()
if first_slope_observation is not None:
slope_dataset.first_time = first_slope_observation.sampling_time_start
slope_dataset.first_value = first_slope_observation.value_quantity
slope_dataset.fk_first_observation_id = first_slope_observation.id
last_slope_observation = pg_session.query(Observation) \
.filter(Observation.fk_dataset_id == slope_dataset.id) \
.order_by(desc('sampling_time_start')) \
.first()
if last_slope_observation is not None:
slope_dataset.last_time = last_slope_observation.sampling_time_start
slope_dataset.last_value = last_slope_observation.value_quantity
slope_dataset.fk_last_observation_id = last_slope_observation.id
pg_session.commit()
pg_session.close()
def create_observations(sensor: str, slope_dataset: Dataset):
''' create_observations method for given sensor '''
pg_session: session = create_pg_session()
# create access token
token_api = os.environ.get("TOKEN_API")
test_api = MyApi(token_api)
# The size of each step in days
# consider the start date as 2021-february 1 st
start_date = datetime.today()
query_date = start_date.strftime('%Y-%m-%d')
create_db_observations(sensor, query_date, test_api,
pg_session, slope_dataset)
pg_session.commit()
def create_db_observations(sensor: str, query_date, test_api, pg_session, dataset: Dataset):
''' parse each observation '''
query_date_obj = datetime.strptime(query_date, "%Y-%m-%d")
data = test_api.getSensorData(sensor, query_date, query_date)
observation_array = (data['FeatureCollection']
['Features'][0]['geometry']['properties'][0])
# print(observation_array)
max_id = pg_session.query(func.max(Observation.id)).scalar()
if max_id is None:
max_id = -1
# pg_session.bulk_save_objects(observations)
for observation_json in observation_array:
ob_date_time = observation_json.get('DateTime')
datetime_obj = datetime.strptime(ob_date_time, "%Y-%m-%dT%H:%M:%S.%fZ")
if datetime_obj.date() != query_date_obj.date():
continue
ob_value = observation_json.get('Value')
if ob_value is None:
continue
# max_id = max_id + 1
max_id = create_observation(
observation_json, pg_session, max_id, dataset)
# pg_session.commit()
print("observations for date " + query_date + " and sensor " + sensor +
" succesfully imported \n")
def create_observation(observation_json: ObservationSchema, db_session, max_id, dataset: Dataset):
"""
This function creates a new observation in the people structure
based on the passed-in observation data
:param observation: person to create in people structure
:return: 201 on success, observation on person exists
"""
ob_id: str = str(observation_json.get('id'))
# db_session = create_pg_session()
existing_observation: bool = (
db_session.query(Observation)
.filter(Observation.value_identifier == ob_id, Observation.fk_dataset_id == dataset.id)
.one_or_none()
)
# Can we insert this observation?
if existing_observation is None:
max_id += 1
# Create a person instance using the schema and the passed in person
schema = ObservationSchema()
# deserialize to python object
new_observation: Observation = schema.load(observation_json)
# new_observation.id = max_id
new_observation.sta_identifier = str(uuid.uuid4())
new_observation.sampling_time_start = new_observation.result_time
new_observation.sampling_time_end = new_observation.result_time
new_observation.fk_dataset_id = dataset.id
# Add the person to the database
db_session.add(new_observation)
# dataset.observations.append(new_observation)
# db_session.commit()
# Serialize and return the newly created person in the response
# data = schema.dump(new_observation)
# return data, 201
return max_id
# Otherwise, nope, person exists already
else:
print(409, f'Observation {ob_id} exists already')
return max_id
if __name__ == "__main__":
load_dotenv(find_dotenv())
sensor_list1 = os.environ.get('GLASFASER_GSCHLIEFGRABEN_SENSORS', [])
print(f'sensors: {sensor_list1} .')
main()