ExportToDatabase.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Wed Jul 14 13:43:45 2021

export the stationlist to database

@author: geopeter
"""

import psycopg2
import psycopg2.extras
from psycopg2 import sql
from psycopg2.extensions import ISOLATION_LEVEL_AUTOCOMMIT
import configparser
#from api.GetAverageData import get_interpolation_data_for_point

cfg = configparser.ConfigParser()
cfg.read('./config.ini')
assert "POSTGRES" in cfg, "missing POSTGRES in config.ini"
param_postgres = cfg["POSTGRES"]
param_interpol = cfg["INTERPOLATION"]
stationGPD = None


# get_average_data_for_point(52.5, 13.4)

# Create DB "temperatures_berteld_morstein"
def create_db(cursor):
    print("Creating DB: ", param_postgres['dbName'])
    create_db_query = sql.SQL("""CREATE DATABASE {};""".format(param_postgres['dbName']))
    cursor.execute(create_db_query)
    print('Done')


def drop_db(db_name):
    if dbexists(db_name):
        print("Drop DB: ", db_name)
        try:
            with psycopg2.connect(database=db_name, user=param_postgres["user"], password=param_postgres["password"], host=param_postgres["host"], port=param_postgres["port"]) as connection:
                with connection.cursor() as cursor:
                    cursor.execute("DROP DATABASE {};".format(db_name))
        except psycopg2.DatabaseError as error:
            # do nothing, because test db is clean
            print(error.message)
            return


# Checks, if database "temperatures_berteld_morstein" exists
def dbexists(cursor):
    try:
        cursor.execute("SELECT datname FROM pg_database WHERE datname LIKE '{}';".format(param_postgres['dbName']))
        db_exists = cursor.fetchall()[0]
        print('DB existing')
        print('Updating data')
        return True
    except(Exception, psycopg2.DatabaseError) as error:
        print('No DB found')
        return False


# If database "temperatures_berteld_morstein" not exists, create it
def check_for_db_existence(cursor):
    print("Checking for database existence")
    if dbexists(cursor):
        return True
    else: 
        create_db(cursor)
        return False


# Connect to DB "temperatures_berteld_morstein" to create table "temperatures". Also installs PostGIS
def create_table(station_list, cursor):
    print('Creating table stations')
    df_columns = list(station_list)
    columns = ['id INTEGER', 'lon NUMERIC', 'lat NUMERIC', 'country TEXT', 'file TEXT']
    for column in df_columns:
        if str(column).startswith('19') or str(column).startswith('20'):
            columns.append('"{}" NUMERIC'.format(column))
    columns_clean = str(columns).strip('[]').replace("'", "")

    query = sql.SQL("""CREATE TABLE stations ({});""".format(columns_clean))
    #print(query)
    cursor.execute(query)
    cursor.execute('CREATE EXTENSION postgis;')
    print('Done')


# Loading matrix coordinates from csv and writing it into table "stations" in db "temperatures_berteld_morstein"
def insert_empty_matrix_into_db(cursor):
    print('Inserting empty matrix into database')
    matrix_density = param_interpol['matrix_density']

    with open('clipped_matrix_{}x{}.csv'.format(matrix_density, matrix_density), 'r') as matrix:
        matrix_data = matrix.readlines()
        matrix_points = 0
        for line in matrix_data[1:]:
            matrix_points += 1
            values = ''  # Used in second parameter of cursor.execute() (Avoids SQL injection)
            data = line.split(';')
            id = int("9999" + data[0].replace('"', ''))
            lon = float(data[1])
            lat = float(data[2].replace('\n', ''))
            for n in [id, lon, lat]:
                values = (*values, n)  # adding n to existing tuple

            query = sql.SQL("INSERT INTO STATIONS (id, lon, lat, country) "
                            "VALUES ({id}, {lon}, {lat}, 'Germany');").format(id=sql.Placeholder(), lon=sql.Placeholder(), lat=sql.Placeholder())
            # print(query.as_string(cursor))
            # print(values)
            cursor.execute(query, values)
        print('Done')
        return matrix_points


# Calculating interpolation data for matrix using the according function from the API, and writing it into the database in bulk
def create_matrix_data(cursor, amount_points):
    print('Calculating interpolation data for matrix')
    # start_time = time.time()
    cursor.execute("SELECT id, lon, lat FROM stations WHERE file is NULL;")
    matrix_points = cursor.fetchall()
    update_data = []
    for i, point in enumerate(matrix_points):
        id = point[0]
        lon = point[1]
        lat = point[2]
        interpolation_data = get_interpolation_data_for_point(lat, lon, sql.SQL('*'), cursor)

        for year, value in interpolation_data.items():
            update_data.append({'year': year, 'value': value, 'id': id})
        if i % round((amount_points / 10)) == 0:
            finished = i / amount_points * 100
            print(round(finished), end="% ... ")
    print('', end=' 100%, Done. \n')

    query = sql.SQL("""UPDATE stations SET "%(year)s" = %(value)s WHERE id = %(id)s; """)
    print('Writing interpolation data to database')
    psycopg2.extras.execute_batch(cursor, query, update_data)   # Multiple times faster than using execute() in a for loop
    # print((time.time() - start_time), 'seconds')
    print('Done')


# Dumping all existing data from database. Inserting station data into database in bulk.
def createInsertStatement(station_list):
    # create INSERT INTO table (columns) VALUES('%s',...)
    station_list.columns.astype(str)
    df_columns = list(station_list)
    station_list = station_list.round(decimals=3)
    values = "VALUES({})".format(",".join(["%s" for _ in df_columns]))
    # create INSERT INTO table (columns) VALUES('%s',...)
    insert_stmt = """INSERT INTO {} ({}) {}""".format('stations', df_columns, values)
    
    return insert_stmt

def insert_data(station_list, cursor):
    print('Inserting data into database')
    if len(station_list) > 0:
        # print(stationList)
        cursor.execute("DELETE FROM stations;")

        #df_columns = list(station_list)
        # create (col1,col2,...)

        # As integers like 2018, 2017, etc. are not possible as column names, double quotes have to be added. This requires some tricks and cleanups
        #columns = ['"' + column + '"' for column in df_columns]
        # for column in df_columns:
        #     columns.append('"' + column + '"')
        #columns = str(columns).replace('[', '').replace(']', '').replace("'", "").replace('\n', '').replace(' ', '')
        station_list = station_list.round(decimals=3)

        # create VALUES('%s', '%s",...) one '%s' per column
        #values = "VALUES({})".format(",".join(["%s" for _ in df_columns]))

        # create INSERT INTO table (columns) VALUES('%s',...)
        #insert_stmt = """INSERT INTO {} ({}) {}""".format('stations', columns, values)
        insert_stmt = createInsertStatement(station_list)
        
        psycopg2.extras.execute_batch(cursor, insert_stmt, station_list.values)
    print('Done')


def export(station_list):
    db_name, user, pw, host, port = param_postgres['dbName'], param_postgres["user"], param_postgres["password"], param_postgres["host"], param_postgres["port"]

    with psycopg2.connect(database='postgres', user=user, password=pw, host=param_postgres["host"], port=port) as connection:
        connection.set_isolation_level(ISOLATION_LEVEL_AUTOCOMMIT)  # Needs to be in AUTOCOMMIT mode for creating database
        with connection.cursor() as cursor:
            new_db = False if check_for_db_existence(cursor) else True

    with psycopg2.connect(database=db_name, user=user, password=pw, host=param_postgres["host"], port=port) as connection:
        with connection.cursor() as cursor:
            if new_db: create_table(station_list, cursor)
            insert_data(station_list, cursor)
            amount_points = insert_empty_matrix_into_db(cursor)

    with psycopg2.connect(database=db_name, user=user, password=pw, host=param_postgres["host"], port=port,
                          keepalives=1, keepalives_idle=30, keepalives_interval=10, keepalives_count=5) as connection:
        with connection.cursor() as cursor:
            create_matrix_data(cursor, amount_points)
    print('Installation successful. You can run the api.py now.')