from codegreen_core.tools.loadshift_location import predict_optimal_location,predict_optimal_location_now

from datetime import datetime,timedelta

import pandas as pd

import pytz

def test_location_now():

a,b,c,d = predict_optimal_location_now(["DE","HU","AT","FR","AU","NO"],5,0,50,datetime(2024,9,13))

print(a,b,c,d)

# test_location_now()

def fetch_data(month_no,countries):

data = pd.read_csv("data/prediction_testing_data.csv")

forecast_data = {}

for c in countries:

filter = data["file_id"] == c+""+str(month_no)

d = data[filter].copy()

if(len(d)>0):

forecast_data[c] = d

return forecast_data

def test_locations():

cases = [

{

"month":1,

"c":["DE","NO","SW","ES","IT"],

"h":5,

"m":0,

"p":50,

"s":"2024-01-05 02:00:00",

"e": 10

}

]

for case in cases:

data = fetch_data(case["month"],case["c"])

start_utc = datetime.strptime(case["s"], '%Y-%m-%d %H:%M:%S')

start_utc = pytz.UTC.localize(start_utc)

start = start_utc.astimezone(pytz.timezone('Europe/Berlin'))

end = (start + timedelta(hours=case["e"]))

a,b,c,d = predict_optimal_location(data,case["h"],case["m"],case["p"],end,start)

print(a,b,c,d)

test_locations()