# this file contains the methods to fetch country data to be used to test prediction times

from codegreen_core.data import energy

from codegreen_core.utilities.metadata import get_country_metadata

from codegreen_core.data.entsoe import renewableSources, nonRenewableSources

from datetime import datetime

import pandas as pd

import numpy as np

import traceback

def gen_test_case(start, end, label):

country_list = get_country_metadata()

cases = []

for ci in country_list.keys():

cdata = country_list[ci]

cdata["country"] = ci

cdata["start_time"] = start

cdata["end_time"] = end

cdata["file"] = ci + label

cases.append(cdata)

return cases

def fetch_data(case):

data = energy(case["country"], case["start_time"], case["end_time"])["data"]

data.to_csv("./data/" + case["file"] + ".csv")

print(case["file"])

# test_cases_1 = gen_test_case(datetime(2024,1,1),datetime(2024,1,5),"1")

# for c in test_cases_1:

# fetch_data(c)

# test_cases_2 = gen_test_case(datetime(2023,7,5),datetime(2023,7,10),"2")

# for c in test_cases_2:

# print(c)

# fetch_data(c)

def test_cases_3():

cases = [

{

"country": "GR",

"start_time": datetime(2024, 1, 1),

"end_time": datetime(2024, 6, 30),

"file": "GR3",

},

{

"country": "LT",

"start_time": datetime(2024, 1, 1),

"end_time": datetime(2024, 6, 30),

"file": "LT3",

},

{

"country": "DE",

"start_time": datetime(2024, 1, 1),

"end_time": datetime(2024, 6, 30),

"file": "DE3",

},

]

for c in cases:

fetch_data(c)

# test_cases_3()

# Defining a function to convert and format the datetime

def convert_format(date_str):

# Convert string to datetime

date_time_obj = datetime.strptime(date_str, "%d.%m.%Y %H:%M")

# Format datetime object to the desired format

return date_time_obj.strftime("%Y%m%d%H%M")

def compute_rrs_error(downloaded, fetched):

d = pd.read_csv("./data/" + downloaded + ".csv")

d[["startTimeUTC", "end"]] = d["MTU"].str.extract(

r"(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) - (\d{2}\.\d{2}\.\d{4} \d{2}:\d{2})"

)

# Applying the conversion function to the start and end columns

d["startTimeUTC"] = d["startTimeUTC"].apply(convert_format)

d["startTimeUTC"] = d["startTimeUTC"].astype("int64")

d["end"] = d["end"].apply(convert_format)

f = pd.read_csv("./data/" + fetched + ".csv")

all_e = set(renewableSources + nonRenewableSources)

e_cols = set(f.columns.tolist())

e_present = list(all_e & e_cols)

combined = f.merge(d, on="startTimeUTC")

summary = {}

for e in e_present:

# print(f.iloc[0][e])

d_col = e + " - Actual Aggregated [MW]"

res_col = "residual-" + e

combined[res_col] = combined[d_col] - combined[e]

summary[e] = np.sqrt(np.sum(combined[res_col]))

# print(d.iloc[0][d_col])

print(summary)

return summary

# compute_rrs_error("gr_24_actual_downloaded","GR3")

# compute_rrs_error("de_24_actual_downloaded","DE3")

# compute_rrs_error("lt_24_actual_downloaded","LT3")

def get_forecast_for_testing():

try:

dates1 = [

[datetime(2024, 1, 5), datetime(2024, 1, 10), 1],

[datetime(2024, 3, 15), datetime(2024, 3, 20), 3],

[datetime(2024, 5, 10), datetime(2024, 5, 15), 5],

[datetime(2024, 8, 1), datetime(2024, 8, 10), 8],

]

clist = gen_test_case(datetime(2024, 7, 5), datetime(2024, 7, 10), "")

test_data = pd.DataFrame()

for c in clist:

for r in dates1:

try:

data = energy(c["country"], r[0], r[1], type="forecast")

print(c["country"], " ", r[2])

# data["data"].to_csv("data/"+c["country"]+str(r[2])+"_forecast.csv")

data["data"]["file_id"] = c["country"] + str(r[2])

print(data)

test_data = pd.concat([test_data, data["data"]], ignore_index=True)

except Exception as e:

print(traceback.format_exc())

print(e)

test_data.to_csv("data/prediction_testing_data.csv")

except Exception:

print(Exception)

# get_forecast_for_testing()

data = energy("DE", datetime(2024, 9, 11), datetime(2024, 9, 12), "generation", False)[

"data"

]

print(data)