## Imports ## 
import requests
import pandas as pd
from datetime import datetime
from datetime import timezone
from pages.Location import *

class ApiCall: 
    def getData(self):
        ## Api Variablen für NINA
        api_variablen = {
            "hochwasser": "/lhp/mapData",
            "polizei": "/police/mapData",
            "wetter": "/dwd/mapData",
            "katwarn": "/katwarn/mapData",
            "mowas": "/mowas/mapData"
        }

        # Dataframe Collumns definieren  
        columns = ['ID', 'Urgency','Area', 'Titel', 'Event', 'Datum']

        # NINA Api URLs 
        ninaBaseUrl = "https://warnung.bund.de/api31"
        ninaWarningsUrl = "https://nina.api.proxy.bund.dev/api31/warnings/"

        # Einzelne NINA Warnings abfragen 
        def get_api_warning(meldung):
            response = requests.get(ninaBaseUrl+meldung+".json")
            return response.json()

        # Details zu NINA Warning abfragen: Return DataFrame mit allen Ergebnissen 
        def get_api_details(warning):
            n = 0
            response = get_api_warning(warning)
            df = pd.DataFrame(columns=columns)
            # GetDetails for warning 
            for responses in response:
                id = responses["id"]
                warningDetails = requests.get(ninaWarningsUrl+id+".json").json() 
                meldungsText = warningDetails["info"][0]["headline"]+ ": "+warningDetails["info"][0]["description"]
                warnung = warningDetails["info"][0]["event"]
                start = warningDetails["sent"]
                urgency = warningDetails["info"][0]["urgency"]
                area = warningDetails["info"][0]["area"][0]["areaDesc"]
                d = datetime.fromisoformat(start).astimezone(timezone.utc)
                d.strftime('%Y-%m-%d %H:%M:')
                time = d.replace(tzinfo=None)
                df.loc[n] = [id, urgency, area, meldungsText, warnung, time]
                n = n+1
            return df

        ## Dataframe to CSV 
        def df_to_csv(data, filename): 
            data.to_csv(filename+'.csv') 
        
        # Dataframe definieren    
        data = pd.DataFrame(columns=columns) 

        ## API abfragen für alle NINA Variablen/Warnings ##
        for x in api_variablen:
            data_api = get_api_details(api_variablen[x])
            data = pd.concat([data, data_api])

        # DataFrame df2 erstellen
        data['Area'] = data['Area'].str.split(', ')
        df2 = data.explode('Area')
        
        # Neue Zeile mit Plz erstellen
        def berechnung_func(row):
            loca = row['Area']
            ort = loca.replace("Gemeinde", "")
            ort = loca.replace("Stadt", "")
            ort = loca.replace("Mitgliedsgemeinde in Verwaltungsgemeinschaft", "")
            
            ###### TODO #####
            # Orte klein schreiben und filtern nach und ... # 
            ###### TODO #####
            location = Location(ort.strip()).getPostalCode()
            plz = location.iloc[0]['name']
            
            return plz

        df2['Plz'] = df2.apply(berechnung_func, axis=1)
        
        # API Daten in CSV speichern 
        df_to_csv(df2, "Data")

        # Return 
        return df2

# Testing    
# l1 = ApiCall().getData()