try:

    from datetime import timedelta
    from airflow import DAG
    
    from airflow.operators.python_operator import PythonOperator
    from datetime import datetime
    from pandas.io.json import json_normalize
    from airflow.operators.postgres_operator import PostgresOperator

    import pandas as pd
    import json
    import requests
    import numpy as np

    import psycopg2
    from sqlalchemy import create_engine

except Exception as e:
    print("Error {} ".format(e))

dRoW_api_end_url = "https://uat2.drow.cloud"

def getDrowToken(**context):

    response = requests.post(
    url=f"{dRoW_api_end_url}/api/auth/authenticate",
    data={
    "username": "icwp2@drow.cloud",
    "password": "dGVzdDAxQHRlc3QuY29t"
    }
    ).json()
    context["ti"].xcom_push(key="token", value=response['token'])

def getMongoDB(**context):
    token = context.get("ti").xcom_pull(key="token")
    response = requests.get(
    url=f"{dRoW_api_end_url}/api/module/document-export/airflow/workflow/605d53f21fb4b60ca76fb242?export_type=0",
    headers={
    "x-access-token": f"Bearer {token}",
    "ICWPxAccessKey": "nd@201907ICWP_[1AG:4UdI){n=b~"
    }
    )

    RISC_Data = json.loads(response.text)
    Mapping= {
    "Inspection Date" : "a01a_inspection_date",
    # "SupD signature time": "f2_checked_by_supd_on_date",
    }

    host                  = 'drowdatewarehouse.crlwwhgepgi7.ap-east-1.rds.amazonaws.com'  

    # User name of the database server
    dbUserName            = 'dRowAdmin'  

    # Password for the database user
    dbUserPassword        = 'drowsuper'  

    # Name of the database 
    database              = 'drowDateWareHouse'

    # Character set
    port                  = "5432"

    # Character set
    charSet               = "utf8mb4"  

    conn_string = ('postgres://' +
                           dbUserName + ':' + 
                           dbUserPassword +
                           '@' + host + ':' + port +
                           '/' + database)
    
    db = create_engine(conn_string)
    conn = db.connect()
    with conn as conn:
        df = pd.DataFrame()
        for x in RISC_Data:
            df_nested_list = json_normalize(x['data'])

            df2 = df_nested_list.reindex(columns=Mapping.keys())
            if len(x['ApproveLogSummary']) > 0:
                request_data = [data for data in x['ApproveLogSummary'] if data.get('statusName')=="B : RSS check"]
                if len(request_data) > 0 and 'from' in request_data[-1]:
                    df2['f2_checked_by_supd_on_date'] = request_data[-1]['from']
                else:
                    df2['f2_checked_by_supd_on_date'] = None
                if len(request_data) > 0 and 'to' in request_data[-1]:
                    df2['d4_submission_date'] = request_data[-1]['to']
                else:
                    df2['d4_submission_date'] = None
            else:
                df2['f2_checked_by_supd_on_date'] = None
                df2['d4_submission_date'] = None
                
            df2["report_name"] = df2["Inspection Date"].astype(str).str[:10]
            if len([data for data in x['ApproveLogSummary'] if data.get('statusName')=="Z : END"])>0 or len([data for data in x['ApproveLogSummary'] if data.get('statusName')=="C : Contractor Acknowledge and Archive"])>0:
                df2['report_complete_or_incomplete'] = 'complete'
            else:
                df2['report_complete_or_incomplete'] = 'incomplete'
            if 'data' in x and isinstance(x['data'], dict):
                for key in x['data']:
                    if 'Checklist' in key:
                        total_report = 0
                        total_x = 0
                        for item in x['data'][key]:
                            for item_key in item:
                                if item[item_key] == 'N/A' or item[item_key] == '✓':
                                    total_report += 1
                                if item[item_key] == '✘':
                                    total_x += 1
                                    total_report += 1
                        df2['nc_report_item'] = total_x
                        df2['total_report_item'] = total_report

            if (not df2['f2_checked_by_supd_on_date'].isnull().bool() and not df2['Inspection Date'].isnull().bool()):
                df2['complete_time_in_days'] = (((df2['f2_checked_by_supd_on_date'].astype('datetime64[ns]') - 
                df2['Inspection Date'].astype('datetime64[ns]'))/ np.timedelta64(1, 'h'))/24).round(2)
                if df2['complete_time_in_days'].isnull().bool() or df2['complete_time_in_days'].lt(0).bool():
                    df2['complete_time_in_days'] = 0
            else:
                df2['complete_time_in_days'] = 0

            df2.rename(columns=Mapping, inplace=True)
            df = df.append(df2)
        df['a01a_inspection_date']=df['a01a_inspection_date'].apply(pd.to_datetime)
        df['d4_submission_date']=df['d4_submission_date'].apply(pd.to_datetime)
        df['f2_checked_by_supd_on_date']=df['f2_checked_by_supd_on_date'].apply(pd.to_datetime)
        df.to_sql('cleansing_dc201911', con=conn, if_exists='replace', index= False)
    

# id SERIAL
create_table_sql_query = """ 
    CREATE TABLE IF NOT EXISTS cleansing_dc201911 (
    f2_checked_by_supd_on_date TIMESTAMP,
    a01a_inspection_date TIMESTAMP,
    d4_submission_date TIMESTAMP,
    report_name VARCHAR (100),
    report_complete_or_incomplete VARCHAR (100),
    nc_report BOOLEAN,
    complete_time_in_days NUMERIC(10,2)
    );
    """


# */2 * * * * Execute every two minute 
with DAG(
        dag_id="dc201911_cleaning",
        schedule_interval="0 15 * * *",
        default_args={
            "owner": "airflow",
            "retries": 1,
            "retry_delay": timedelta(minutes=5),
            "start_date": datetime(2023, 1, 17)
        },
        catchup=False) as f:
    
    getMongoDB = PythonOperator(
        task_id="getMongoDB",
        python_callable=getMongoDB,
        op_kwargs={"name": "Dylan"},
        provide_context=True,
    )

    getDrowToken = PythonOperator(
        task_id="getDrowToken",
        python_callable=getDrowToken,
        provide_context=True,
    )

    create_table = PostgresOperator(
        sql = create_table_sql_query,
        task_id = "create_table_task",
        postgres_conn_id = "postgres_rds",
    )

create_table >> getDrowToken >> getMongoDB