noklam

noklam

import time
from datetime import datetime, timedelta
from airflow import DAG
from airflow.operators.python_operator import PythonOperator

default_args = {
    'owner': 'Meng Lee',
    'start_date': datetime(2100, 1, 1, 0, 0),
    'schedule_interval': '@daily',
    'retries': 2,

noklam

import time
from datetime import datetime, timedelta
from airflow import DAG
from airflow.operators.python_operator import PythonOperator, BranchPythonOperator
from airflow.operators.dummy_operator import DummyOperator
from airflow.operators.slack_operator import SlackAPIPostOperator

default_args = {
    'owner': 'Meng Lee',
    'start_date': datetime(2100, 1, 1, 0, 0),

noklam

import os
import time
import json
import logging
from datetime import datetime, timedelta
from selenium import webdriver
from airflow import DAG
from airflow.operators.python_operator import PythonOperator, BranchPythonOperator
from airflow.operators.dummy_operator import DummyOperator
from airflow.operators.slack_operator import SlackAPIPostOperator

noklam

#!/usr/bin/env python3
# coding: utf-8
"""
Author       : noklam
Created Time : 2019-11-09 12:30:04
Base on: https://gist.github.com/weaming/32b7b62956b304ff66fbae396ca9c86b
Prerequisite:
    python3 -m pip install cson arrow
"""
import json

noklam

noklam

noklam

noklam

noklam

# # Using pandas merge_as_of (similar to a merge join) to do inexact join (join the nearest key instead)
import pandas as pd

# %%
table_a = pd.DataFrame([('2020-01-01'), ('2020-01-03'), ('2020-01-06')],
                       columns=['PK'])
table_b = pd.DataFrame([('2020-01-01', 'A'), ('2020-01-02', 'A'),
                        ('2020-01-04', 'B'), ('2020-01-05', 'B')],
                       columns=['FK', 'Category'])
# %%