data.py

import os
import pandas as pd


class Olist:

    def get_data(self):
        """
        01-01 > This function returns all Olist datasets
        as DataFrames within a Python dict.
        """
        # Hint: You will need to find the absolute path of the csv folder in order to call this method from anywhere.
        # Hint 2: look at python __file__ attribute
        csv_path = os.path.join(os.path.dirname(__file__), '..', 'data', 'csv')
        csv_files = [path.replace('.csv', '') for path in os.listdir(csv_path) if path.endswith('.csv')]
        data = {}
        for name in csv_files:
            data[name] = pd.read_csv(os.path.join(csv_path, f'{name}.csv'))
        return data


    def get_matching_table(self):
        """
        01-01 > This function returns a matching table between
        columns [ "order_id", "review_id", "customer_id", "product_id", "seller_id"]
        """
        # Get data
        data = self.get_data()
        # Filter only on order_status delivered
        orders = data['olist_orders_dataset']
        orders = orders[orders['order_status']=='delivered']
        reviews = data['olist_order_reviews_dataset']
        order_items = data['olist_order_items_dataset']
        # Select only the columns of interest
        orders = orders[['order_id', 'customer_id']]
        reviews = reviews[['review_id', 'order_id']]
        order_items = order_items[['order_id', 'product_id', 'seller_id']]
        # Merge DataFrame
        merge_df = orders.merge(reviews, on='order_id', how='outer').merge(order_items, on='order_id', how='outer')

        return merge_df

    def ping(self):
        """
        You call ping I print pong.
        """
        print('pong')