import pandas as pd

import numpy as np

import re

productList = pd.read_csv('Products_2K.csv')

# productList = pd.read_csv('productlist.csv')

productList.shape

#productList['product_Name'] = productList['product_Name'].str.translate(None, ",~`!@#$%^&*()_-+=?.{}[]|\/<>:;")

# productList['product_Name'] = productList['product_Name'].str.replace('\W','')

# drop all rows where product_name is empty

productList.drop(productList[~productList.product_name.notnull()].index, inplace = True)

# Sort values based on product_name

productList.sort_values("product_name", axis = 0, ascending = True, inplace = True, na_position = 'first')

# dropping ALL duplicte values

data.drop_duplicates(subset ="First Name", keep = False, inplace = True)

uniqueSKUs = set(productList.product_name.str.lower())

len(uniqueSKUs)

productList.shape

re.findall(r"\W+",'Caress Velvet Bliss Ultra Silkening Beauty Bar Qty 6 Ct')

uniqueSKUs = pd.DataFrame(uniqueSKUs)

from quantulum3 import parser

item = parser.parse('Caress Velvet Bliss Ultra Silkening Beauty Bar Qty 6 Ct')

" ".join([(x.surface) for x in item])

item

x = productList.groupby(['product_name', 'generic_name', 'quantity', 'brands', 'brands_tags', 'categories', 'categories_tags', 'categories_en'])['product_name'].count()

x.shape

# sort the data in the dataframe

productList.sort_values(['product_name', 'pid'], inplace = True)

# Duplicate the product_name column

# Split the Duplicate into and convert to lowercase

# Weight into separate column

# Stopwords into separate columns

# Name

x = productList.groupby('product_name')['product_name'].count().sort_values(ascending=False)