customer = read.csv("/Users/adamtomasov/Desktop/data/out.c_hackathon.customer_entity.csv",header = T, na.strings = c("","NA"))
colnames(customer)[colnames(customer)=="entity_id"] <- "customer_id"
colnames(customer)[colnames(customer)=="created_at"] <- "customer_created_at"
colnames(customer)[colnames(customer)=="updated_at"] <- "customer_updated_at"

customer_address = read.csv("/Users/adamtomasov/Desktop/data/out.c_hackathon.customer_address_entity.csv")
colnames(customer_address)[colnames(customer_address)=="entity_id"] <- "customer_address_id"
colnames(customer_address)[colnames(customer_address)=="parent_id"] <- "customer_address_customer_id"
colnames(customer_address)[colnames(customer_address)=="created_at"] <- "cusaddr_created_at"
colnames(customer_address)[colnames(customer_address)=="updated_at"] <- "cusaddr_created_at"

#merge customer dataframes
customer_df = merge(customer, customer_address,by.x = "customer_id" ,by.y="customer_address_customer_id")

#merge sales
data_sales_order = read.csv("/Users/adamtomasov/Desktop/data/out.c_hackathon.train_sales_order.csv")
colnames(data_sales_order)[colnames(data_sales_order)=="entity_id"] <- "order_id"
colnames(data_sales_order)[colnames(data_sales_order)=="created_at"] <- "order_created_at"
colnames(data_sales_order)[colnames(data_sales_order)=="updated_at"] <- "order_created_at"
data_sales_order_item = read.csv("/Users/adamtomasov/Desktop/data/out.c_hackathon.train_sales_order_item.csv")
colnames(data_sales_order_item)[colnames(data_sales_order_item)=="created_at"] <- "order_item_created_at"
colnames(data_sales_order_item)[colnames(data_sales_order_item)=="updated_at"] <- "order_item_created_at"
colnames(data_sales_order_item)[colnames(data_sales_order_item)=="product_id"] <- "product_id_ordered"
sales_df = merge(data_sales_order, data_sales_order_item, by = "order_id")


#merge data about order with product information
products = read.csv("/Users/adamtomasov/Desktop/data/out.c_hackathon.product_list.csv",header = T, na.strings = c("","NA"))
colnames(products)[colnames(products)=="category_id"] <- "product_category_id"
colnames(products)[colnames(products)=="parent_id"] <- "parent_prod_id"
ps_df = merge(sales_df, products,by.x = "product_id_ordered" ,by.y="product_id")

#final merge customer data with product sales data
finall_df = merge(customer_df, ps_df, by.x = "email", by.y = "customer_email")



#Vyjebanie store_id.y, x lebo je to to iste ako store name
#Store.id obsahuje chybajuce hodnoty pricom store_name nema prazdne hodnoty
drops <- c("store_id.x", "store_id.y")
finall_df = finall_df[ , !(names(finall_df) %in% drops)]
head(finall_df)
unique(product_type.x)
unique(product_type.y)

library(dplyr)
finall_df %>% group_by(email) %>% summarise(Freq=n())
zgrupovane = group_by(finall_df,email, order)$email

status

#vytvorenie noveho atributu price_in_EUR a odstranenie base_subtotal, currency atd
change_cur = finall_df$base_subtotal * finall_df$base_to_global_rate
finall_df$price_in_EUR = change_cur
finall_df$base_subtotal = NULL
finall_df$base_currency_code = NULL
finall_df$global_currency_code = NULL

#replace NA -> 0
finall_df$gender = replace(finall_df$gender, is.na(finall_df$gender),0)


write.csv(finall_df, file = "/Users/macbook/Desktop/GymBeam/GymBeam.csv")