How to speed up iteration through array in ruby - arrays

Question

I have multiple csv files that have the name and the price of products. There may be or may not be products that are in both files. I have to find the highest and the lowest price across these files for each product.
I joined products from both files into one array:
Dir["./*.csv"].each do |file|
CSV.foreach(file, headers:true) do |row|
tmpRow = row.to_s.chomp + "," + file #saving name of the input file
list.push(tmpRow.chomp.split(","))
end
end
The array list looks like this:
[["5893105","2.38", "weightOrSomethingIrrelevant", "./FIAT_2.csv"]]
This is the main algorithm:
while list[0] do
if list[0] != nil
tmpPart = list[0][0]
tmpParts = list.select{ |part, price| part == tmpPart}
tmpParts.each do |tp|
tmpPrices.push(tp[1])
end
list[0][2].to_f != 0.0 ? tmpWeight = list[0][2].to_s : tmpWeight = "Undefined"
tmpMaxPrice = tmpParts.select{|part, price| part == tmpPart && price == tmpPrices.max}
tmpMinPrice = tmpParts.select{|part, price| part == tmpPart && price == tmpPrices.min}
result.push([tmpPart, tmpWeight, tmpPrices.max, tmpMaxPrice[0].last, tmpPrices.min, tmpMinPrice[0].last)
tmpPart = ""
list = list - tmpParts
tmpParts = []
tmpPrices = []
tmpMaxPrice = []
tmpMinPrice = []
tmpWeight = ""
end
end
The input files are huge (over 200 000 rows), so I am having problems with efficiency of my algorithm (as it processes one row in half a second).
I am wondering if there is any better way to write this app.

Answer 1

I would split this into several parts:
1) I suggest you have a table which represents files (the file name, location, line number etc) and connected to that a product table (the row data from that file)
2) script / function to ingest files and store rows as DB records
3) script / function to analyse rows and find products by name, using the DB and pulling price info out using Min/max.
This could later be improved to deal with naming inconsistencies products vs product occurrences etc.