Calculate elements in CSV file - ruby

Question

I am new to Ruby and now I have an issue while I try to calculate some elements.
I've got 6 CSV files with the same headers and the question is how to find the total amount of payments for each payed month.
01-test.csv
Payment date,Payable month,House,Apartment,Amount of payment
2014-09-14,2014-08,Panel,84,5839.77
2014-09-14,2014-08,Brick,118,4251.63
2014-09-14,2014-08,Brick,97,471.5
2014-09-14,2014-08,Panel,53,236.22
2014-09-14,2014-08,Panel,83,4220.77
.......
02-test.csv
Payment date,Payable month,House,Apartment,Amount of payment
2014-10-01,2014-08,Brick,34,1522.59
2014-10-01,2014-08,Brick,117,1285.57
2014-10-01,2014-08,Brick,136,1925.97
2014-10-01,2014-08,Brick,24,1032.95
2014-10-01,2014-08,Brick,113,957.01
.......
Here is my code:
def create_month_array(payments)
months = []
months = payments.uniq { |a| a[:payed_for]
months
end
def payed_for_each_month(payments, months)
sums = Array.new(months.length){|a| a = 0}
months.each{|a|
if(a[:payed_for] == payments.each{|x| x[:payed_for]})
.....
end
}
p sum
sum.round(2)
end
Thanks for any hints.

Answer 1

Suppose the data were read from files into strings.
str1 =<<_
2014-09-14,2014-08,Panel,84,5839.77
2014-09-14,2014-08,Brick,118,4251.63
2014-09-14,2014-09,Brick,97,471.5
2014-09-14,2014-10,Panel,53,236.22
2014-09-14,2014-10,Panel,83,4220.77
_
str2 =<<_
2014-10-01,2014-08,Brick,34,1522.59
2014-10-01,2014-09,Brick,117,1285.57
2014-10-01,2014-09,Brick,136,1925.97
2014-10-01,2014-10,Brick,24,1032.95
2014-10-01,2014-11,Brick,113,957.01
_
We can then combine the strings into a single string, convert it an array of lines and then use a counting hash to aggregate values for each payable month, which I assume to be the values of the second field. See Hash::new, specifically when new is assigned an argument equal to the default value (here 0).
(str1 + str2).lines.each_with_object(Hash.new(0)) do |line,h|
_, payable_month, _, _, amount = line.split(',')
h[payable_month] += amount.to_f
end
#=> {"2014-08"=>11613.990000000002, (5839.77 + 4251.63 + 1522.59)
# "2014-09"=>3683.04, ( 471.5 + 1285.57 + 1925.97)
# "2014-10"=>5489.9400000000005, ( 236.22 + 4220.77 + 1032.95)
# "2014-11"=>957.01} ( 957.01)
If a hash h is defined
h = Hash.new(0)
Ruby expands h[payable_month] += amount.to_f to
h[payable_month] = h[payable_month] + amount.to_f
If h has no key payable_month, h[payable_month] on the right of the equality sign returns the default value. Hence,
h[payable_month] = 0 + amount.to_f
#=> amount.to_f
Note we could have alternatively written
(str1.lines + str2.lines).each_with_object(Hash.new(0))...
or we could have read each file line-by-line and written all those lines to one file.

Answer 2

To combine all CSV data across multiple files use the following:
csv_files = ["01-test.csv", "02-test.csv", "03-test.csv", "04-test.csv", "05-test.csv", "06-test.csv"]
csv_data = CSV.generate(headers: :first_row) do |csv|
csv << CSV.open(csv_files.first).readline
csv_files.each do |csv_file|
CSV.read(csv_file)[1..-1].each { |row| csv << row }
end
end
To then to the calculate the sum of each "Payable month" (or "Payment date",
it was not clear which was the payed month), you do the following
Interpret the data, using Ruby's CSV library
data = CSV.parse(csv_data, headers: true)
Group the data by the payed month
month_array = data.group_by { |row| row["Payable month"] }
# month_array = data.group_by { |row| row["Payment date"][0..6] }
Chose either line and comment out the other
For each month get the sum/reduce of all the "Amount of payment" into a
total for that month within our collection of totals
payed_for_each_month = month_array.each_with_object({}) do |(month, rows), totals|
totals[month] = rows.reduce(0.0) { |sum, row| sum + row["Amount of payment"].to_f }
end
This produces the final result with the presented data
payed_for_each_month
# => {"2014-08"=>21743.98}
If "Payment date" month was used instead the totals would produce the following:
month_array = data.group_by { |row| row["Payment date"][0..6] }
# ...
payed_for_each_month
# => {"2014-09"=>15019.890000000001,
# "2014-10"=>6724.09}
All the code together:
data = CSV.parse(csv_data, headers: true)
month_array = data.group_by { |row| row["Payable month"] }
# month_array = data.group_by { |row| row["Payment date"][0..6] }
payed_for_each_month = month_array.each_with_object({}) do |(month, rows), totals|
totals[month] = rows.reduce(0.0) { |sum, row| sum + row["Amount of payment"].to_f }
end
payed_for_each_month
# => {"2014-08"=>21743.98}
References:
group_by
reduce
each_with_object