Question

I figured already out how to work with ar_of_ar, but can’t select multiple elements of only the second column.

temp_ary=CSV.read(file, {col_sep->";"})  

the mounted file has the structure:

[Date, value]  
june6;1200.02  
jul6;500.04    
dec06; 3400.07

after having imported, I want to count the values which are greater than 1000. However, the normal adressing I’m used to does not work, thus selecting values 1..3 of column2:

temp_ary[2..3][1]  

only returns the first value?

I’m not sure whatever values I can expect, but I want to count occurring nil, null, 0, space values and so on, which are no real numbers. How can I efficiently do that?

Answer 1

You could do something like this:

require "csv"

def csv
  @csv ||= CSV.read("./test.csv", :col_sep => ';')
end

puts csv.count {|x| x.last.to_i > 1000}

You can pass a “block” to the count method to specify a predicate to count on.
See http://www.ruby-doc.org/core-1.9.3/Array.html#method-i-count

Also you might not be interested by the method “csv” and so on. The hard work is done by the count method.

EDIT: If you just wanted to select certain values of you CSV file. I suggest taking a look at the method “values_at”: http://www.ruby-doc.org/core-1.9.3/Array.html#method-i-values_at

csv.values_at(0..1).count {|x| x.last.to_i > 1000}