Sha256: c82d186a204efb3624cec903484c8d32f6413faaba18e51595e650ad1c2f44b3
Contents?: true
Size: 1.77 KB
Versions: 29
Compression:
Stored size: 1.77 KB
Contents
module Eco
class CSV
class Stream
include Eco::Language::AuxiliarLogger
attr_reader :filename
def initialize(filename, **kargs)
raise ArgumentError, "File '#{filename}' does not exist" unless ::File.exist?(filename)
@filename = filename
@params = {
headers: true,
skip_blanks: true
}.merge(kargs)
init
end
def for_each(start_at_idx: 0)
raise ArgumentError, 'Expecting block, but not given.' unless block_given?
move_to_idx(start_at_idx)
yield(row, next_idx) while (self.row = csv.shift)
rescue StandardError => err
self.exception = err
raise
ensure
(fd.close; @fd = nil) if fd.is_a?(::File) # rubocop:disable Style/Semicolon
if exception
# Give some feedback if it crashes
msg = []
msg << "Last row IDX: #{idx}"
msg << "Last row content: #{row.to_h.pretty_inspect}"
puts msg
log(:debug) { msg.join("\n") }
end
end
def move_to_idx(start_at_idx)
start_at_idx ||= 0
next_idx while (idx < start_at_idx) && (self.row = csv.shift)
end
private
attr_reader :params
attr_reader :idx, :fd
attr_accessor :row, :exception
def next_idx
idx.tap { @idx += 1 }
end
# see https://dalibornasevic.com/posts/68-processing-large-csv-files-with-ruby
def csv
return @csv if instance_variable_defined?(:@csv)
@fd = ::File.open(filename, 'r')
@csv = Eco::CSV.new(fd, **params)
end
def init
@idx ||= 0 # rubocop:disable Naming/MemoizedInstanceVariableName
end
end
end
end
Version data entries
29 entries across 29 versions & 1 rubygems