Created
March 7, 2011 17:00
-
-
Save mps/858787 to your computer and use it in GitHub Desktop.
Refactored using FasterCSV
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/env ruby | |
require 'optparse' | |
require 'rubygems' | |
#require 'excelsior' | |
require 'mongo' | |
require 'fastercsv' | |
load 'progressbar.rb' | |
options = { :server => "localhost", :port => "27017", :db => "AccuAuto" } | |
optparse = OptionParser.new do |opts| | |
opts.banner = "Usage: import-vins.rb [options] file-to-import.csv" | |
opts.on("-s", "--server SERVER_URL", "The URL of the server to connect to. The default is 'localhost'.") do |server| | |
options[:server] = server | |
end | |
opts.on("-p", "--port SERVER_PORT", "The port number to connect on. The default is '27017'.") do |port| | |
options[:port] = port | |
end | |
opts.on("-d", "--db DATABASE_NAME", "The database to import to. The default is 'AccuAuto'.") do |db| | |
options[:db] = db | |
end | |
opts.on("-h", "--help', 'Display this screen") do | |
puts opts | |
exit | |
end | |
end | |
optparse.parse! | |
if (ARGV.count == 0) | |
puts optparse | |
exit | |
end | |
csv_file = ARGV[0] | |
puts "Parsing #{csv_file}" | |
rows = FasterCSV.read(csv_file) | |
total = rows.count - 1; | |
puts "Found #{total} rows." | |
puts "Connecting to mongodb://#{options[:server]}:#{options[:port]}/#{options[:db]}" | |
conn = Mongo::ReplSetConnection.new([options[:server], options[:port]], :read_secondary => true) | |
db = conn.db(options[:db]) | |
vins = db["VinInfo"] | |
puts "Removing #{vins.count} existing VINs." | |
vins.drop unless vins.count == 0 | |
bar = ProgressBar.new("Adding VINs", rows.count) | |
(1..total).each do |i| | |
r = rows[i] | |
r.each {|f| f.strip! unless f.nil?} | |
vin = { | |
"vin" => r[7], | |
"yr" => r[0].to_i, | |
"mka" => r[1], | |
"mk" => r[2], | |
"md" => r[3], | |
"sm" => r[4], | |
"mn" => r[5].to_i, | |
"t" => r[6], | |
"bipd" => r[8], | |
"pipmp" => r[9], | |
"cmp" => r[10], | |
"col" => r[11], | |
"pns" => r[12], | |
"bs" => r[13], | |
"ss" => r[14], | |
"cyl" => r[15], | |
"es" => r[16], | |
"eh" => r[17], | |
"wb" => r[19], | |
"ht" => r[20], | |
"cw" => r[21], | |
"gw" => r[22], | |
"cls" => r[23], | |
"n" => r[24] | |
} | |
svchars = r[18] | |
vchars = [] | |
idx = 0 | |
until idx >= svchars.length | |
vc = svchars[idx,4] | |
if !vc.nil? | |
vc.strip! | |
vchars << vc | |
end | |
idx += 4 | |
end | |
vin["vchars"] = vchars | |
if r[25] =~ %r{(\d+)/(\d+)/(\d+)} | |
vin["cd"] = Time.utc($3.to_i, $1.to_i, $2.to_i) | |
end | |
vins.insert(vin) | |
bar.inc | |
end | |
bar.finish | |
puts "Inserted #{vins.count} VINs into the collection." |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment