http://forrst.com/posts/Quickly_Adding_300_000_Record_CSV_to_SQL_DB-1R6

cleanup_violation_data.sh

#!/bin/sh
iconv -t utf8 -c data/WebExtract.csv > data/web_extract_utf8.csv
iconv -t utf8 -c data/cuisine.csv > data/cuisine_utf8.csv

health_department.rb

#!/usr/bin/env ruby
# encoding: utf-8
require 'csv'
require 'mongo'

db = Mongo::Connection.new.db("health_department")

cuisines = {}
CSV.foreach('data/cuisine_utf8.csv') do |row|
  next if row[0] == 'CUISINECODE'
  cuisines[row[0]] = row[1]
end

db['inspections'].drop
db['inspections'].ensure_index('camis')

CSV.foreach('data/web_extract_utf8.csv') do |row|
  db['inspections'].insert({
    'camis'               => row[0],
    'dba'                 => row[1],
    'boro'                => row[2],
    'building'            => row[3],
    'street'              => row[4],
    'zip_code'            => row[5],
    'phone'               => row[6],
    'cuisine_code'        => row[7],
    'cuisine_description' => cuisines[row[7]],
    'inspection_date'     => row[8],
    'action'              => row[9],
    'violation_code'      => row[10],
    'score'               => row[11],
    'current_grade'       => row[12],
    'grade_date'          => row[13],
    'record_date'         => row[14]
  })
end

db['inspections'].remove({ 'camis' => 'CAMIS' })