importCsv2DDB.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-

import boto
import sys
import csv

"""
Import CSV data to your DynamoDB.
usage : importCSV2DDB.py -f <csv filepath>

prerequirements:
- you need to setup boto before you run. see also  http://boto.readthedocs.org/en/latest/boto_config_tut.html
- you need to create table before you run.
- the table name which you want to load data to must be same name with CSV file name (e.g. tablename 'rating' for 'rating.csv')
- the table Primary Key's attribute name must be 'id'. this script does not support 'Hash and Range' type.
- WARN!: This script is NOT consitered about your DynamoDB's provisiond throughput.
"""
if __name__ == '__main__':
	from optparse import OptionParser
	parser = OptionParser()
	parser.add_option('-f', '--file', dest='file', help='csv file path.')
	options, args = parser.parse_args(sys.argv[1:])
	if not options.file:   # if filename is not given
		parser.error('Filename not given')

	filepath = options.file
	tablename = filepath[filepath.rindex('/')+1:filepath.rindex('.')]

	conn = boto.connect_dynamodb()
	table = conn.get_table(tablename)

	i = 0
	header = []
	for row in csv.reader(open(filepath)):
		# 1行目はヘッダーとして扱う
		if i == 0:
			i = i+1
			header = row
			continue

		item_data = {}
		n = 0
		for fname in header:
			# 先頭のカラムは id なのでスキップ
			if fname == 'id':
				continue
			n = n + 1

			# DynamoDB の仕様上、空の値は入れられない。
			val = row[n]
			if val == None or val == '':
				continue
			item_data[fname] = row[n]

		item = table.new_item(
		        hash_key=row[0],
		        attrs=item_data
		    )
		item.put()