ISBNをAmazon APIに投げて自炊した本のタイトル向けの文字列を出力

isbn2title.py

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

# http://docs.amazonwebservices.com/AWSECommerceService/latest/DG/

from sys import argv
from time import gmtime, strftime
from urllib import urlencode, urlopen
from xml.etree import ElementTree as ET
from base64 import b64encode
import hashlib, hmac
from StringIO import StringIO


API_HOST = 'webservices.amazon.co.jp'
API_PATH = '/onca/xml'
SERVICE = 'AWSECommerceService'
OPERATION = 'ItemLookup'
VERSION = '2010-11-01'
NAMESPACE = "{http://webservices.amazon.com/%s/%s}" % (SERVICE, VERSION)


if len(argv) != 4:
    print("Usage: %s AccessKeyID SecretAccessKey ISBN" % argv[0])
    exit(-1)

ACCESS_KEY_ID = argv[1]
SECRET_ACCESS_KEY = argv[2]
ISBN = argv[3].replace('-', '')

query = sorted([
    ('Service', SERVICE),
    ('Version', VERSION),
    ('Operation',OPERATION),
    ('Timestamp', strftime("%Y-%m-%dT%H:%M:%SZ", gmtime())),
    ('AWSAccessKeyId', ACCESS_KEY_ID),
    ('IdType', "ISBN"),
    ('ItemId', ISBN),
    ('SearchIndex', "Books"),
])

string_to_sign = '\n'.join(['GET', API_HOST, API_PATH, urlencode(query)])
signature = b64encode(hmac.new(SECRET_ACCESS_KEY, string_to_sign, hashlib.sha256).digest())
query.append(('Signature', signature))

r = urlopen('http://'+API_HOST+API_PATH, urlencode(query))
dom = ET.parse(StringIO(r.read()))

if dom.findall('//'+NAMESPACE+"Error"):
    print("Error: ISBN=%s" % ISBN)
    exit(-1)

manufacturer = dom.findtext('//'+NAMESPACE+"Manufacturer")
title = dom.findtext('//'+NAMESPACE+"Title")
author = []
illustrator = []
for creator in dom.findall('//'+NAMESPACE+"Creator"):
    if creator.attrib['Role'] == u'著':
        author.append(creator.text)
    elif creator.attrib['Role'].find(u"イラスト") >= 0:
        illustrator.append(creator.text)
else:
    if not author:
        author = [dom.findtext('//'+NAMESPACE+"Author")]

filename = ("[%s][%s][%s]%s.pdf"
            % (manufacturer, ','.join(author), ','.join(illustrator), title))
filename.replace(u'/', u'／')
# print(filename)
print(filename.encode('utf8'))