openmedialibrary/oml/meta/amazon.py
2017-07-30 11:37:44 +02:00

106 lines
3.8 KiB
Python

from ox.cache import read_url
from ox import decode_html, strip_tags, find_re, fix_bad_unicode
import json
import re
from urllib.parse import unquote
import lxml.html
import stdnum.isbn
def info(key, value):
if key not in ('isbn',):
raise IOError('unknwon key %s' % key)
if len(value) == 13:
value = stdnum.isbn.to_isbn10(value)
if len(value) != 10:
raise IOError('invalid isbn %s' % value)
url = 'http://www.amazon.com/dp/' + value
data = read_url(url).decode()
doc = lxml.html.document_fromstring(data)
info = {}
if '<title>404 - Document Not Found</title>' in data:
return info
if 'To discuss automated access to Amazon data please' in data:
return info
for l in doc.xpath('//link[@rel="canonical" and @href]'):
info['asin'] = [l.get('href').rpartition('/')[-1]]
break
info['title'] = strip_tags(decode_html(doc.xpath('//span[@id="productTitle"]')[0].text))
info['title'] = re.sub(' \([^\)]+? Classics\)', '', info['title'])
info['title'] = re.sub(' \([^\)]+? Collection\)', '', info['title'])
d = re.compile('encodedDescription\' : "(.*?)",').findall(data)
if d:
info['description'] = strip_tags(decode_html(unquote(d[0])))
info['description'] = fix_bad_unicode(info['description'])
else:
info['description'] = ''
content = doc.xpath('//div[@class="content"]')[0]
content_info = {}
for li in content.xpath('.//li'):
v = li.text_content()
if ': ' in v:
k, v = li.text_content().split(': ', 1)
content_info[k.strip()] = v.strip()
if 'Language' in content_info:
info['language'] = content_info['Language']
if 'Publisher' in content_info:
if ' (' in content_info['Publisher']:
info['date'] = find_re(content_info['Publisher'].split(' (')[-1], '\d{4}')
info['publisher'] = content_info['Publisher'].split(' (')[0]
if '; ' in info['publisher']:
info['publisher'], info['edition'] = info['publisher'].split('; ', 1)
if 'ISBN-13' in content_info:
info['isbn'] = content_info['ISBN-13'].replace('-', '')
elif 'ISBN-10' in content_info:
info['isbn'] = stdnum.isbn.to_isbn13(content_info['ISBN-10'])
a = doc.xpath('//span[@class="a-size-medium"]')
if a:
for span in a:
r = span.getchildren()[0].text.strip()
role = get_role(r)
if role not in info:
info[role] = []
info[role].append(span.text.strip())
else:
for span in doc.xpath('//span[@class="author notFaded"]'):
author = [x.strip() for x in span.text_content().strip().split('\n') if x.strip()]
role = get_role(author[-1])
if role not in info:
info[role] = []
info[role].append(author[0])
covers = re.compile('data-a-dynamic-image="({.+?})"').findall(data)[0]
covers = json.loads(decode_html(covers))
last = [0, 0]
for url in covers:
if covers[url] > last:
last = covers[url]
info['cover'] = re.sub('(\._SX.+?_\.)', '.', url)
return info
def get_price(asin, currency='EUR'):
if currency == 'EUR':
url = 'http://www.amazon.de/dp/' + asin
else:
url = 'http://www.amazon.com/dp/' + asin
data = read_url(url).decode()
doc = lxml.html.document_fromstring(data)
for price in doc.xpath("//span[contains(@class, 'a-color-price')]"):
price = price.text_content().strip()
if currency == 'EUR':
price = price.replace('EUR ', '').replace(',', '.')
else:
price = price.replace('$', '').strip()
price = float(price)
return price
def get_role(value):
if 'Translator' in value:
role = 'translator'
elif 'Editor' in value:
role = 'editor'
else:
role = 'author'
return role