az fixes

oml/meta/amazon.py

@@ -27,8 +27,13 @@ def info(key, value):
     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'])
-    info['description'] = strip_tags(decode_html(unquote(re.compile('encodedDescription\' : "(.*?)",').findall(data)[0])))
-    info['description'] = fix_bad_unicode(info['description'])
+    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'):
@@ -47,7 +52,6 @@ def info(key, value):
 
     if 'ISBN-13' in content_info:
         info['isbn'] = content_info['ISBN-13'].replace('-', '')
-        info['isbn'].append(content_info['ISBN-13'].replace('-', ''))
     elif 'ISBN-10' in content_info:
         info['isbn'] = stdnum.isbn.to_isbn13(content_info['ISBN-10'])
 
@@ -56,24 +60,42 @@ def info(key, value):
         for span in a:
             r = span.getchildren()[0].text.strip()
             role = get_role(r)
-            if not role in info: info[role] = []
+            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 not role in info: info[role] = []
+            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]
+    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'