@@ -56,10 +56,14 @@ def response(resp):
5656 link = result.xpath('.//div[@class="newstitle"]/a')[0]
5757 url = link.attrib.get('href')
5858 title = ' '.join(link.xpath('.//text()'))
59− content = escape(' '.join(result.xpath('.//div[@class="sn_txt"]/div//span[@class="sn_snip"]//text()')))
60−
59+ contentXPath = result.xpath('.//div[@class="sn_txt"]/div//span[@class="sn_snip"]//text()')
60+ if contentXPath != None:
61+ content = escape(' '.join(contentXPath))
62+
6163 # parse publishedDate
62− publishedDate = escape(' '.join(result.xpath('.//div[@class="sn_txt"]/div//span[@class="sn_ST"]//span[@class="sn_tm"]//text()')))
64+ publishedDateXPath = result.xpath('.//div[@class="sn_txt"]/div//span[contains(@class,"sn_ST")]//span[contains(@class,"sn_tm")]//text()')
65+ if publishedDateXPath != None:
66+ publishedDate = escape(' '.join(publishedDateXPath))
6367
6468 if re.match("^[0-9]+ minute(s|) ago$", publishedDate):
6569 timeNumbers = re.findall(r'\d+', publishedDate)
@@ -74,9 +78,18 @@ def response(resp):
7478 publishedDate = datetime.now()\
7579 - timedelta(hours=int(timeNumbers[0]))\
7680 - timedelta(minutes=int(timeNumbers[1]))
81+ elif re.match("^[0-9]+ day(s|) ago$", publishedDate):
82+ timeNumbers = re.findall(r'\d+', publishedDate)
83+ publishedDate = datetime.now()\
84+ - timedelta(days=int(timeNumbers[0]))
7785 else:
78− publishedDate = parser.parse(publishedDate)
79−
86+ try:
87+ # FIXME use params['language'] to parse either mm/dd or dd/mm
88+ publishedDate = parser.parse(publishedDate, dayfirst=False)
89+ except TypeError:
90+ # FIXME
91+ publishedDate = datetime.now()
92+
8093 # append result
8194 results.append({'url': url,
8295 'title': title,