zaclys/searxng
1
0
Fork 1
mirror of https://github.com/searxng/searxng synced 2024-01-01 18:24:07 +00:00
Code Issues Packages Projects Releases Wiki Activity

[fix] bing_new engine : fix published date parsing

Browse Source
This commit is contained in:
Dalf 2014-09-07 18:10:05 +02:00
parent 7f1e2ba211
commit a92e3ba446
1 changed files with 18 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
searx/engines/bing_news.py
View File

@ -56,10 +56,14 @@ def response(resp):
link = result.xpath('.//div[@class="newstitle"]/a')[0] link = result.xpath('.//div[@class="newstitle"]/a')[0]
url = link.attrib.get('href') url = link.attrib.get('href')
title = ' '.join(link.xpath('.//text()')) title = ' '.join(link.xpath('.//text()'))
content = escape(' '.join(result.xpath('.//div[@class="sn_txt"]/div//span[@class="sn_snip"]//text()'))) contentXPath = result.xpath('.//div[@class="sn_txt"]/div//span[@class="sn_snip"]//text()')
if contentXPath != None:
content = escape(' '.join(contentXPath))
# parse publishedDate # parse publishedDate
publishedDate = escape(' '.join(result.xpath('.//div[@class="sn_txt"]/div//span[@class="sn_ST"]//span[@class="sn_tm"]//text()'))) publishedDateXPath = result.xpath('.//div[@class="sn_txt"]/div//span[contains(@class,"sn_ST")]//span[contains(@class,"sn_tm")]//text()')
if publishedDateXPath != None:
publishedDate = escape(' '.join(publishedDateXPath))
if re.match("^[0-9]+ minute(s|) ago$", publishedDate): if re.match("^[0-9]+ minute(s|) ago$", publishedDate):
timeNumbers = re.findall(r'\d+', publishedDate) timeNumbers = re.findall(r'\d+', publishedDate)
@ -74,9 +78,18 @@ def response(resp):
publishedDate = datetime.now()\ publishedDate = datetime.now()\
- timedelta(hours=int(timeNumbers[0]))\ - timedelta(hours=int(timeNumbers[0]))\
- timedelta(minutes=int(timeNumbers[1])) - timedelta(minutes=int(timeNumbers[1]))
elif re.match("^[0-9]+ day(s|) ago$", publishedDate):
timeNumbers = re.findall(r'\d+', publishedDate)
publishedDate = datetime.now()\
- timedelta(days=int(timeNumbers[0]))
else: else:
publishedDate = parser.parse(publishedDate) try:
# FIXME use params['language'] to parse either mm/dd or dd/mm
publishedDate = parser.parse(publishedDate, dayfirst=False)
except TypeError:
# FIXME
publishedDate = datetime.now()
# append result # append result
results.append({'url': url, results.append({'url': url,
'title': title, 'title': title,