Icycoide/searxng
1
0
Fork 1
mirror of https://github.com/searxng/searxng.git synced 2025-08-03 02:22:22 +02:00
Code Issues Projects Releases Packages Wiki Activity

Bing news' unit test

Browse source 
I have no idea why coverage tell 97% and 2 misses in branches. If anyone has an idea...
This commit is contained in:
Cqoicebordel 2015-01-29 20:56:57 +01:00
parent dad0434f34
commit efde2c21c8
3 changed files with 249 additions and 17 deletions
Download patch file Download diff file Expand all files Collapse all files

29
searx/engines/bing_news.py
View file

@ -15,6 +15,7 @@ from lxml import html
from datetime import datetime, timedelta
from dateutil import parser
import re
from searx.engines.xpath import extract_text
# engine dependent config
categories = ['news']
@ -42,6 +43,7 @@ def request(query, params):
params['cookies']['_FP'] = "ui=en-US"
params['url'] = base_url + search_path
return params
@ -55,44 +57,37 @@ def response(resp):
for result in dom.xpath('//div[@class="sn_r"]'):
link = result.xpath('.//div[@class="newstitle"]/a')[0]
url = link.attrib.get('href')
title = ' '.join(link.xpath('.//text()'))
contentXPath = result.xpath('.//div[@class="sn_txt"]/div'
'//span[@class="sn_snip"]//text()')
title = extract_text(link)
contentXPath = result.xpath('.//div[@class="sn_txt"]/div//span[@class="sn_snip"]')
if contentXPath is not None:
content = escape(' '.join(contentXPath))
content = escape(extract_text(contentXPath))
# parse publishedDate
publishedDateXPath = result.xpath('.//div[@class="sn_txt"]/div'
'//span[contains(@class,"sn_ST")]'
'//span[contains(@class,"sn_tm")]'
'//text()')
'//span[contains(@class,"sn_tm")]')
if publishedDateXPath is not None:
publishedDate = escape(' '.join(publishedDateXPath))
publishedDate = escape(extract_text(publishedDateXPath))
if re.match("^[0-9]+ minute(s|) ago$", publishedDate):
timeNumbers = re.findall(r'\d+', publishedDate)
publishedDate = datetime.now()\
- timedelta(minutes=int(timeNumbers[0]))
publishedDate = datetime.now() - timedelta(minutes=int(timeNumbers[0]))
elif re.match("^[0-9]+ hour(s|) ago$", publishedDate):
timeNumbers = re.findall(r'\d+', publishedDate)
publishedDate = datetime.now()\
- timedelta(hours=int(timeNumbers[0]))
elif re.match("^[0-9]+ hour(s|),"
" [0-9]+ minute(s|) ago$", publishedDate):
publishedDate = datetime.now() - timedelta(hours=int(timeNumbers[0]))
elif re.match("^[0-9]+ hour(s|), [0-9]+ minute(s|) ago$", publishedDate):
timeNumbers = re.findall(r'\d+', publishedDate)
publishedDate = datetime.now()\
- timedelta(hours=int(timeNumbers[0]))\
- 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]))
publishedDate = datetime.now() - timedelta(days=int(timeNumbers[0]))
else:
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