mirror of
https://github.com/searxng/searxng.git
synced 2024-11-23 12:31:25 +01:00
94 lines
2.9 KiB
Python
94 lines
2.9 KiB
Python
# Yahoo (News)
|
|
#
|
|
# @website https://news.yahoo.com
|
|
# @provide-api yes (https://developer.yahoo.com/boss/search/)
|
|
# $0.80/1000 queries
|
|
#
|
|
# @using-api no (because pricing)
|
|
# @results HTML (using search portal)
|
|
# @stable no (HTML can change)
|
|
# @parse url, title, content, publishedDate
|
|
|
|
from urllib import urlencode
|
|
from lxml import html
|
|
from searx.engines.xpath import extract_text, extract_url
|
|
from searx.engines.yahoo import parse_url
|
|
from datetime import datetime, timedelta
|
|
import re
|
|
from dateutil import parser
|
|
|
|
# engine dependent config
|
|
categories = ['news']
|
|
paging = True
|
|
language_support = True
|
|
|
|
# search-url
|
|
search_url = 'https://news.search.yahoo.com/search?{query}&b={offset}&fl=1&vl=lang_{lang}' # noqa
|
|
|
|
# specific xpath variables
|
|
results_xpath = '//div[@class="res"]'
|
|
url_xpath = './/h3/a/@href'
|
|
title_xpath = './/h3/a'
|
|
content_xpath = './/div[@class="abstr"]'
|
|
publishedDate_xpath = './/span[@class="timestamp"]'
|
|
suggestion_xpath = '//div[@id="satat"]//a'
|
|
|
|
|
|
# do search-request
|
|
def request(query, params):
|
|
offset = (params['pageno'] - 1) * 10 + 1
|
|
|
|
if params['language'] == 'all':
|
|
language = 'en'
|
|
else:
|
|
language = params['language'].split('_')[0]
|
|
|
|
params['url'] = search_url.format(offset=offset,
|
|
query=urlencode({'p': query}),
|
|
lang=language)
|
|
|
|
# TODO required?
|
|
params['cookies']['sB'] = 'fl=1&vl=lang_{lang}&sh=1&rw=new&v=1'\
|
|
.format(lang=language)
|
|
return params
|
|
|
|
|
|
# get response from search-request
|
|
def response(resp):
|
|
results = []
|
|
|
|
dom = html.fromstring(resp.text)
|
|
|
|
# parse results
|
|
for result in dom.xpath(results_xpath):
|
|
url = parse_url(extract_url(result.xpath(url_xpath), search_url))
|
|
title = extract_text(result.xpath(title_xpath)[0])
|
|
content = extract_text(result.xpath(content_xpath)[0])
|
|
|
|
# parse publishedDate
|
|
publishedDate = extract_text(result.xpath(publishedDate_xpath)[0])
|
|
|
|
if re.match("^[0-9]+ minute(s|) ago$", publishedDate):
|
|
publishedDate = datetime.now() - timedelta(minutes=int(re.match(r'\d+', publishedDate).group())) # noqa
|
|
else:
|
|
if 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]))
|
|
else:
|
|
publishedDate = parser.parse(publishedDate)
|
|
|
|
if publishedDate.year == 1900:
|
|
publishedDate = publishedDate.replace(year=datetime.now().year)
|
|
|
|
# append result
|
|
results.append({'url': url,
|
|
'title': title,
|
|
'content': content,
|
|
'publishedDate': publishedDate})
|
|
|
|
# return results
|
|
return results
|