searxng/searx/engines/digg.py
Alexandre Flament 78edc16e66 [enh] reduce the number of http outgoing connections.
engines that still use http : gigablast, bing image for thumbnails, 1x and dbpedia autocompleter
2015-05-02 11:43:12 +02:00

73 lines
2.1 KiB
Python

## Digg (News, Social media)
#
# @website https://digg.com/
# @provide-api no
#
# @using-api no
# @results HTML (using search portal)
# @stable no (HTML can change)
# @parse url, title, content, publishedDate, thumbnail
from urllib import quote_plus
from json import loads
from lxml import html
from cgi import escape
from dateutil import parser
# engine dependent config
categories = ['news', 'social media']
paging = True
# search-url
base_url = 'https://digg.com/'
search_url = base_url+'api/search/{query}.json?position={position}&format=html'
# specific xpath variables
results_xpath = '//article'
link_xpath = './/small[@class="time"]//a'
title_xpath = './/h2//a//text()'
content_xpath = './/p//text()'
pubdate_xpath = './/time'
# do search-request
def request(query, params):
offset = (params['pageno'] - 1) * 10
params['url'] = search_url.format(position=offset,
query=quote_plus(query))
return params
# get response from search-request
def response(resp):
results = []
search_result = loads(resp.text)
if 'html' not in search_result or search_result['html'] == '':
return results
dom = html.fromstring(search_result['html'])
# parse results
for result in dom.xpath(results_xpath):
url = result.attrib.get('data-contenturl')
thumbnail = result.xpath('.//img')[0].attrib.get('src')
title = ''.join(result.xpath(title_xpath))
content = escape(''.join(result.xpath(content_xpath)))
pubdate = result.xpath(pubdate_xpath)[0].attrib.get('datetime')
publishedDate = parser.parse(pubdate)
# http to https
thumbnail = thumbnail.replace("http://static.digg.com", "https://static.digg.com")
# append result
results.append({'url': url,
'title': title,
'content': content,
'template': 'videos.html',
'publishedDate': publishedDate,
'thumbnail': thumbnail})
# return results
return results