searxng/searx/engines/ina.py

87 lines
2.6 KiB
Python
Raw Normal View History

2016-07-09 15:44:27 +00:00
# INA (Videos)
#
# @website https://www.ina.fr/
# @provide-api no
#
# @using-api no
# @results HTML (using search portal)
# @stable no (HTML can change)
# @parse url, title, content, publishedDate, thumbnail
#
# @todo set content-parameter with correct data
# @todo embedded (needs some md5 from video page)
from json import loads
from html import unescape
from urllib.parse import urlencode
2016-07-09 15:44:27 +00:00
from lxml import html
from dateutil import parser
from searx.utils import extract_text
2016-11-30 17:43:03 +00:00
2016-07-09 15:44:27 +00:00
# engine dependent config
categories = ['videos']
paging = True
page_size = 48
# search-url
base_url = 'https://www.ina.fr'
search_url = base_url + '/layout/set/ajax/recherche/result?autopromote=&hf={ps}&b={start}&type=Video&r=&{query}'
# specific xpath variables
2020-01-02 21:29:28 +00:00
results_xpath = '//div[contains(@class,"search-results--list")]//div[@class="media-body"]'
2016-07-09 15:44:27 +00:00
url_xpath = './/a/@href'
title_xpath = './/h3[@class="h3--title media-heading"]'
thumbnail_xpath = './/img/@src'
publishedDate_xpath = './/span[@class="broadcast"]'
content_xpath = './/p[@class="media-body__summary"]'
# do search-request
def request(query, params):
params['url'] = search_url.format(ps=page_size,
start=params['pageno'] * page_size,
query=urlencode({'q': query}))
return params
# get response from search-request
def response(resp):
results = []
# we get html in a JSON container...
response = loads(resp.text)
if "content" not in response:
return []
dom = html.fromstring(response["content"])
# parse results
for result in dom.xpath(results_xpath):
videoid = result.xpath(url_xpath)[0]
url = base_url + videoid
title = unescape(extract_text(result.xpath(title_xpath)))
2020-01-02 21:29:28 +00:00
try:
thumbnail = extract_text(result.xpath(thumbnail_xpath)[0])
except:
thumbnail = ''
if thumbnail and thumbnail[0] == '/':
2016-07-09 15:44:27 +00:00
thumbnail = base_url + thumbnail
d = extract_text(result.xpath(publishedDate_xpath)[0])
d = d.split('/')
# force ISO date to avoid wrong parsing
d = "%s-%s-%s" % (d[2], d[1], d[0])
publishedDate = parser.parse(d)
content = extract_text(result.xpath(content_xpath))
# append result
results.append({'url': url,
'title': title,
'content': content,
'template': 'videos.html',
'publishedDate': publishedDate,
'thumbnail': thumbnail})
# return results
return results