mirror of
https://github.com/searxng/searxng.git
synced 2024-11-27 13:21:04 +00:00
d6e511fc2f
There is a commented line of a test that I didn't succed to make it work. It's an issue of unicode, utf-8, ascii, latin1... I think I tried everything, but if you have an idea... I'm still a newbie in python...
77 lines
2.2 KiB
Python
77 lines
2.2 KiB
Python
## Twitter (Social media)
|
|
#
|
|
# @website https://twitter.com/
|
|
# @provide-api yes (https://dev.twitter.com/docs/using-search)
|
|
#
|
|
# @using-api no
|
|
# @results HTML (using search portal)
|
|
# @stable no (HTML can change)
|
|
# @parse url, title, content
|
|
#
|
|
# @todo publishedDate
|
|
|
|
from urlparse import urljoin
|
|
from urllib import urlencode
|
|
from lxml import html
|
|
from datetime import datetime
|
|
from searx.engines.xpath import extract_text
|
|
|
|
# engine dependent config
|
|
categories = ['social media']
|
|
language_support = True
|
|
|
|
# search-url
|
|
base_url = 'https://twitter.com/'
|
|
search_url = base_url + 'search?'
|
|
|
|
# specific xpath variables
|
|
results_xpath = '//li[@data-item-type="tweet"]'
|
|
link_xpath = './/small[@class="time"]//a'
|
|
title_xpath = './/span[@class="username js-action-profile-name"]'
|
|
content_xpath = './/p[@class="js-tweet-text tweet-text"]'
|
|
timestamp_xpath = './/span[contains(@class,"_timestamp")]'
|
|
|
|
|
|
# do search-request
|
|
def request(query, params):
|
|
params['url'] = search_url + urlencode({'q': query})
|
|
|
|
# set language if specified
|
|
if params['language'] != 'all':
|
|
params['cookies']['lang'] = params['language'].split('_')[0]
|
|
else:
|
|
params['cookies']['lang'] = 'en'
|
|
|
|
return params
|
|
|
|
|
|
# get response from search-request
|
|
def response(resp):
|
|
results = []
|
|
|
|
dom = html.fromstring(resp.text)
|
|
|
|
# parse results
|
|
for tweet in dom.xpath(results_xpath):
|
|
link = tweet.xpath(link_xpath)[0]
|
|
url = urljoin(base_url, link.attrib.get('href'))
|
|
title = extract_text(tweet.xpath(title_xpath))
|
|
content = extract_text(tweet.xpath(content_xpath)[0])
|
|
|
|
pubdate = tweet.xpath(timestamp_xpath)
|
|
if len(pubdate) > 0:
|
|
timestamp = float(pubdate[0].attrib.get('data-time'))
|
|
publishedDate = datetime.fromtimestamp(timestamp, None)
|
|
# append result
|
|
results.append({'url': url,
|
|
'title': title,
|
|
'content': content,
|
|
'publishedDate': publishedDate})
|
|
else:
|
|
# append result
|
|
results.append({'url': url,
|
|
'title': title,
|
|
'content': content})
|
|
|
|
# return results
|
|
return results
|