searxng/searx/engines/wolframalpha_noapi.py

96 lines
2.6 KiB
Python
Raw Normal View History

2015-12-30 02:59:51 +00:00
# WolframAlpha (Maths)
#
# @website http://www.wolframalpha.com/
2016-01-02 04:02:10 +00:00
# @provide-api yes (http://api.wolframalpha.com/v2/)
2015-12-30 02:59:51 +00:00
#
# @using-api no
# @results HTML
2015-12-30 02:59:51 +00:00
# @stable no
# @parse answer
from cgi import escape
from json import loads
from time import time
2015-12-30 02:59:51 +00:00
from urllib import urlencode
from searx.poolrequests import get as http_get
2015-12-30 02:59:51 +00:00
# search-url
url = 'https://www.wolframalpha.com/'
2016-01-18 17:34:38 +00:00
search_url = url + 'input/?{query}'
2016-01-02 07:49:32 +00:00
search_url = url + 'input/json.jsp'\
'?async=true'\
'&banners=raw'\
'&debuggingdata=false'\
'&format=image,plaintext,imagemap,minput,moutput'\
'&formattimeout=2'\
'&{query}'\
'&output=JSON'\
'&parsetimeout=2'\
'&proxycode={token}'\
'&scantimeout=0.5'\
'&sponsorcategories=true'\
'&statemethod=deploybutton'
2016-01-02 07:49:32 +00:00
# xpath variables
scripts_xpath = '//script'
title_xpath = '//title'
failure_xpath = '//p[attribute::class="pfail"]'
token = {'value': '',
'last_updated': None}
# seems, wolframalpha resets its token in every hour
def obtain_token():
update_time = time() - (time() % 3600)
try:
token_response = http_get('https://www.wolframalpha.com/input/api/v1/code?ts=9999999999999999999', timeout=2.0)
token['value'] = loads(token_response.text)['code']
token['last_updated'] = update_time
except:
pass
return token
obtain_token()
2015-12-30 02:59:51 +00:00
# do search-request
def request(query, params):
# obtain token if last update was more than an hour
if time() - token['last_updated'] > 3600:
obtain_token()
params['url'] = search_url.format(query=urlencode({'input': query}), token=token['value'])
params['headers']['Referer'] = 'https://www.wolframalpha.com/input/?i=' + query
2015-12-30 02:59:51 +00:00
return params
# get response from search-request
def response(resp):
resp_json = loads(resp.text)
if not resp_json['queryresult']['success']:
return []
# TODO handle resp_json['queryresult']['assumptions']
result_chunks = []
for pod in resp_json['queryresult']['pods']:
pod_title = pod.get('title', '')
if 'subpods' not in pod:
continue
for subpod in pod['subpods']:
if 'img' in subpod:
result_chunks.append(u'<p>{0}<br /><img src="{1}" alt="{2}" /></p>'
.format(escape(pod_title or subpod['img']['alt']),
escape(subpod['img']['src']),
escape(subpod['img']['alt'])))
if not result_chunks:
return []
return [{'url': resp.request.headers['Referer'],
'title': 'Wolframalpha',
'content': ''.join(result_chunks)}]