searxng/searx/engines/wolframalpha_noapi.py

123 lines
3.5 KiB
Python
Raw Normal View History

# Wolfram|Alpha (Science)
2015-12-30 02:59:51 +00:00
#
# @website https://www.wolframalpha.com/
# @provide-api yes (https://api.wolframalpha.com/v2/)
2015-12-30 02:59:51 +00:00
#
# @using-api no
# @results JSON
2015-12-30 02:59:51 +00:00
# @stable no
# @parse url, infobox
2015-12-30 02:59:51 +00:00
2016-07-07 23:41:33 +00:00
from cgi import escape
from json import loads
from time import time
2015-12-30 02:59:51 +00:00
from urllib import urlencode
2016-07-07 23:41:33 +00:00
from lxml.etree import XML
from searx.poolrequests import get as http_get
2015-12-30 02:59:51 +00:00
# search-url
url = 'https://www.wolframalpha.com/'
2016-01-02 07:49:32 +00:00
search_url = url + 'input/json.jsp'\
'?async=false'\
'&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'
referer_url = url + 'input/?{query}'
token = {'value': '',
2016-07-07 23:41:33 +00:00
'last_updated': None}
# pods to display as image in infobox
# this pods do return a plaintext, but they look better and are more useful as images
2016-02-28 06:47:36 +00:00
image_pods = {'VisualRepresentation',
'Illustration',
'Symbol'}
# 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'] = referer_url.format(query=urlencode({'i': query}))
2015-12-30 02:59:51 +00:00
return params
# get response from search-request
def response(resp):
results = []
resp_json = loads(resp.text)
if not resp_json['queryresult']['success']:
return []
# TODO handle resp_json['queryresult']['assumptions']
result_chunks = []
2016-07-07 23:41:33 +00:00
infobox_title = ""
result_content = ""
for pod in resp_json['queryresult']['pods']:
2016-02-28 06:47:36 +00:00
pod_id = pod.get('id', '')
pod_title = pod.get('title', '')
pod_is_result = pod.get('primary', None)
if 'subpods' not in pod:
continue
2016-02-28 06:47:36 +00:00
if pod_id == 'Input' or not infobox_title:
infobox_title = pod['subpods'][0]['plaintext']
for subpod in pod['subpods']:
2016-02-28 06:47:36 +00:00
if subpod['plaintext'] != '' and pod_id not in image_pods:
# append unless it's not an actual answer
if subpod['plaintext'] != '(requires interactivity)':
result_chunks.append({'label': pod_title, 'value': subpod['plaintext']})
2016-07-07 23:41:33 +00:00
if pod_is_result or not result_content:
if pod_id != "Input":
result_content = pod_title + ': ' + subpod['plaintext']
elif 'img' in subpod:
result_chunks.append({'label': pod_title, 'image': subpod['img']})
if not result_chunks:
return []
results.append({'infobox': infobox_title,
'attributes': result_chunks,
'urls': [{'title': 'Wolfram|Alpha', 'url': resp.request.headers['Referer'].decode('utf8')}]})
results.append({'url': resp.request.headers['Referer'].decode('utf8'),
2016-07-07 23:41:33 +00:00
'title': 'Wolfram|Alpha (' + infobox_title + ')',
'content': result_content})
return results