2022-01-19 17:25:24 +01:00
|
|
|
# SPDX-License-Identifier: AGPL-3.0-or-later
|
|
|
|
# lint: pylint
|
|
|
|
"""This module implements functions needed for the autocompleter.
|
2014-09-13 18:47:28 +02:00
|
|
|
|
2022-01-19 17:25:24 +01:00
|
|
|
"""
|
2023-02-10 13:40:12 +01:00
|
|
|
# pylint: disable=use-dict-literal
|
2014-09-13 18:47:28 +02:00
|
|
|
|
2022-12-04 22:57:22 +01:00
|
|
|
import json
|
2020-08-06 17:42:46 +02:00
|
|
|
from urllib.parse import urlencode
|
|
|
|
|
2022-12-04 22:57:22 +01:00
|
|
|
import lxml
|
2021-03-18 19:59:01 +01:00
|
|
|
from httpx import HTTPError
|
|
|
|
|
2015-04-10 00:59:25 +02:00
|
|
|
from searx import settings
|
2022-12-04 22:57:22 +01:00
|
|
|
from searx.engines import (
|
|
|
|
engines,
|
|
|
|
google,
|
|
|
|
)
|
[httpx] replace searx.poolrequests by searx.network
settings.yml:
* outgoing.networks:
* can contains network definition
* propertiers: enable_http, verify, http2, max_connections, max_keepalive_connections,
keepalive_expiry, local_addresses, support_ipv4, support_ipv6, proxies, max_redirects, retries
* retries: 0 by default, number of times searx retries to send the HTTP request (using different IP & proxy each time)
* local_addresses can be "192.168.0.1/24" (it supports IPv6)
* support_ipv4 & support_ipv6: both True by default
see https://github.com/searx/searx/pull/1034
* each engine can define a "network" section:
* either a full network description
* either reference an existing network
* all HTTP requests of engine use the same HTTP configuration (it was not the case before, see proxy configuration in master)
2021-04-05 10:43:33 +02:00
|
|
|
from searx.network import get as http_get
|
2021-02-22 18:13:50 +01:00
|
|
|
from searx.exceptions import SearxEngineResponseException
|
2018-01-19 03:51:27 +01:00
|
|
|
|
2015-04-10 00:59:25 +02:00
|
|
|
|
|
|
|
def get(*args, **kwargs):
|
2015-05-02 15:45:17 +02:00
|
|
|
if 'timeout' not in kwargs:
|
2015-08-02 19:38:27 +02:00
|
|
|
kwargs['timeout'] = settings['outgoing']['request_timeout']
|
2021-02-22 18:13:50 +01:00
|
|
|
kwargs['raise_for_httperror'] = True
|
2015-04-10 00:59:25 +02:00
|
|
|
return http_get(*args, **kwargs)
|
2015-01-10 16:42:57 +01:00
|
|
|
|
|
|
|
|
2022-01-19 17:25:24 +01:00
|
|
|
def brave(query, _lang):
|
2022-01-02 23:05:49 +01:00
|
|
|
# brave search autocompleter
|
2022-01-19 17:25:24 +01:00
|
|
|
url = 'https://search.brave.com/api/suggest?'
|
|
|
|
url += urlencode({'q': query})
|
|
|
|
country = 'all'
|
|
|
|
# if lang in _brave:
|
|
|
|
# country = lang
|
|
|
|
kwargs = {'cookies': {'country': country}}
|
|
|
|
resp = get(url, **kwargs)
|
2022-01-02 23:05:49 +01:00
|
|
|
|
|
|
|
results = []
|
|
|
|
|
|
|
|
if resp.ok:
|
2022-01-23 17:22:13 +01:00
|
|
|
data = resp.json()
|
2022-01-02 23:05:49 +01:00
|
|
|
for item in data[1]:
|
|
|
|
results.append(item)
|
|
|
|
return results
|
|
|
|
|
|
|
|
|
2022-01-19 17:25:24 +01:00
|
|
|
def dbpedia(query, _lang):
|
2015-05-02 11:43:12 +02:00
|
|
|
# dbpedia autocompleter, no HTTPS
|
2020-12-04 16:47:43 +01:00
|
|
|
autocomplete_url = 'https://lookup.dbpedia.org/api/search.asmx/KeywordSearch?'
|
2014-03-29 16:30:49 +01:00
|
|
|
|
2016-01-18 12:47:31 +01:00
|
|
|
response = get(autocomplete_url + urlencode(dict(QueryString=query)))
|
2014-03-29 16:30:49 +01:00
|
|
|
|
|
|
|
results = []
|
|
|
|
|
|
|
|
if response.ok:
|
2022-12-04 22:57:22 +01:00
|
|
|
dom = lxml.etree.fromstring(response.content)
|
2020-12-04 16:47:43 +01:00
|
|
|
results = dom.xpath('//Result/Label//text()')
|
2014-03-29 16:30:49 +01:00
|
|
|
|
|
|
|
return results
|
|
|
|
|
|
|
|
|
2022-11-05 15:10:52 +01:00
|
|
|
def duckduckgo(query, sxng_locale):
|
|
|
|
"""Autocomplete from DuckDuckGo. Supports DuckDuckGo's languages"""
|
|
|
|
|
|
|
|
traits = engines['duckduckgo'].traits
|
|
|
|
args = {
|
|
|
|
'q': query,
|
|
|
|
'kl': traits.get_region(sxng_locale, traits.all_locale),
|
|
|
|
}
|
|
|
|
|
|
|
|
url = 'https://duckduckgo.com/ac/?type=list&' + urlencode(args)
|
|
|
|
resp = get(url)
|
|
|
|
|
|
|
|
ret_val = []
|
|
|
|
if resp.ok:
|
|
|
|
j = resp.json()
|
|
|
|
if len(j) > 1:
|
|
|
|
ret_val = j[1]
|
|
|
|
return ret_val
|
2014-09-07 23:56:06 +02:00
|
|
|
|
|
|
|
|
2022-12-04 22:57:22 +01:00
|
|
|
def google_complete(query, sxng_locale):
|
|
|
|
"""Autocomplete from Google. Supports Google's languages and subdomains
|
|
|
|
(:py:obj:`searx.engines.google.get_google_info`) by using the async REST
|
|
|
|
API::
|
2014-03-29 16:30:49 +01:00
|
|
|
|
2022-12-04 22:57:22 +01:00
|
|
|
https://{subdomain}/complete/search?{args}
|
2014-03-29 16:30:49 +01:00
|
|
|
|
2022-12-04 22:57:22 +01:00
|
|
|
"""
|
2014-03-29 16:30:49 +01:00
|
|
|
|
2022-12-04 22:57:22 +01:00
|
|
|
google_info = google.get_google_info({'searxng_locale': sxng_locale}, engines['google'].traits)
|
2014-03-29 16:30:49 +01:00
|
|
|
|
2022-12-04 22:57:22 +01:00
|
|
|
url = 'https://{subdomain}/complete/search?{args}'
|
|
|
|
args = urlencode(
|
|
|
|
{
|
|
|
|
'q': query,
|
|
|
|
'client': 'gws-wiz',
|
|
|
|
'hl': google_info['params']['hl'],
|
|
|
|
}
|
|
|
|
)
|
|
|
|
results = []
|
|
|
|
resp = get(url.format(subdomain=google_info['subdomain'], args=args))
|
|
|
|
if resp.ok:
|
|
|
|
json_txt = resp.text[resp.text.find('[') : resp.text.find(']', -3) + 1]
|
|
|
|
data = json.loads(json_txt)
|
|
|
|
for item in data[0]:
|
|
|
|
results.append(lxml.html.fromstring(item[0]).text_content())
|
2014-03-29 16:30:49 +01:00
|
|
|
return results
|
|
|
|
|
|
|
|
|
2022-04-14 03:02:05 +02:00
|
|
|
def seznam(query, _lang):
|
|
|
|
# seznam search autocompleter
|
|
|
|
url = 'https://suggest.seznam.cz/fulltext/cs?{query}'
|
|
|
|
|
2022-05-07 18:23:10 +02:00
|
|
|
resp = get(
|
|
|
|
url.format(
|
|
|
|
query=urlencode(
|
|
|
|
{'phrase': query, 'cursorPosition': len(query), 'format': 'json-2', 'highlight': '1', 'count': '6'}
|
|
|
|
)
|
|
|
|
)
|
|
|
|
)
|
2022-04-14 03:02:05 +02:00
|
|
|
|
|
|
|
if not resp.ok:
|
|
|
|
return []
|
|
|
|
|
|
|
|
data = resp.json()
|
2022-05-07 18:23:10 +02:00
|
|
|
return [
|
|
|
|
''.join([part.get('text', '') for part in item.get('text', [])])
|
|
|
|
for item in data.get('result', [])
|
|
|
|
if item.get('itemType', None) == 'ItemType.TEXT'
|
|
|
|
]
|
|
|
|
|
2022-04-14 03:02:05 +02:00
|
|
|
|
2022-10-30 11:23:20 +01:00
|
|
|
def startpage(query, sxng_locale):
|
|
|
|
"""Autocomplete from Startpage. Supports Startpage's languages"""
|
|
|
|
lui = engines['startpage'].traits.get_language(sxng_locale, 'english')
|
2021-11-13 13:26:47 +01:00
|
|
|
url = 'https://startpage.com/suggestions?{query}'
|
|
|
|
resp = get(url.format(query=urlencode({'q': query, 'segment': 'startpage.udog', 'lui': lui})))
|
|
|
|
data = resp.json()
|
|
|
|
return [e['text'] for e in data.get('suggestions', []) if 'text' in e]
|
2015-06-01 20:45:18 +02:00
|
|
|
|
|
|
|
|
2022-01-19 17:25:24 +01:00
|
|
|
def swisscows(query, _lang):
|
2020-02-14 19:19:24 +01:00
|
|
|
# swisscows autocompleter
|
|
|
|
url = 'https://swisscows.ch/api/suggest?{query}&itemsCount=5'
|
|
|
|
|
2022-12-04 22:57:22 +01:00
|
|
|
resp = json.loads(get(url.format(query=urlencode({'query': query}))).text)
|
2020-02-14 19:19:24 +01:00
|
|
|
return resp
|
|
|
|
|
|
|
|
|
2022-10-03 22:42:58 +02:00
|
|
|
def qwant(query, sxng_locale):
|
|
|
|
"""Autocomplete from Qwant. Supports Qwant's regions."""
|
2016-03-02 12:54:06 +01:00
|
|
|
results = []
|
|
|
|
|
2022-10-03 22:42:58 +02:00
|
|
|
locale = engines['qwant'].traits.get_region(sxng_locale, 'en_US')
|
|
|
|
url = 'https://api.qwant.com/v3/suggest?{query}'
|
|
|
|
resp = get(url.format(query=urlencode({'q': query, 'locale': locale, 'version': '2'})))
|
|
|
|
|
2016-03-02 12:54:06 +01:00
|
|
|
if resp.ok:
|
2022-10-03 22:42:58 +02:00
|
|
|
data = resp.json()
|
2016-03-02 12:54:06 +01:00
|
|
|
if data['status'] == 'success':
|
|
|
|
for item in data['data']['items']:
|
|
|
|
results.append(item['value'])
|
|
|
|
|
|
|
|
return results
|
|
|
|
|
|
|
|
|
2022-10-28 19:12:59 +02:00
|
|
|
def wikipedia(query, sxng_locale):
|
|
|
|
"""Autocomplete from Wikipedia. Supports Wikipedia's languages (aka netloc)."""
|
|
|
|
results = []
|
|
|
|
eng_traits = engines['wikipedia'].traits
|
|
|
|
wiki_lang = eng_traits.get_language(sxng_locale, 'en')
|
|
|
|
wiki_netloc = eng_traits.custom['wiki_netloc'].get(wiki_lang, 'en.wikipedia.org')
|
|
|
|
|
|
|
|
url = 'https://{wiki_netloc}/w/api.php?{args}'
|
|
|
|
args = urlencode(
|
|
|
|
{
|
|
|
|
'action': 'opensearch',
|
|
|
|
'format': 'json',
|
|
|
|
'formatversion': '2',
|
|
|
|
'search': query,
|
|
|
|
'namespace': '0',
|
|
|
|
'limit': '10',
|
|
|
|
}
|
|
|
|
)
|
|
|
|
resp = get(url.format(args=args, wiki_netloc=wiki_netloc))
|
|
|
|
if resp.ok:
|
|
|
|
data = resp.json()
|
|
|
|
if len(data) > 1:
|
|
|
|
results = data[1]
|
2014-03-29 16:30:49 +01:00
|
|
|
|
2022-10-28 19:12:59 +02:00
|
|
|
return results
|
2014-03-29 16:30:49 +01:00
|
|
|
|
|
|
|
|
2022-09-09 22:42:44 +02:00
|
|
|
def yandex(query, _lang):
|
|
|
|
# yandex autocompleter
|
|
|
|
url = "https://suggest.yandex.com/suggest-ff.cgi?{0}"
|
|
|
|
|
2022-12-04 22:57:22 +01:00
|
|
|
resp = json.loads(get(url.format(urlencode(dict(part=query)))).text)
|
2022-09-09 22:42:44 +02:00
|
|
|
if len(resp) > 1:
|
|
|
|
return resp[1]
|
|
|
|
return []
|
|
|
|
|
|
|
|
|
2021-12-27 09:26:22 +01:00
|
|
|
backends = {
|
|
|
|
'dbpedia': dbpedia,
|
|
|
|
'duckduckgo': duckduckgo,
|
2022-12-04 22:57:22 +01:00
|
|
|
'google': google_complete,
|
2022-04-14 03:02:05 +02:00
|
|
|
'seznam': seznam,
|
2021-12-27 09:26:22 +01:00
|
|
|
'startpage': startpage,
|
|
|
|
'swisscows': swisscows,
|
|
|
|
'qwant': qwant,
|
|
|
|
'wikipedia': wikipedia,
|
2022-01-02 23:05:49 +01:00
|
|
|
'brave': brave,
|
2022-09-09 22:42:44 +02:00
|
|
|
'yandex': yandex,
|
2021-12-27 09:26:22 +01:00
|
|
|
}
|
2021-02-22 18:13:50 +01:00
|
|
|
|
|
|
|
|
2022-09-29 20:54:46 +02:00
|
|
|
def search_autocomplete(backend_name, query, sxng_locale):
|
2021-02-22 18:13:50 +01:00
|
|
|
backend = backends.get(backend_name)
|
|
|
|
if backend is None:
|
|
|
|
return []
|
|
|
|
try:
|
2022-09-29 20:54:46 +02:00
|
|
|
return backend(query, sxng_locale)
|
2021-03-18 19:59:01 +01:00
|
|
|
except (HTTPError, SearxEngineResponseException):
|
2021-02-22 18:13:50 +01:00
|
|
|
return []
|