mirror of
https://github.com/searxng/searxng.git
synced 2024-06-11 01:39:35 +00:00
86b4d2f2d0
We have been using a static type checker (pyright) for a long time, but its check was not yet a prerequisite for passing the quality gate. It was checked in the CI, but the error messages were only logged. As is always the case in life, with checks that you have to do but which have no consequences; you neglect them :-) We didn't activate the checks back then because we (even today) have too much monkey patching in our code (not only in the engines, httpx and others objects are also affected). We want to replace monkey patching with clear interfaces for a long time, the basis for this is increased typing and we can only achieve this if we make type checking an integral part of the quality gate. This PR activates the type check; in order to pass the check, a few typings were corrected in the code, but most type inconsistencies were deactivated via inline comments. This was particularly necessary in places where the code uses properties that stick to the objects (monkey patching). The sticking of properties only happens in a few places, but the access to these properties extends over the entire code, which is why there are many `# type: ignore` markers in the code ... which we will hopefully be able to remove again successively in the future. Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
263 lines
7.3 KiB
Python
263 lines
7.3 KiB
Python
# SPDX-License-Identifier: AGPL-3.0-or-later
|
|
"""This module implements functions needed for the autocompleter.
|
|
|
|
"""
|
|
# pylint: disable=use-dict-literal
|
|
|
|
import json
|
|
from urllib.parse import urlencode, quote_plus
|
|
|
|
import lxml
|
|
from httpx import HTTPError
|
|
|
|
from searx import settings
|
|
from searx.engines import (
|
|
engines,
|
|
google,
|
|
)
|
|
from searx.network import get as http_get, post as http_post
|
|
from searx.exceptions import SearxEngineResponseException
|
|
|
|
|
|
def update_kwargs(**kwargs):
|
|
if 'timeout' not in kwargs:
|
|
kwargs['timeout'] = settings['outgoing']['request_timeout']
|
|
kwargs['raise_for_httperror'] = True
|
|
|
|
|
|
def get(*args, **kwargs):
|
|
update_kwargs(**kwargs)
|
|
return http_get(*args, **kwargs)
|
|
|
|
|
|
def post(*args, **kwargs):
|
|
update_kwargs(**kwargs)
|
|
return http_post(*args, **kwargs)
|
|
|
|
|
|
def brave(query, _lang):
|
|
# brave search autocompleter
|
|
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)
|
|
|
|
results = []
|
|
|
|
if resp.ok: # type: ignore
|
|
data = resp.json() # type: ignore
|
|
for item in data[1]:
|
|
results.append(item)
|
|
return results
|
|
|
|
|
|
def dbpedia(query, _lang):
|
|
# dbpedia autocompleter, no HTTPS
|
|
autocomplete_url = 'https://lookup.dbpedia.org/api/search.asmx/KeywordSearch?'
|
|
|
|
response = get(autocomplete_url + urlencode(dict(QueryString=query)))
|
|
|
|
results = []
|
|
|
|
if response.ok: # type: ignore
|
|
dom = lxml.etree.fromstring(response.content) # type: ignore
|
|
results = dom.xpath('//Result/Label//text()')
|
|
|
|
return results
|
|
|
|
|
|
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: # type: ignore
|
|
j = resp.json() # type: ignore
|
|
if len(j) > 1:
|
|
ret_val = j[1]
|
|
return ret_val
|
|
|
|
|
|
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::
|
|
|
|
https://{subdomain}/complete/search?{args}
|
|
|
|
"""
|
|
|
|
google_info = google.get_google_info({'searxng_locale': sxng_locale}, engines['google'].traits)
|
|
|
|
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: # type: ignore
|
|
json_txt = resp.text[resp.text.find('[') : resp.text.find(']', -3) + 1] # type: ignore
|
|
data = json.loads(json_txt)
|
|
for item in data[0]:
|
|
results.append(lxml.html.fromstring(item[0]).text_content()) # type: ignore
|
|
return results
|
|
|
|
|
|
def mwmbl(query, _lang):
|
|
"""Autocomplete from Mwmbl_."""
|
|
|
|
# mwmbl autocompleter
|
|
url = 'https://api.mwmbl.org/search/complete?{query}'
|
|
|
|
results = get(url.format(query=urlencode({'q': query}))).json()[1] # type: ignore
|
|
|
|
# results starting with `go:` are direct urls and not useful for auto completion
|
|
return [result for result in results if not result.startswith("go: ") and not result.startswith("search: ")]
|
|
|
|
|
|
def seznam(query, _lang):
|
|
# seznam search autocompleter
|
|
url = 'https://suggest.seznam.cz/fulltext/cs?{query}'
|
|
|
|
resp = get(
|
|
url.format(
|
|
query=urlencode(
|
|
{'phrase': query, 'cursorPosition': len(query), 'format': 'json-2', 'highlight': '1', 'count': '6'}
|
|
)
|
|
)
|
|
)
|
|
|
|
if not resp.ok: # type: ignore
|
|
return []
|
|
|
|
data = resp.json() # type: ignore
|
|
return [
|
|
''.join([part.get('text', '') for part in item.get('text', [])])
|
|
for item in data.get('result', [])
|
|
if item.get('itemType', None) == 'ItemType.TEXT'
|
|
]
|
|
|
|
|
|
def stract(query, _lang):
|
|
# stract autocompleter (beta)
|
|
url = f"https://stract.com/beta/api/autosuggest?q={quote_plus(query)}"
|
|
|
|
resp = post(url)
|
|
|
|
if not resp.ok: # type: ignore
|
|
return []
|
|
|
|
return [suggestion['raw'] for suggestion in resp.json()] # type: ignore
|
|
|
|
|
|
def startpage(query, sxng_locale):
|
|
"""Autocomplete from Startpage. Supports Startpage's languages"""
|
|
lui = engines['startpage'].traits.get_language(sxng_locale, 'english')
|
|
url = 'https://startpage.com/suggestions?{query}'
|
|
resp = get(url.format(query=urlencode({'q': query, 'segment': 'startpage.udog', 'lui': lui})))
|
|
data = resp.json() # type: ignore
|
|
return [e['text'] for e in data.get('suggestions', []) if 'text' in e]
|
|
|
|
|
|
def swisscows(query, _lang):
|
|
# swisscows autocompleter
|
|
url = 'https://swisscows.ch/api/suggest?{query}&itemsCount=5'
|
|
|
|
resp = json.loads(get(url.format(query=urlencode({'query': query}))).text) # type: ignore
|
|
return resp
|
|
|
|
|
|
def qwant(query, sxng_locale):
|
|
"""Autocomplete from Qwant. Supports Qwant's regions."""
|
|
results = []
|
|
|
|
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'})))
|
|
|
|
if resp.ok: # type: ignore
|
|
data = resp.json() # type: ignore
|
|
if data['status'] == 'success':
|
|
for item in data['data']['items']:
|
|
results.append(item['value'])
|
|
|
|
return results
|
|
|
|
|
|
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') # type: ignore
|
|
|
|
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: # type: ignore
|
|
data = resp.json() # type: ignore
|
|
if len(data) > 1:
|
|
results = data[1]
|
|
|
|
return results
|
|
|
|
|
|
def yandex(query, _lang):
|
|
# yandex autocompleter
|
|
url = "https://suggest.yandex.com/suggest-ff.cgi?{0}"
|
|
|
|
resp = json.loads(get(url.format(urlencode(dict(part=query)))).text) # type: ignore
|
|
if len(resp) > 1:
|
|
return resp[1]
|
|
return []
|
|
|
|
|
|
backends = {
|
|
'dbpedia': dbpedia,
|
|
'duckduckgo': duckduckgo,
|
|
'google': google_complete,
|
|
'mwmbl': mwmbl,
|
|
'seznam': seznam,
|
|
'startpage': startpage,
|
|
'stract': stract,
|
|
'swisscows': swisscows,
|
|
'qwant': qwant,
|
|
'wikipedia': wikipedia,
|
|
'brave': brave,
|
|
'yandex': yandex,
|
|
}
|
|
|
|
|
|
def search_autocomplete(backend_name, query, sxng_locale):
|
|
backend = backends.get(backend_name)
|
|
if backend is None:
|
|
return []
|
|
try:
|
|
return backend(query, sxng_locale)
|
|
except (HTTPError, SearxEngineResponseException):
|
|
return []
|