Ponysearch/searx/search.py
Alexandre Flament f0054d67f1 [fix] wikipedia engine: don't raise an error when the query is not found
Add a new parameter "raise_for_status", set by default to True.
When True, any HTTP status code >= 300 raise an exception ( #2332 )
When False, the engine can manage the HTTP status code by itself.
2020-12-04 20:04:39 +01:00

529 lines
19 KiB
Python

'''
searx is free software: you can redistribute it and/or modify
it under the terms of the GNU Affero General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
searx is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU Affero General Public License for more details.
You should have received a copy of the GNU Affero General Public License
along with searx. If not, see < http://www.gnu.org/licenses/ >.
(C) 2013- by Adam Tauber, <asciimoo@gmail.com>
'''
import typing
import gc
import threading
from time import time
from uuid import uuid4
from urllib.parse import urlparse
from _thread import start_new_thread
import requests.exceptions
import searx.poolrequests as requests_lib
from searx.engines import engines, settings
from searx.answerers import ask
from searx.external_bang import get_bang_url
from searx.utils import gen_useragent
from searx.results import ResultContainer
from searx import logger
from searx.plugins import plugins
from searx.exceptions import SearxEngineCaptchaException
from searx.metrology.error_recorder import record_exception, record_error
logger = logger.getChild('search')
max_request_timeout = settings.get('outgoing', {}).get('max_request_timeout' or None)
if max_request_timeout is None:
logger.info('max_request_timeout={0}'.format(max_request_timeout))
else:
if isinstance(max_request_timeout, float):
logger.info('max_request_timeout={0} second(s)'.format(max_request_timeout))
else:
logger.critical('outgoing.max_request_timeout if defined has to be float')
import sys
sys.exit(1)
class EngineRef:
__slots__ = 'name', 'category', 'from_bang'
def __init__(self, name: str, category: str, from_bang: bool=False):
self.name = name
self.category = category
self.from_bang = from_bang
def __repr__(self):
return "EngineRef({!r}, {!r}, {!r})".format(self.name, self.category, self.from_bang)
def __eq__(self, other):
return self.name == other.name and self.category == other.category and self.from_bang == other.from_bang
class SearchQuery:
"""container for all the search parameters (query, language, etc...)"""
__slots__ = 'query', 'engineref_list', 'categories', 'lang', 'safesearch', 'pageno', 'time_range',\
'timeout_limit', 'external_bang'
def __init__(self,
query: str,
engineref_list: typing.List[EngineRef],
categories: typing.List[str],
lang: str,
safesearch: int,
pageno: int,
time_range: typing.Optional[str],
timeout_limit: typing.Optional[float]=None,
external_bang: typing.Optional[str]=None):
self.query = query
self.engineref_list = engineref_list
self.categories = categories
self.lang = lang
self.safesearch = safesearch
self.pageno = pageno
self.time_range = time_range
self.timeout_limit = timeout_limit
self.external_bang = external_bang
def __repr__(self):
return "SearchQuery({!r}, {!r}, {!r}, {!r}, {!r}, {!r}, {!r}, {!r}, {!r})".\
format(self.query, self.engineref_list, self.categories, self.lang, self.safesearch,
self.pageno, self.time_range, self.timeout_limit, self.external_bang)
def __eq__(self, other):
return self.query == other.query\
and self.engineref_list == other.engineref_list\
and self.categories == self.categories\
and self.lang == other.lang\
and self.safesearch == other.safesearch\
and self.pageno == other.pageno\
and self.time_range == other.time_range\
and self.timeout_limit == other.timeout_limit\
and self.external_bang == other.external_bang
def send_http_request(engine, request_params):
# create dictionary which contain all
# informations about the request
request_args = dict(
headers=request_params['headers'],
cookies=request_params['cookies'],
verify=request_params['verify'],
auth=request_params['auth']
)
# setting engine based proxies
if hasattr(engine, 'proxies'):
request_args['proxies'] = requests_lib.get_proxies(engine.proxies)
# max_redirects
max_redirects = request_params.get('max_redirects')
if max_redirects:
request_args['max_redirects'] = max_redirects
# soft_max_redirects
soft_max_redirects = request_params.get('soft_max_redirects', max_redirects or 0)
# specific type of request (GET or POST)
if request_params['method'] == 'GET':
req = requests_lib.get
else:
req = requests_lib.post
request_args['data'] = request_params['data']
# send the request
response = req(request_params['url'], **request_args)
# check HTTP status
if request_params.get('raise_for_status'):
response.raise_for_status()
# check soft limit of the redirect count
if len(response.history) > soft_max_redirects:
# unexpected redirect : record an error
# but the engine might still return valid results.
status_code = str(response.status_code or '')
reason = response.reason or ''
hostname = str(urlparse(response.url or '').netloc)
record_error(engine.name,
'{} redirects, maximum: {}'.format(len(response.history), soft_max_redirects),
(status_code, reason, hostname))
return response
def search_one_http_request(engine, query, request_params):
# update request parameters dependent on
# search-engine (contained in engines folder)
engine.request(query, request_params)
# ignoring empty urls
if request_params['url'] is None:
return None
if not request_params['url']:
return None
# send request
response = send_http_request(engine, request_params)
# parse the response
response.search_params = request_params
return engine.response(response)
def search_one_http_request_safe(engine_name, query, request_params, result_container, start_time, timeout_limit):
# set timeout for all HTTP requests
requests_lib.set_timeout_for_thread(timeout_limit, start_time=start_time)
# reset the HTTP total time
requests_lib.reset_time_for_thread()
#
engine = engines[engine_name]
# suppose everything will be alright
requests_exception = False
try:
# send requests and parse the results
search_results = search_one_http_request(engine, query, request_params)
# check if the engine accepted the request
if search_results is not None:
# yes, so add results
result_container.extend(engine_name, search_results)
# update engine time when there is no exception
engine_time = time() - start_time
page_load_time = requests_lib.get_time_for_thread()
result_container.add_timing(engine_name, engine_time, page_load_time)
with threading.RLock():
engine.stats['engine_time'] += engine_time
engine.stats['engine_time_count'] += 1
# update stats with the total HTTP time
engine.stats['page_load_time'] += page_load_time
engine.stats['page_load_count'] += 1
except Exception as e:
record_exception(engine_name, e)
# Timing
engine_time = time() - start_time
page_load_time = requests_lib.get_time_for_thread()
result_container.add_timing(engine_name, engine_time, page_load_time)
# Record the errors
with threading.RLock():
engine.stats['errors'] += 1
if (issubclass(e.__class__, requests.exceptions.Timeout)):
result_container.add_unresponsive_engine(engine_name, 'HTTP timeout')
# requests timeout (connect or read)
logger.error("engine {0} : HTTP requests timeout"
"(search duration : {1} s, timeout: {2} s) : {3}"
.format(engine_name, engine_time, timeout_limit, e.__class__.__name__))
requests_exception = True
elif (issubclass(e.__class__, requests.exceptions.RequestException)):
result_container.add_unresponsive_engine(engine_name, 'HTTP error')
# other requests exception
logger.exception("engine {0} : requests exception"
"(search duration : {1} s, timeout: {2} s) : {3}"
.format(engine_name, engine_time, timeout_limit, e))
requests_exception = True
elif (issubclass(e.__class__, SearxEngineCaptchaException)):
result_container.add_unresponsive_engine(engine_name, 'CAPTCHA required')
logger.exception('engine {0} : CAPTCHA')
else:
result_container.add_unresponsive_engine(engine_name, 'unexpected crash')
# others errors
logger.exception('engine {0} : exception : {1}'.format(engine_name, e))
else:
if getattr(threading.current_thread(), '_timeout', False):
record_error(engine_name, 'Timeout')
# suspend or not the engine if there are HTTP errors
with threading.RLock():
if requests_exception:
# update continuous_errors / suspend_end_time
engine.continuous_errors += 1
engine.suspend_end_time = time() + min(settings['search']['max_ban_time_on_fail'],
engine.continuous_errors * settings['search']['ban_time_on_fail'])
else:
# no HTTP error (perhaps an engine error)
# anyway, reset the suspend variables
engine.continuous_errors = 0
engine.suspend_end_time = 0
def record_offline_engine_stats_on_error(engine, result_container, start_time):
engine_time = time() - start_time
result_container.add_timing(engine.name, engine_time, engine_time)
with threading.RLock():
engine.stats['errors'] += 1
def search_one_offline_request(engine, query, request_params):
return engine.search(query, request_params)
def search_one_offline_request_safe(engine_name, query, request_params, result_container, start_time, timeout_limit):
engine = engines[engine_name]
try:
search_results = search_one_offline_request(engine, query, request_params)
if search_results:
result_container.extend(engine_name, search_results)
engine_time = time() - start_time
result_container.add_timing(engine_name, engine_time, engine_time)
with threading.RLock():
engine.stats['engine_time'] += engine_time
engine.stats['engine_time_count'] += 1
except ValueError as e:
record_exception(engine_name, e)
record_offline_engine_stats_on_error(engine, result_container, start_time)
logger.exception('engine {0} : invalid input : {1}'.format(engine_name, e))
except Exception as e:
record_exception(engine_name, e)
record_offline_engine_stats_on_error(engine, result_container, start_time)
result_container.add_unresponsive_engine(engine_name, 'unexpected crash', str(e))
logger.exception('engine {0} : exception : {1}'.format(engine_name, e))
else:
if getattr(threading.current_thread(), '_timeout', False):
record_error(engine_name, 'Timeout')
def search_one_request_safe(engine_name, query, request_params, result_container, start_time, timeout_limit):
if engines[engine_name].offline:
return search_one_offline_request_safe(engine_name, query, request_params, result_container, start_time, timeout_limit) # noqa
return search_one_http_request_safe(engine_name, query, request_params, result_container, start_time, timeout_limit)
def search_multiple_requests(requests, result_container, start_time, timeout_limit):
search_id = uuid4().__str__()
for engine_name, query, request_params in requests:
th = threading.Thread(
target=search_one_request_safe,
args=(engine_name, query, request_params, result_container, start_time, timeout_limit),
name=search_id,
)
th._timeout = False
th._engine_name = engine_name
th.start()
for th in threading.enumerate():
if th.name == search_id:
remaining_time = max(0.0, timeout_limit - (time() - start_time))
th.join(remaining_time)
if th.is_alive():
th._timeout = True
result_container.add_unresponsive_engine(th._engine_name, 'timeout')
logger.warning('engine timeout: {0}'.format(th._engine_name))
# get default reqest parameter
def default_request_params():
return {
'method': 'GET',
'headers': {},
'data': {},
'url': '',
'cookies': {},
'verify': True,
'auth': None,
'raise_for_status': True
}
class Search:
"""Search information container"""
__slots__ = "search_query", "result_container", "start_time", "actual_timeout"
def __init__(self, search_query):
# init vars
super().__init__()
self.search_query = search_query
self.result_container = ResultContainer()
self.start_time = None
self.actual_timeout = None
def search_external_bang(self):
"""
Check if there is a external bang.
If yes, update self.result_container and return True
"""
if self.search_query.external_bang:
self.result_container.redirect_url = get_bang_url(self.search_query)
# This means there was a valid bang and the
# rest of the search does not need to be continued
if isinstance(self.result_container.redirect_url, str):
return True
return False
def search_answerers(self):
"""
Check if an answer return a result.
If yes, update self.result_container and return True
"""
answerers_results = ask(self.search_query)
if answerers_results:
for results in answerers_results:
self.result_container.extend('answer', results)
return True
return False
def _is_accepted(self, engine_name, engine):
# skip suspended engines
if engine.suspend_end_time >= time():
logger.debug('Engine currently suspended: %s', engine_name)
return False
# if paging is not supported, skip
if self.search_query.pageno > 1 and not engine.paging:
return False
# if time_range is not supported, skip
if self.search_query.time_range and not engine.time_range_support:
return False
return True
def _get_params(self, engineref, user_agent):
if engineref.name not in engines:
return None, None
engine = engines[engineref.name]
if not self._is_accepted(engineref.name, engine):
return None, None
# set default request parameters
request_params = {}
if not engine.offline:
request_params = default_request_params()
request_params['headers']['User-Agent'] = user_agent
if hasattr(engine, 'language') and engine.language:
request_params['language'] = engine.language
else:
request_params['language'] = self.search_query.lang
request_params['safesearch'] = self.search_query.safesearch
request_params['time_range'] = self.search_query.time_range
request_params['category'] = engineref.category
request_params['pageno'] = self.search_query.pageno
with threading.RLock():
engine.stats['sent_search_count'] += 1
return request_params, engine.timeout
# do search-request
def _get_requests(self):
# init vars
requests = []
# set default useragent
# user_agent = request.headers.get('User-Agent', '')
user_agent = gen_useragent()
# max of all selected engine timeout
default_timeout = 0
# start search-reqest for all selected engines
for engineref in self.search_query.engineref_list:
# set default request parameters
request_params, engine_timeout = self._get_params(engineref, user_agent)
if request_params is None:
continue
# append request to list
requests.append((engineref.name, self.search_query.query, request_params))
# update default_timeout
default_timeout = max(default_timeout, engine_timeout)
# adjust timeout
actual_timeout = default_timeout
query_timeout = self.search_query.timeout_limit
if max_request_timeout is None and query_timeout is None:
# No max, no user query: default_timeout
pass
elif max_request_timeout is None and query_timeout is not None:
# No max, but user query: From user query except if above default
actual_timeout = min(default_timeout, query_timeout)
elif max_request_timeout is not None and query_timeout is None:
# Max, no user query: Default except if above max
actual_timeout = min(default_timeout, max_request_timeout)
elif max_request_timeout is not None and query_timeout is not None:
# Max & user query: From user query except if above max
actual_timeout = min(query_timeout, max_request_timeout)
logger.debug("actual_timeout={0} (default_timeout={1}, ?timeout_limit={2}, max_request_timeout={3})"
.format(actual_timeout, default_timeout, query_timeout, max_request_timeout))
return requests, actual_timeout
def search_standard(self):
"""
Update self.result_container, self.actual_timeout
"""
requests, self.actual_timeout = self._get_requests()
# send all search-request
if requests:
search_multiple_requests(requests, self.result_container, self.start_time, self.actual_timeout)
start_new_thread(gc.collect, tuple())
# return results, suggestions, answers and infoboxes
return True
# do search-request
def search(self):
self.start_time = time()
if not self.search_external_bang():
if not self.search_answerers():
self.search_standard()
return self.result_container
class SearchWithPlugins(Search):
"""Similar to the Search class but call the plugins."""
__slots__ = 'ordered_plugin_list', 'request'
def __init__(self, search_query, ordered_plugin_list, request):
super().__init__(search_query)
self.ordered_plugin_list = ordered_plugin_list
self.request = request
def search(self):
if plugins.call(self.ordered_plugin_list, 'pre_search', self.request, self):
super().search()
plugins.call(self.ordered_plugin_list, 'post_search', self.request, self)
results = self.result_container.get_ordered_results()
for result in results:
plugins.call(self.ordered_plugin_list, 'on_result', self.request, self, result)
return self.result_container