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(self, search_query: SearchQuery) -> ResultContainer:
result_container = ResultContainer()
engineref_category = search_query.engineref_list[0].category
params = self.processor.get_params(search_query, engineref_category)
if params is not None:
self.processor.search(search_query.query, params, result_container, time(), 5)
return result_container
def search(self, search_query: SearchQuery) -> ResultContainer:
result_container = ResultContainer()
engineref_category = search_query.engineref_list[0].category
params = self.processor.get_params(search_query, engineref_category)
if params is not None:
counter_inc('engine', search_query.engineref_list[0].name, 'search', 'count', 'sent')
self.processor.search(search_query.query, params, result_container, default_timer(), 5)
return result_container
def test_result_merge(self):
c = ResultContainer()
c.extend('wikipedia', [fake_result()])
c.extend('wikidata',
[fake_result(),
fake_result(url='https://example.com/')])
self.assertEqual(c.results_length(), 2)
def search(self, search_query: SearchQuery) -> ResultContainer:
result_container = ResultContainer()
engineref_category = search_query.engineref_list[0].category
params = self.processor.get_params(search_query, engineref_category)
if params is not None:
with threading.RLock():
self.processor.engine.stats['sent_search_count'] += 1
self.processor.search(search_query.query, params, result_container, time(), 5)
return result_container
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
# do search-request
def _get_requests(self):
# init vars
requests = []
# max of all selected engine timeout
default_timeout = 0
# start search-reqest for all selected engines
for engineref in self.search_query.engineref_list:
processor = processors[engineref.name]
# stop the request now if the engine is suspend
if processor.extend_container_if_suspended(self.result_container):
continue
# set default request parameters
request_params = processor.get_params(self.search_query, engineref.category)
if request_params is None:
continue
counter_inc('engine', engineref.name, 'search', 'count', 'sent')
# append request to list
requests.append((engineref.name, self.search_query.query, request_params))
# update default_timeout
default_timeout = max(default_timeout, processor.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_multiple_requests(self, requests):
search_id = uuid4().__str__()
for engine_name, query, request_params in requests:
th = threading.Thread(
target=processors[engine_name].search,
args=(query, request_params, self.result_container, self.start_time, self.actual_timeout),
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, self.actual_timeout - (default_timer() - self.start_time))
th.join(remaining_time)
if th.is_alive():
th._timeout = True
self.result_container.add_unresponsive_engine(th._engine_name, 'timeout')
logger.warning('engine timeout: {0}'.format(th._engine_name))
def search_standard(self):
"""
Update self.result_container, self.actual_timeout
"""
requests, self.actual_timeout = self._get_requests()
# send all search-request
if requests:
self.search_multiple_requests(requests)
start_new_thread(gc.collect, tuple())
# return results, suggestions, answers and infoboxes
return True
# do search-request
def search(self):
self.start_time = default_timer()
if not self.search_external_bang():
if not self.search_answerers():
self.search_standard()
return self.result_container
def __init__(self, request):
# init vars
super(Search, self).__init__()
self.query = None
self.engines = []
self.categories = []
self.paging = False
self.pageno = 1
self.lang = 'all'
# set blocked engines
self.blocked_engines = request.preferences.engines.get_disabled()
self.result_container = ResultContainer()
self.request_data = {}
# set specific language if set
self.lang = request.preferences.get_value('language')
# set request method
if request.method == 'POST':
self.request_data = request.form
else:
self.request_data = request.args
# TODO better exceptions
if not self.request_data.get('q'):
raise Exception('noquery')
# set pagenumber
pageno_param = self.request_data.get('pageno', '1')
if not pageno_param.isdigit() or int(pageno_param) < 1:
pageno_param = 1
self.pageno = int(pageno_param)
# parse query, if tags are set, which change
# the serch engine or search-language
query_obj = Query(self.request_data['q'], self.blocked_engines)
query_obj.parse_query()
# set query
self.query = query_obj.getSearchQuery()
# get last selected language in query, if possible
# TODO support search with multible languages
if len(query_obj.languages):
self.lang = query_obj.languages[-1]
self.engines = query_obj.engines
self.categories = []
# if engines are calculated from query,
# set categories by using that informations
if self.engines and query_obj.specific:
self.categories = list(
set(engine['category'] for engine in self.engines))
# otherwise, using defined categories to
# calculate which engines should be used
else:
# set categories/engines
load_default_categories = True
for pd_name, pd in self.request_data.items():
if pd_name == 'categories':
self.categories.extend(
categ for categ in map(unicode.strip, pd.split(','))
if categ in categories)
elif pd_name == 'engines':
pd_engines = [{
'category': engines[engine].categories[0],
'name': engine
} for engine in map(unicode.strip, pd.split(','))
if engine in engines]
if pd_engines:
self.engines.extend(pd_engines)
load_default_categories = False
elif pd_name.startswith('category_'):
category = pd_name[9:]
# if category is not found in list, skip
if category not in categories:
continue
if pd != 'off':
# add category to list
self.categories.append(category)
elif category in self.categories:
# remove category from list if property is set to 'off'
self.categories.remove(category)
if not load_default_categories:
if not self.categories:
self.categories = list(
set(engine['category'] for engine in self.engines))
return
# if no category is specified for this search,
# using user-defined default-configuration which
# (is stored in cookie)
if not self.categories:
cookie_categories = request.cookies.get('categories', '')
cookie_categories = cookie_categories.split(',')
for ccateg in cookie_categories:
if ccateg in categories:
self.categories.append(ccateg)
# if still no category is specified, using general
# as default-category
if not self.categories:
self.categories = ['general']
# using all engines for that search, which are
# declared under the specific categories
for categ in self.categories:
self.engines.extend({
'category': categ,
'name': engine.name
} for engine in categories[categ]
if (engine.name,
categ) not in self.blocked_engines)
# remove suspended engines
self.engines = [
e for e in self.engines
if engines[e['name']].suspend_end_time <= time()
]
def test_result_merge(self):
c = ResultContainer()
c.extend('wikipedia', [fake_result()])
c.extend('wikidata', [fake_result(), fake_result(url='https://example.com/')])
self.assertEqual(c.results_length(), 2)
def test_one_suggestion(self):
c = ResultContainer()
c.extend('wikipedia', [fake_result(suggestion=True)])
self.assertEqual(len(c.suggestions), 1)
self.assertEqual(c.results_length(), 0)
def test_one_result(self):
c = ResultContainer()
c.extend('wikipedia', [fake_result()])
self.assertEqual(c.results_length(), 1)
class Search:
"""Search information container"""
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):
if not self.search_query.preferences.validate_token(engine):
return False
# 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, selected_engine, user_agent):
if selected_engine['name'] not in engines:
return None, None
engine = engines[selected_engine['name']]
if not self._is_accepted(selected_engine['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'] = selected_engine['category']
request_params['pageno'] = self.search_query.pageno
return request_params, engine.timeout
# do search-request
def _get_requests(self):
global number_of_searches
# 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 selected_engine in self.search_query.engines:
# set default request parameters
request_params, engine_timeout = self._get_params(
selected_engine, user_agent)
if request_params is None:
continue
# append request to list
requests.append((selected_engine['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(self.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
def __init__(self, search_query):
# init vars
super(Search, self).__init__()
self.search_query = search_query
self.result_container = ResultContainer()
class Search(object):
"""Search information container"""
def __init__(self, search_query):
# init vars
super(Search, self).__init__()
self.search_query = search_query
self.result_container = ResultContainer()
# do search-request
def search(self):
global number_of_searches
# start time
start_time = time()
# answeres ?
answerers_results = ask(self.search_query)
if answerers_results:
for results in answerers_results:
self.result_container.extend('answer', results)
return self.result_container
# init vars
requests = []
# increase number of searches
number_of_searches += 1
# set default useragent
# user_agent = request.headers.get('User-Agent', '')
user_agent = gen_useragent()
search_query = self.search_query
# max of all selected engine timeout
timeout_limit = 0
# start search-reqest for all selected engines
for selected_engine in search_query.engines:
if selected_engine['name'] not in engines:
continue
engine = engines[selected_engine['name']]
# skip suspended engines
if engine.suspend_end_time >= time():
logger.debug('Engine currently suspended: %s', selected_engine['name'])
continue
# if paging is not supported, skip
if search_query.pageno > 1 and not engine.paging:
continue
# if time_range is not supported, skip
if search_query.time_range and not engine.time_range_support:
continue
# set default request parameters
request_params = default_request_params()
request_params['headers']['User-Agent'] = user_agent
request_params['category'] = selected_engine['category']
request_params['pageno'] = search_query.pageno
if hasattr(engine, 'language') and engine.language:
request_params['language'] = engine.language
else:
request_params['language'] = search_query.lang
# 0 = None, 1 = Moderate, 2 = Strict
request_params['safesearch'] = search_query.safesearch
request_params['time_range'] = search_query.time_range
# append request to list
requests.append((selected_engine['name'], search_query.query.encode('utf-8'), request_params))
# update timeout_limit
timeout_limit = max(timeout_limit, engine.timeout)
if requests:
# send all search-request
search_multiple_requests(requests, self.result_container, start_time, timeout_limit)
start_new_thread(gc.collect, tuple())
# return results, suggestions, answers and infoboxes
return self.result_container
class Search(object):
"""Search information container"""
def __init__(self, search_query):
# init vars
super(Search, self).__init__()
self.search_query = search_query
self.result_container = ResultContainer()
self.actual_timeout = None
# do search-request
def search(self):
global number_of_searches
# start time
start_time = time()
# answeres ?
answerers_results = ask(self.search_query)
if answerers_results:
for results in answerers_results:
self.result_container.extend('answer', results)
return self.result_container
# init vars
requests = []
# increase number of searches
number_of_searches += 1
# set default useragent
# user_agent = request.headers.get('User-Agent', '')
user_agent = gen_useragent()
search_query = self.search_query
# max of all selected engine timeout
default_timeout = 0
# start search-reqest for all selected engines
for selected_engine in search_query.engines:
if selected_engine['name'] not in engines:
continue
engine = engines[selected_engine['name']]
# skip suspended engines
if engine.suspend_end_time >= time():
logger.debug('Engine currently suspended: %s', selected_engine['name'])
continue
# if paging is not supported, skip
if search_query.pageno > 1 and not engine.paging:
continue
# if time_range is not supported, skip
if search_query.time_range and not engine.time_range_support:
continue
# 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'] = search_query.lang
request_params['safesearch'] = search_query.safesearch
request_params['time_range'] = search_query.time_range
request_params['category'] = selected_engine['category']
request_params['pageno'] = search_query.pageno
# append request to list
requests.append((selected_engine['name'], search_query.query, request_params))
# update default_timeout
default_timeout = max(default_timeout, engine.timeout)
# adjust timeout
self.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
self.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
self.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
self.actual_timeout = min(query_timeout, max_request_timeout)
logger.debug("actual_timeout={0} (default_timeout={1}, ?timeout_limit={2}, max_request_timeout={3})"
.format(self.actual_timeout, default_timeout, query_timeout, max_request_timeout))
# send all search-request
if requests:
search_multiple_requests(requests, self.result_container, start_time, self.actual_timeout)
start_new_thread(gc.collect, tuple())
# return results, suggestions, answers and infoboxes
return self.result_container
class Search(object):
"""Search information container"""
def __init__(self, search_query):
# init vars
super(Search, self).__init__()
self.search_query = search_query
self.result_container = ResultContainer()
# do search-request
def search(self):
global number_of_searches
answerers_results = ask(self.search_query)
if answerers_results:
for results in answerers_results:
self.result_container.extend('answer', results)
return self.result_container
# init vars
requests = []
# increase number of searches
number_of_searches += 1
# set default useragent
# user_agent = request.headers.get('User-Agent', '')
user_agent = gen_useragent()
search_query = self.search_query
# start search-reqest for all selected engines
for selected_engine in search_query.engines:
if selected_engine['name'] not in engines:
continue
engine = engines[selected_engine['name']]
# skip suspended engines
if engine.suspend_end_time >= time():
logger.debug('Engine currently suspended: %s',
selected_engine['name'])
continue
# if paging is not supported, skip
if search_query.pageno > 1 and not engine.paging:
continue
# if search-language is set and engine does not
# provide language-support, skip
if search_query.lang != 'all' and not engine.language_support:
continue
# if time_range is not supported, skip
if search_query.time_range and not engine.time_range_support:
continue
# set default request parameters
request_params = default_request_params()
request_params['headers']['User-Agent'] = user_agent
request_params['category'] = selected_engine['category']
request_params['started'] = time()
request_params['pageno'] = search_query.pageno
if hasattr(engine, 'language') and engine.language:
request_params['language'] = engine.language
else:
request_params['language'] = search_query.lang
# 0 = None, 1 = Moderate, 2 = Strict
request_params['safesearch'] = search_query.safesearch
request_params['time_range'] = search_query.time_range
# update request parameters dependent on
# search-engine (contained in engines folder)
engine.request(search_query.query.encode('utf-8'), request_params)
if request_params['url'] is None:
# TODO add support of offline engines
pass
# create a callback wrapper for the search engine results
callback = make_callback(selected_engine['name'], engine.response,
request_params, self.result_container)
# create dictionary which contain all
# informations about the request
request_args = dict(headers=request_params['headers'],
hooks=dict(response=callback),
cookies=request_params['cookies'],
timeout=engine.timeout,
verify=request_params['verify'])
# 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']
# ignoring empty urls
if not request_params['url']:
continue
# append request to list
requests.append((req, request_params['url'], request_args,
selected_engine['name']))
if not requests:
return self.result_container
# send all search-request
threaded_requests(requests)
start_new_thread(gc.collect, tuple())
# return results, suggestions, answers and infoboxes
return self.result_container
class Search(object):
"""Search information container"""
def __init__(self, search_query):
# init vars
super(Search, self).__init__()
self.search_query = search_query
self.result_container = ResultContainer()
# do search-request
def search(self):
global number_of_searches
# start time
start_time = time()
# answeres ?
answerers_results = ask(self.search_query)
if answerers_results:
for results in answerers_results:
self.result_container.extend('answer', results)
return self.result_container
# init vars
requests = []
# increase number of searches
number_of_searches += 1
# set default useragent
# user_agent = request.headers.get('User-Agent', '')
user_agent = gen_useragent()
search_query = self.search_query
# max of all selected engine timeout
timeout_limit = 0
# start search-reqest for all selected engines
for selected_engine in search_query.engines:
if selected_engine['name'] not in engines:
continue
engine = engines[selected_engine['name']]
# skip suspended engines
if engine.suspend_end_time >= time():
logger.debug('Engine currently suspended: %s', selected_engine['name'])
continue
# if paging is not supported, skip
if search_query.pageno > 1 and not engine.paging:
continue
# if time_range is not supported, skip
if search_query.time_range and not engine.time_range_support:
continue
# set default request parameters
request_params = default_request_params()
request_params['headers']['User-Agent'] = user_agent
request_params['category'] = selected_engine['category']
request_params['pageno'] = search_query.pageno
if hasattr(engine, 'language') and engine.language:
request_params['language'] = engine.language
else:
request_params['language'] = search_query.lang
# 0 = None, 1 = Moderate, 2 = Strict
request_params['safesearch'] = search_query.safesearch
request_params['time_range'] = search_query.time_range
# append request to list
requests.append((selected_engine['name'], search_query.query, request_params))
# update timeout_limit
timeout_limit = max(timeout_limit, engine.timeout)
if requests:
# send all search-request
search_multiple_requests(requests, self.result_container, start_time, timeout_limit)
start_new_thread(gc.collect, tuple())
# return results, suggestions, answers and infoboxes
return self.result_container
def __init__(self, search_query):
# init vars
super(Search, self).__init__()
self.search_query = search_query
self.result_container = ResultContainer()
def test_empty(self):
c = ResultContainer()
self.assertEqual(c.get_ordered_results(), [])
