summaryrefslogtreecommitdiff
path: root/searx/search.py
diff options
context:
space:
mode:
authorAlexandre Flament <alex@al-f.net>2020-09-22 13:59:27 +0200
committerAlexandre Flament <alex@al-f.net>2020-09-22 18:59:51 +0200
commit2dbc0de0cd000459ebfdb3d015be8684e737e95c (patch)
treef0087cd5dcf9c349d5a23ebb9bccea57bd875878 /searx/search.py
parent21dbc7e852a5c48097c0067e78f0600972e54823 (diff)
downloadsearxng-2dbc0de0cd000459ebfdb3d015be8684e737e95c.tar.gz
searxng-2dbc0de0cd000459ebfdb3d015be8684e737e95c.zip
[mod] add searx/webadapter.py
* move searx.search.get_search_query_from_webapp to searx.webadapter * move searx.query.SearchQuery to searx.search
Diffstat (limited to 'searx/search.py')
-rw-r--r--searx/search.py186
1 files changed, 21 insertions, 165 deletions
diff --git a/searx/search.py b/searx/search.py
index 96fffd569..d3b131d44 100644
--- a/searx/search.py
+++ b/searx/search.py
@@ -16,26 +16,20 @@ along with searx. If not, see < http://www.gnu.org/licenses/ >.
'''
import gc
-import sys
import threading
from time import time
from uuid import uuid4
from _thread import start_new_thread
-from flask_babel import gettext
import requests.exceptions
import searx.poolrequests as requests_lib
-from searx.engines import (
- categories, engines, settings
-)
+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.query import RawTextQuery, SearchQuery, VALID_LANGUAGE_CODE
from searx.results import ResultContainer
from searx import logger
from searx.plugins import plugins
-from searx.exceptions import SearxParameterException
logger = logger.getChild('search')
@@ -53,6 +47,26 @@ else:
exit(1)
+class SearchQuery:
+ """container for all the search parameters (query, language, etc...)"""
+
+ def __init__(self, query, engines, categories, lang, safesearch, pageno, time_range,
+ timeout_limit=None, preferences=None, external_bang=None):
+ self.query = query
+ self.engines = engines
+ self.categories = categories
+ self.lang = lang
+ self.safesearch = safesearch
+ self.pageno = pageno
+ self.time_range = None if time_range in ('', 'None', None) else time_range
+ self.timeout_limit = timeout_limit
+ self.preferences = preferences
+ self.external_bang = external_bang
+
+ def __str__(self):
+ return self.query + ";" + str(self.engines)
+
+
def send_http_request(engine, request_params):
# create dictionary which contain all
# informations about the request
@@ -247,164 +261,6 @@ def default_request_params():
}
-# remove duplicate queries.
-# FIXME: does not fix "!music !soundcloud", because the categories are 'none' and 'music'
-def deduplicate_query_engines(query_engines):
- uniq_query_engines = {q["category"] + '|' + q["name"]: q for q in query_engines}
- return uniq_query_engines.values()
-
-
-def get_search_query_from_webapp(preferences, form):
- # no text for the query ?
- if not form.get('q'):
- raise SearxParameterException('q', '')
-
- # set blocked engines
- disabled_engines = preferences.engines.get_disabled()
-
- # parse query, if tags are set, which change
- # the serch engine or search-language
- raw_text_query = RawTextQuery(form['q'], disabled_engines)
-
- # set query
- query = raw_text_query.getQuery()
-
- # get and check page number
- pageno_param = form.get('pageno', '1')
- if not pageno_param.isdigit() or int(pageno_param) < 1:
- raise SearxParameterException('pageno', pageno_param)
- query_pageno = int(pageno_param)
-
- # get language
- # set specific language if set on request, query or preferences
- # TODO support search with multible languages
- if len(raw_text_query.languages):
- query_lang = raw_text_query.languages[-1]
- elif 'language' in form:
- query_lang = form.get('language')
- else:
- query_lang = preferences.get_value('language')
-
- # check language
- if not VALID_LANGUAGE_CODE.match(query_lang):
- raise SearxParameterException('language', query_lang)
-
- # get safesearch
- if 'safesearch' in form:
- query_safesearch = form.get('safesearch')
- # first check safesearch
- if not query_safesearch.isdigit():
- raise SearxParameterException('safesearch', query_safesearch)
- query_safesearch = int(query_safesearch)
- else:
- query_safesearch = preferences.get_value('safesearch')
-
- # safesearch : second check
- if query_safesearch < 0 or query_safesearch > 2:
- raise SearxParameterException('safesearch', query_safesearch)
-
- # get time_range
- query_time_range = form.get('time_range')
-
- # check time_range
- if query_time_range not in ('None', None, '', 'day', 'week', 'month', 'year'):
- raise SearxParameterException('time_range', query_time_range)
-
- # query_engines
- query_engines = raw_text_query.engines
-
- # timeout_limit
- query_timeout = raw_text_query.timeout_limit
- if query_timeout is None and 'timeout_limit' in form:
- raw_time_limit = form.get('timeout_limit')
- if raw_time_limit in ['None', '']:
- raw_time_limit = None
- else:
- try:
- query_timeout = float(raw_time_limit)
- except ValueError:
- raise SearxParameterException('timeout_limit', raw_time_limit)
-
- # query_categories
- query_categories = []
-
- # if engines are calculated from query,
- # set categories by using that informations
- if query_engines and raw_text_query.specific:
- additional_categories = set()
- for engine in query_engines:
- if 'from_bang' in engine and engine['from_bang']:
- additional_categories.add('none')
- else:
- additional_categories.add(engine['category'])
- query_categories = list(additional_categories)
-
- # otherwise, using defined categories to
- # calculate which engines should be used
- else:
- # set categories/engines
- load_default_categories = True
- for pd_name, pd in form.items():
- if pd_name == 'categories':
- query_categories.extend(categ for categ in map(str.strip, pd.split(',')) if categ in categories)
- elif pd_name == 'engines':
- pd_engines = [{'category': engines[engine].categories[0],
- 'name': engine}
- for engine in map(str.strip, pd.split(',')) if engine in engines]
- if pd_engines:
- query_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
- query_categories.append(category)
- elif category in query_categories:
- # remove category from list if property is set to 'off'
- query_categories.remove(category)
-
- if not load_default_categories:
- if not query_categories:
- query_categories = list(set(engine['category']
- for engine in query_engines))
- else:
- # if no category is specified for this search,
- # using user-defined default-configuration which
- # (is stored in cookie)
- if not query_categories:
- cookie_categories = preferences.get_value('categories')
- for ccateg in cookie_categories:
- if ccateg in categories:
- query_categories.append(ccateg)
-
- # if still no category is specified, using general
- # as default-category
- if not query_categories:
- query_categories = ['general']
-
- # using all engines for that search, which are
- # declared under the specific categories
- for categ in query_categories:
- query_engines.extend({'category': categ,
- 'name': engine.name}
- for engine in categories[categ]
- if (engine.name, categ) not in disabled_engines)
-
- query_engines = deduplicate_query_engines(query_engines)
- external_bang = raw_text_query.external_bang
-
- return (SearchQuery(query, query_engines, query_categories,
- query_lang, query_safesearch, query_pageno,
- query_time_range, query_timeout, preferences,
- external_bang=external_bang),
- raw_text_query)
-
-
class Search:
"""Search information container"""