mirror of
https://github.com/searxng/searxng
synced 2024-01-01 18:24:07 +00:00
Merge branch 'comments' of https://github.com/pointhi/searx
Conflicts: searx/search.py
This commit is contained in:
commit
bd2db71fa6
@ -1,3 +1,20 @@
|
|||||||
|
'''
|
||||||
|
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>
|
||||||
|
'''
|
||||||
|
|
||||||
from os import environ
|
from os import environ
|
||||||
from os.path import realpath, dirname, join, abspath
|
from os.path import realpath, dirname, join, abspath
|
||||||
try:
|
try:
|
||||||
@ -10,11 +27,14 @@ except:
|
|||||||
searx_dir = abspath(dirname(__file__))
|
searx_dir = abspath(dirname(__file__))
|
||||||
engine_dir = dirname(realpath(__file__))
|
engine_dir = dirname(realpath(__file__))
|
||||||
|
|
||||||
|
# if possible set path to settings using the enviroment variable SEARX_SETTINGS_PATH
|
||||||
if 'SEARX_SETTINGS_PATH' in environ:
|
if 'SEARX_SETTINGS_PATH' in environ:
|
||||||
settings_path = environ['SEARX_SETTINGS_PATH']
|
settings_path = environ['SEARX_SETTINGS_PATH']
|
||||||
|
# otherwise using default path
|
||||||
else:
|
else:
|
||||||
settings_path = join(searx_dir, 'settings.yml')
|
settings_path = join(searx_dir, 'settings.yml')
|
||||||
|
|
||||||
|
|
||||||
|
# load settings
|
||||||
with open(settings_path) as settings_yaml:
|
with open(settings_path) as settings_yaml:
|
||||||
settings = load(settings_yaml)
|
settings = load(settings_yaml)
|
||||||
|
@ -1,3 +1,21 @@
|
|||||||
|
'''
|
||||||
|
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>
|
||||||
|
'''
|
||||||
|
|
||||||
|
|
||||||
from lxml import etree
|
from lxml import etree
|
||||||
from requests import get
|
from requests import get
|
||||||
from json import loads
|
from json import loads
|
||||||
@ -22,7 +40,7 @@ def dbpedia(query):
|
|||||||
|
|
||||||
|
|
||||||
def duckduckgo(query):
|
def duckduckgo(query):
|
||||||
# wikipedia autocompleter
|
# duckduckgo autocompleter
|
||||||
url = 'https://ac.duckduckgo.com/ac/?{0}&type=list'
|
url = 'https://ac.duckduckgo.com/ac/?{0}&type=list'
|
||||||
|
|
||||||
resp = loads(get(url.format(urlencode(dict(q=query)))).text)
|
resp = loads(get(url.format(urlencode(dict(q=query)))).text)
|
||||||
|
@ -1,3 +1,21 @@
|
|||||||
|
'''
|
||||||
|
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>
|
||||||
|
'''
|
||||||
|
|
||||||
|
# list of language codes
|
||||||
language_codes = (
|
language_codes = (
|
||||||
("ar_XA", "Arabic", "Arabia"),
|
("ar_XA", "Arabic", "Arabia"),
|
||||||
("bg_BG", "Bulgarian", "Bulgaria"),
|
("bg_BG", "Bulgarian", "Bulgaria"),
|
||||||
|
118
searx/search.py
118
searx/search.py
@ -1,3 +1,20 @@
|
|||||||
|
'''
|
||||||
|
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 grequests
|
import grequests
|
||||||
from itertools import izip_longest, chain
|
from itertools import izip_longest, chain
|
||||||
from datetime import datetime
|
from datetime import datetime
|
||||||
@ -9,45 +26,65 @@ from searx.engines import (
|
|||||||
from searx.languages import language_codes
|
from searx.languages import language_codes
|
||||||
from searx.utils import gen_useragent
|
from searx.utils import gen_useragent
|
||||||
|
|
||||||
|
|
||||||
number_of_searches = 0
|
number_of_searches = 0
|
||||||
|
|
||||||
|
|
||||||
|
# get default reqest parameter
|
||||||
def default_request_params():
|
def default_request_params():
|
||||||
return {
|
return {
|
||||||
'method': 'GET', 'headers': {}, 'data': {}, 'url': '', 'cookies': {}}
|
'method': 'GET', 'headers': {}, 'data': {}, 'url': '', 'cookies': {}}
|
||||||
|
|
||||||
|
|
||||||
|
# create a callback wrapper for the search engine results
|
||||||
def make_callback(engine_name, results, suggestions, callback, params):
|
def make_callback(engine_name, results, suggestions, callback, params):
|
||||||
|
|
||||||
# creating a callback wrapper for the search engine results
|
# creating a callback wrapper for the search engine results
|
||||||
def process_callback(response, **kwargs):
|
def process_callback(response, **kwargs):
|
||||||
cb_res = []
|
cb_res = []
|
||||||
response.search_params = params
|
response.search_params = params
|
||||||
|
|
||||||
|
# update stats with current page-load-time
|
||||||
engines[engine_name].stats['page_load_time'] += \
|
engines[engine_name].stats['page_load_time'] += \
|
||||||
(datetime.now() - params['started']).total_seconds()
|
(datetime.now() - params['started']).total_seconds()
|
||||||
|
|
||||||
try:
|
try:
|
||||||
search_results = callback(response)
|
search_results = callback(response)
|
||||||
except Exception, e:
|
except Exception, e:
|
||||||
|
# increase errors stats
|
||||||
engines[engine_name].stats['errors'] += 1
|
engines[engine_name].stats['errors'] += 1
|
||||||
results[engine_name] = cb_res
|
results[engine_name] = cb_res
|
||||||
|
|
||||||
|
# print engine name and specific error message
|
||||||
print '[E] Error with engine "{0}":\n\t{1}'.format(
|
print '[E] Error with engine "{0}":\n\t{1}'.format(
|
||||||
engine_name, str(e))
|
engine_name, str(e))
|
||||||
return
|
return
|
||||||
|
|
||||||
for result in search_results:
|
for result in search_results:
|
||||||
result['engine'] = engine_name
|
result['engine'] = engine_name
|
||||||
|
|
||||||
|
# if it is a suggestion, add it to list of suggestions
|
||||||
if 'suggestion' in result:
|
if 'suggestion' in result:
|
||||||
# TODO type checks
|
# TODO type checks
|
||||||
suggestions.add(result['suggestion'])
|
suggestions.add(result['suggestion'])
|
||||||
continue
|
continue
|
||||||
|
|
||||||
|
# append result
|
||||||
cb_res.append(result)
|
cb_res.append(result)
|
||||||
|
|
||||||
results[engine_name] = cb_res
|
results[engine_name] = cb_res
|
||||||
|
|
||||||
return process_callback
|
return process_callback
|
||||||
|
|
||||||
|
|
||||||
|
# score results and remove duplications
|
||||||
def score_results(results):
|
def score_results(results):
|
||||||
|
# calculate scoring parameters
|
||||||
flat_res = filter(
|
flat_res = filter(
|
||||||
None, chain.from_iterable(izip_longest(*results.values())))
|
None, chain.from_iterable(izip_longest(*results.values())))
|
||||||
flat_len = len(flat_res)
|
flat_len = len(flat_res)
|
||||||
engines_len = len(results)
|
engines_len = len(results)
|
||||||
|
|
||||||
results = []
|
results = []
|
||||||
|
|
||||||
# pass 1: deduplication + scoring
|
# pass 1: deduplication + scoring
|
||||||
@ -63,34 +100,53 @@ def score_results(results):
|
|||||||
res['engines'] = [res['engine']]
|
res['engines'] = [res['engine']]
|
||||||
weight = 1.0
|
weight = 1.0
|
||||||
|
|
||||||
|
# get weight of this engine if possible
|
||||||
if hasattr(engines[res['engine']], 'weight'):
|
if hasattr(engines[res['engine']], 'weight'):
|
||||||
weight = float(engines[res['engine']].weight)
|
weight = float(engines[res['engine']].weight)
|
||||||
|
|
||||||
|
# calculate score for that engine
|
||||||
score = int((flat_len - i) / engines_len) * weight + 1
|
score = int((flat_len - i) / engines_len) * weight + 1
|
||||||
|
|
||||||
duplicated = False
|
duplicated = False
|
||||||
|
|
||||||
|
# check for duplicates
|
||||||
for new_res in results:
|
for new_res in results:
|
||||||
|
# remove / from the end of the url if required
|
||||||
p1 = res['parsed_url'].path[:-1] if res['parsed_url'].path.endswith('/') else res['parsed_url'].path # noqa
|
p1 = res['parsed_url'].path[:-1] if res['parsed_url'].path.endswith('/') else res['parsed_url'].path # noqa
|
||||||
p2 = new_res['parsed_url'].path[:-1] if new_res['parsed_url'].path.endswith('/') else new_res['parsed_url'].path # noqa
|
p2 = new_res['parsed_url'].path[:-1] if new_res['parsed_url'].path.endswith('/') else new_res['parsed_url'].path # noqa
|
||||||
|
|
||||||
|
# check if that result is a duplicate
|
||||||
if res['host'] == new_res['host'] and\
|
if res['host'] == new_res['host'] and\
|
||||||
unquote(p1) == unquote(p2) and\
|
unquote(p1) == unquote(p2) and\
|
||||||
res['parsed_url'].query == new_res['parsed_url'].query and\
|
res['parsed_url'].query == new_res['parsed_url'].query and\
|
||||||
res.get('template') == new_res.get('template'):
|
res.get('template') == new_res.get('template'):
|
||||||
duplicated = new_res
|
duplicated = new_res
|
||||||
break
|
break
|
||||||
|
|
||||||
|
# merge duplicates together
|
||||||
if duplicated:
|
if duplicated:
|
||||||
|
# using content with more text
|
||||||
if res.get('content') > duplicated.get('content'):
|
if res.get('content') > duplicated.get('content'):
|
||||||
duplicated['content'] = res['content']
|
duplicated['content'] = res['content']
|
||||||
|
|
||||||
|
# increase result-score
|
||||||
duplicated['score'] += score
|
duplicated['score'] += score
|
||||||
|
|
||||||
|
# add engine to list of result-engines
|
||||||
duplicated['engines'].append(res['engine'])
|
duplicated['engines'].append(res['engine'])
|
||||||
|
|
||||||
|
# using https if possible
|
||||||
if duplicated['parsed_url'].scheme == 'https':
|
if duplicated['parsed_url'].scheme == 'https':
|
||||||
continue
|
continue
|
||||||
elif res['parsed_url'].scheme == 'https':
|
elif res['parsed_url'].scheme == 'https':
|
||||||
duplicated['url'] = res['parsed_url'].geturl()
|
duplicated['url'] = res['parsed_url'].geturl()
|
||||||
duplicated['parsed_url'] = res['parsed_url']
|
duplicated['parsed_url'] = res['parsed_url']
|
||||||
|
|
||||||
|
# if there is no duplicate found, append result
|
||||||
else:
|
else:
|
||||||
res['score'] = score
|
res['score'] = score
|
||||||
results.append(res)
|
results.append(res)
|
||||||
|
|
||||||
results = sorted(results, key=itemgetter('score'), reverse=True)
|
results = sorted(results, key=itemgetter('score'), reverse=True)
|
||||||
|
|
||||||
# pass 2 : group results by category and template
|
# pass 2 : group results by category and template
|
||||||
@ -99,7 +155,7 @@ def score_results(results):
|
|||||||
|
|
||||||
for i, res in enumerate(results):
|
for i, res in enumerate(results):
|
||||||
# FIXME : handle more than one category per engine
|
# FIXME : handle more than one category per engine
|
||||||
category = engines[res['engine']].categories[0] + ':' + '' if 'template' not in res else res['template']
|
category = engines[res['engine']].categories[0] + ':' + '' if 'template' not in res else res['template']
|
||||||
|
|
||||||
current = None if category not in categoryPositions else categoryPositions[category]
|
current = None if category not in categoryPositions else categoryPositions[category]
|
||||||
|
|
||||||
@ -134,6 +190,7 @@ class Search(object):
|
|||||||
"""Search information container"""
|
"""Search information container"""
|
||||||
|
|
||||||
def __init__(self, request):
|
def __init__(self, request):
|
||||||
|
# init vars
|
||||||
super(Search, self).__init__()
|
super(Search, self).__init__()
|
||||||
self.query = None
|
self.query = None
|
||||||
self.engines = []
|
self.engines = []
|
||||||
@ -141,18 +198,23 @@ class Search(object):
|
|||||||
self.paging = False
|
self.paging = False
|
||||||
self.pageno = 1
|
self.pageno = 1
|
||||||
self.lang = 'all'
|
self.lang = 'all'
|
||||||
|
|
||||||
|
# set blocked engines
|
||||||
if request.cookies.get('blocked_engines'):
|
if request.cookies.get('blocked_engines'):
|
||||||
self.blocked_engines = request.cookies['blocked_engines'].split(',') # noqa
|
self.blocked_engines = request.cookies['blocked_engines'].split(',') # noqa
|
||||||
else:
|
else:
|
||||||
self.blocked_engines = []
|
self.blocked_engines = []
|
||||||
|
|
||||||
self.results = []
|
self.results = []
|
||||||
self.suggestions = []
|
self.suggestions = []
|
||||||
self.request_data = {}
|
self.request_data = {}
|
||||||
|
|
||||||
|
# set specific language if set
|
||||||
if request.cookies.get('language')\
|
if request.cookies.get('language')\
|
||||||
and request.cookies['language'] in (x[0] for x in language_codes):
|
and request.cookies['language'] in (x[0] for x in language_codes):
|
||||||
self.lang = request.cookies['language']
|
self.lang = request.cookies['language']
|
||||||
|
|
||||||
|
# set request method
|
||||||
if request.method == 'POST':
|
if request.method == 'POST':
|
||||||
self.request_data = request.form
|
self.request_data = request.form
|
||||||
else:
|
else:
|
||||||
@ -162,51 +224,72 @@ class Search(object):
|
|||||||
if not self.request_data.get('q'):
|
if not self.request_data.get('q'):
|
||||||
raise Exception('noquery')
|
raise Exception('noquery')
|
||||||
|
|
||||||
|
# set query
|
||||||
self.query = self.request_data['q']
|
self.query = self.request_data['q']
|
||||||
|
|
||||||
|
# set pagenumber
|
||||||
pageno_param = self.request_data.get('pageno', '1')
|
pageno_param = self.request_data.get('pageno', '1')
|
||||||
if not pageno_param.isdigit() or int(pageno_param) < 1:
|
if not pageno_param.isdigit() or int(pageno_param) < 1:
|
||||||
raise Exception('wrong pagenumber')
|
raise Exception('wrong pagenumber')
|
||||||
|
|
||||||
self.pageno = int(pageno_param)
|
self.pageno = int(pageno_param)
|
||||||
|
|
||||||
|
# parse query, if tags are set, which change the serch engine or search-language
|
||||||
self.parse_query()
|
self.parse_query()
|
||||||
|
|
||||||
self.categories = []
|
self.categories = []
|
||||||
|
|
||||||
|
# if engines are calculated from query, set categories by using that informations
|
||||||
if self.engines:
|
if self.engines:
|
||||||
self.categories = list(set(engine['category']
|
self.categories = list(set(engine['category']
|
||||||
for engine in self.engines))
|
for engine in self.engines))
|
||||||
|
|
||||||
|
# otherwise, using defined categories to calculate which engines should be used
|
||||||
else:
|
else:
|
||||||
|
# set used categories
|
||||||
for pd_name, pd in self.request_data.items():
|
for pd_name, pd in self.request_data.items():
|
||||||
if pd_name.startswith('category_'):
|
if pd_name.startswith('category_'):
|
||||||
category = pd_name[9:]
|
category = pd_name[9:]
|
||||||
|
# if category is not found in list, skip
|
||||||
if not category in categories:
|
if not category in categories:
|
||||||
continue
|
continue
|
||||||
|
|
||||||
|
# add category to list
|
||||||
self.categories.append(category)
|
self.categories.append(category)
|
||||||
|
|
||||||
|
# if no category is specified for this search, using user-defined default-configuration which (is stored in cookie)
|
||||||
if not self.categories:
|
if not self.categories:
|
||||||
cookie_categories = request.cookies.get('categories', '')
|
cookie_categories = request.cookies.get('categories', '')
|
||||||
cookie_categories = cookie_categories.split(',')
|
cookie_categories = cookie_categories.split(',')
|
||||||
for ccateg in cookie_categories:
|
for ccateg in cookie_categories:
|
||||||
if ccateg in categories:
|
if ccateg in categories:
|
||||||
self.categories.append(ccateg)
|
self.categories.append(ccateg)
|
||||||
|
|
||||||
|
# if still no category is specified, using general as default-category
|
||||||
if not self.categories:
|
if not self.categories:
|
||||||
self.categories = ['general']
|
self.categories = ['general']
|
||||||
|
|
||||||
|
# using all engines for that search, which are declared under the specific categories
|
||||||
for categ in self.categories:
|
for categ in self.categories:
|
||||||
self.engines.extend({'category': categ,
|
self.engines.extend({'category': categ,
|
||||||
'name': x.name}
|
'name': x.name}
|
||||||
for x in categories[categ]
|
for x in categories[categ]
|
||||||
if not x.name in self.blocked_engines)
|
if not x.name in self.blocked_engines)
|
||||||
|
|
||||||
|
# parse query, if tags are set, which change the serch engine or search-language
|
||||||
def parse_query(self):
|
def parse_query(self):
|
||||||
query_parts = self.query.split()
|
query_parts = self.query.split()
|
||||||
modified = False
|
modified = False
|
||||||
|
|
||||||
|
# check if language-prefix is set
|
||||||
if query_parts[0].startswith(':'):
|
if query_parts[0].startswith(':'):
|
||||||
lang = query_parts[0][1:].lower()
|
lang = query_parts[0][1:].lower()
|
||||||
|
|
||||||
|
# check if any language-code is equal with declared language-codes
|
||||||
for lc in language_codes:
|
for lc in language_codes:
|
||||||
lang_id, lang_name, country = map(str.lower, lc)
|
lang_id, lang_name, country = map(str.lower, lc)
|
||||||
|
|
||||||
|
# if correct language-code is found, set it as new search-language
|
||||||
if lang == lang_id\
|
if lang == lang_id\
|
||||||
or lang_id.startswith(lang)\
|
or lang_id.startswith(lang)\
|
||||||
or lang == lang_name\
|
or lang == lang_name\
|
||||||
@ -215,56 +298,78 @@ class Search(object):
|
|||||||
modified = True
|
modified = True
|
||||||
break
|
break
|
||||||
|
|
||||||
|
# check if category/engine prefix is set
|
||||||
elif query_parts[0].startswith('!'):
|
elif query_parts[0].startswith('!'):
|
||||||
prefix = query_parts[0][1:].replace('_', ' ')
|
prefix = query_parts[0][1:].replace('_', ' ')
|
||||||
|
|
||||||
|
# check if prefix is equal with engine shortcut
|
||||||
if prefix in engine_shortcuts\
|
if prefix in engine_shortcuts\
|
||||||
and not engine_shortcuts[prefix] in self.blocked_engines:
|
and not engine_shortcuts[prefix] in self.blocked_engines:
|
||||||
modified = True
|
modified = True
|
||||||
self.engines.append({'category': 'none',
|
self.engines.append({'category': 'none',
|
||||||
'name': engine_shortcuts[prefix]})
|
'name': engine_shortcuts[prefix]})
|
||||||
|
|
||||||
|
# check if prefix is equal with engine name
|
||||||
elif prefix in engines\
|
elif prefix in engines\
|
||||||
and not prefix in self.blocked_engines:
|
and not prefix in self.blocked_engines:
|
||||||
modified = True
|
modified = True
|
||||||
self.engines.append({'category': 'none',
|
self.engines.append({'category': 'none',
|
||||||
'name': prefix})
|
'name': prefix})
|
||||||
|
|
||||||
|
# check if prefix is equal with categorie name
|
||||||
elif prefix in categories:
|
elif prefix in categories:
|
||||||
modified = True
|
modified = True
|
||||||
|
# using all engines for that search, which are declared under that categorie name
|
||||||
self.engines.extend({'category': prefix,
|
self.engines.extend({'category': prefix,
|
||||||
'name': engine.name}
|
'name': engine.name}
|
||||||
for engine in categories[prefix]
|
for engine in categories[prefix]
|
||||||
if not engine in self.blocked_engines)
|
if not engine in self.blocked_engines)
|
||||||
|
|
||||||
|
# if language, category or engine were specificed in this query, search for more tags which does the same
|
||||||
if modified:
|
if modified:
|
||||||
self.query = self.query.replace(query_parts[0], '', 1).strip()
|
self.query = self.query.replace(query_parts[0], '', 1).strip()
|
||||||
self.parse_query()
|
self.parse_query()
|
||||||
|
|
||||||
|
# do search-request
|
||||||
def search(self, request):
|
def search(self, request):
|
||||||
global number_of_searches
|
global number_of_searches
|
||||||
|
|
||||||
|
# init vars
|
||||||
requests = []
|
requests = []
|
||||||
results = {}
|
results = {}
|
||||||
suggestions = set()
|
suggestions = set()
|
||||||
|
|
||||||
|
# increase number of searches
|
||||||
number_of_searches += 1
|
number_of_searches += 1
|
||||||
|
|
||||||
|
# set default useragent
|
||||||
#user_agent = request.headers.get('User-Agent', '')
|
#user_agent = request.headers.get('User-Agent', '')
|
||||||
user_agent = gen_useragent()
|
user_agent = gen_useragent()
|
||||||
|
|
||||||
|
# start search-reqest for all selected engines
|
||||||
for selected_engine in self.engines:
|
for selected_engine in self.engines:
|
||||||
if selected_engine['name'] not in engines:
|
if selected_engine['name'] not in engines:
|
||||||
continue
|
continue
|
||||||
|
|
||||||
engine = engines[selected_engine['name']]
|
engine = engines[selected_engine['name']]
|
||||||
|
|
||||||
|
# if paging is not supported, skip
|
||||||
if self.pageno > 1 and not engine.paging:
|
if self.pageno > 1 and not engine.paging:
|
||||||
continue
|
continue
|
||||||
|
|
||||||
|
# if search-language is set and engine does not provide language-support, skip
|
||||||
if self.lang != 'all' and not engine.language_support:
|
if self.lang != 'all' and not engine.language_support:
|
||||||
continue
|
continue
|
||||||
|
|
||||||
|
# set default request parameters
|
||||||
request_params = default_request_params()
|
request_params = default_request_params()
|
||||||
request_params['headers']['User-Agent'] = user_agent
|
request_params['headers']['User-Agent'] = user_agent
|
||||||
request_params['category'] = selected_engine['category']
|
request_params['category'] = selected_engine['category']
|
||||||
request_params['started'] = datetime.now()
|
request_params['started'] = datetime.now()
|
||||||
request_params['pageno'] = self.pageno
|
request_params['pageno'] = self.pageno
|
||||||
request_params['language'] = self.lang
|
request_params['language'] = self.lang
|
||||||
|
|
||||||
|
# update request parameters dependent on search-engine (contained in engines folder)
|
||||||
request_params = engine.request(self.query.encode('utf-8'),
|
request_params = engine.request(self.query.encode('utf-8'),
|
||||||
request_params)
|
request_params)
|
||||||
|
|
||||||
@ -272,6 +377,7 @@ class Search(object):
|
|||||||
# TODO add support of offline engines
|
# TODO add support of offline engines
|
||||||
pass
|
pass
|
||||||
|
|
||||||
|
# create a callback wrapper for the search engine results
|
||||||
callback = make_callback(
|
callback = make_callback(
|
||||||
selected_engine['name'],
|
selected_engine['name'],
|
||||||
results,
|
results,
|
||||||
@ -280,6 +386,7 @@ class Search(object):
|
|||||||
request_params
|
request_params
|
||||||
)
|
)
|
||||||
|
|
||||||
|
# create dictionary which contain all informations about the request
|
||||||
request_args = dict(
|
request_args = dict(
|
||||||
headers=request_params['headers'],
|
headers=request_params['headers'],
|
||||||
hooks=dict(response=callback),
|
hooks=dict(response=callback),
|
||||||
@ -287,6 +394,7 @@ class Search(object):
|
|||||||
timeout=engine.timeout
|
timeout=engine.timeout
|
||||||
)
|
)
|
||||||
|
|
||||||
|
# specific type of request (GET or POST)
|
||||||
if request_params['method'] == 'GET':
|
if request_params['method'] == 'GET':
|
||||||
req = grequests.get
|
req = grequests.get
|
||||||
else:
|
else:
|
||||||
@ -297,17 +405,25 @@ class Search(object):
|
|||||||
if not request_params['url']:
|
if not request_params['url']:
|
||||||
continue
|
continue
|
||||||
|
|
||||||
|
# append request to list
|
||||||
requests.append(req(request_params['url'], **request_args))
|
requests.append(req(request_params['url'], **request_args))
|
||||||
|
|
||||||
|
# send all search-request
|
||||||
grequests.map(requests)
|
grequests.map(requests)
|
||||||
|
|
||||||
|
# update engine-specific stats
|
||||||
for engine_name, engine_results in results.items():
|
for engine_name, engine_results in results.items():
|
||||||
engines[engine_name].stats['search_count'] += 1
|
engines[engine_name].stats['search_count'] += 1
|
||||||
engines[engine_name].stats['result_count'] += len(engine_results)
|
engines[engine_name].stats['result_count'] += len(engine_results)
|
||||||
|
|
||||||
|
# score results and remove duplications
|
||||||
results = score_results(results)
|
results = score_results(results)
|
||||||
|
|
||||||
|
# update engine stats, using calculated score
|
||||||
for result in results:
|
for result in results:
|
||||||
for res_engine in result['engines']:
|
for res_engine in result['engines']:
|
||||||
engines[result['engine']]\
|
engines[result['engine']]\
|
||||||
.stats['score_count'] += result['score']
|
.stats['score_count'] += result['score']
|
||||||
|
|
||||||
|
# return results and suggestions
|
||||||
return results, suggestions
|
return results, suggestions
|
||||||
|
Loading…
Reference in New Issue
Block a user