1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
|
'''
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.path import realpath, dirname, splitext, join
from imp import load_source
import grequests
from itertools import izip_longest, chain
from operator import itemgetter
from urlparse import urlparse
from searx import settings
import ConfigParser
import sys
from datetime import datetime
engine_dir = dirname(realpath(__file__))
searx_dir = join(engine_dir, '../../')
engines_config = ConfigParser.SafeConfigParser()
engines_config.read(join(searx_dir, 'engines.cfg'))
number_of_searches = 0
engines = {}
categories = {'general': []}
def load_module(filename):
modname = splitext(filename)[0]
if modname in sys.modules:
del sys.modules[modname]
filepath = join(engine_dir, filename)
module = load_source(modname, filepath)
module.name = modname
return module
if not engines_config.sections():
print '[E] Error no engines found. Edit your engines.cfg'
exit(2)
for section in engines_config.sections():
engine_data = engines_config.options(section)
engine = load_module(engines_config.get(section, 'engine')+'.py')
engine.name = section
for param_name in engine_data:
if param_name == 'engine':
continue
if param_name == 'categories':
engine.categories = map(str.strip, engines_config.get(section, param_name).split(','))
continue
setattr(engine, param_name, engines_config.get(section, param_name))
for engine_attr in dir(engine):
if engine_attr.startswith('_'):
continue
if getattr(engine, engine_attr) == None:
print '[E] Engine config error: Missing attribute "{0}.{1}"'.format(engine.name, engine_attr)
sys.exit(1)
engines[engine.name] = engine
engine.stats = {'result_count': 0, 'search_count': 0, 'page_load_time': 0, 'score_count': 0, 'errors': 0}
if hasattr(engine, 'categories'):
for category_name in engine.categories:
categories.setdefault(category_name, []).append(engine)
else:
categories['general'].append(engine)
def default_request_params():
return {'method': 'GET', 'headers': {}, 'data': {}, 'url': '', 'cookies': {}}
def make_callback(engine_name, results, callback, params):
def process_callback(response, **kwargs):
cb_res = []
response.search_params = params
engines[engine_name].stats['page_load_time'] += (datetime.now() - params['started']).total_seconds()
try:
search_results = callback(response)
except Exception, e:
engines[engine_name].stats['errors'] += 1
results[engine_name] = cb_res
print '[E] Error with engine "{0}":\n\t{1}'.format(engine_name, str(e))
return
for result in search_results:
result['engine'] = engine_name
cb_res.append(result)
results[engine_name] = cb_res
return process_callback
def search(query, request, selected_engines):
global engines, categories, number_of_searches
requests = []
results = {}
number_of_searches += 1
user_agent = request.headers.get('User-Agent', '')
for selected_engine in selected_engines:
if selected_engine['name'] not in engines:
continue
engine = engines[selected_engine['name']]
request_params = default_request_params()
request_params['headers']['User-Agent'] = user_agent
request_params['category'] = selected_engine['category']
request_params['started'] = datetime.now()
request_params = engine.request(query, request_params)
callback = make_callback(selected_engine['name'], results, engine.response, request_params)
request_args = dict(headers = request_params['headers']
,hooks = dict(response=callback)
,cookies = request_params['cookies']
,timeout = settings.request_timeout
)
if request_params['method'] == 'GET':
req = grequests.get
else:
req = grequests.post
request_args['data'] = request_params['data']
# ignoring empty urls
if not request_params['url']:
continue
requests.append(req(request_params['url'], **request_args))
grequests.map(requests)
for engine_name,engine_results in results.items():
engines[engine_name].stats['search_count'] += 1
engines[engine_name].stats['result_count'] += len(engine_results)
flat_res = filter(None, chain.from_iterable(izip_longest(*results.values())))
flat_len = len(flat_res)
engines_len = len(selected_engines)
results = []
# deduplication + scoring
for i,res in enumerate(flat_res):
res['parsed_url'] = urlparse(res['url'])
res['engines'] = [res['engine']]
score = int((flat_len - i)/engines_len)*settings.weights.get(res['engine'], 1)+1
duplicated = False
for new_res in results:
p1 = res['parsed_url'].path[:-1] if res['parsed_url'].path.endswith('/') else res['parsed_url'].path
p2 = new_res['parsed_url'].path[:-1] if new_res['parsed_url'].path.endswith('/') else new_res['parsed_url'].path
if res['parsed_url'].netloc == new_res['parsed_url'].netloc and\
p1 == p2 and\
res['parsed_url'].query == new_res['parsed_url'].query and\
res.get('template') == new_res.get('template'):
duplicated = new_res
break
if duplicated:
if len(res.get('content', '')) > len(duplicated.get('content', '')):
duplicated['content'] = res['content']
duplicated['score'] += score
duplicated['engines'].append(res['engine'])
if duplicated['parsed_url'].scheme == 'https':
continue
elif res['parsed_url'].scheme == 'https':
duplicated['url'] = res['parsed_url'].geturl()
duplicated['parsed_url'] = res['parsed_url']
else:
res['score'] = score
results.append(res)
for result in results:
for res_engine in result['engines']:
engines[result['engine']].stats['score_count'] += result['score']
return sorted(results, key=itemgetter('score'), reverse=True)
def get_engines_stats():
pageloads = []
results = []
scores = []
errors = []
max_pageload = max_results = max_score = max_errors = 0
for engine in engines.values():
if engine.stats['search_count'] == 0:
continue
results_num = engine.stats['result_count']/float(engine.stats['search_count'])
load_times = engine.stats['page_load_time']/float(engine.stats['search_count'])
if results_num:
score = engine.stats['score_count'] / float(engine.stats['search_count'])
else:
score = 0
max_results = max(results_num, max_results)
max_pageload = max(load_times, max_pageload)
max_score = max(score, max_score)
max_errors = max(max_errors, engine.stats['errors'])
pageloads.append({'avg': load_times, 'name': engine.name})
results.append({'avg': results_num, 'name': engine.name})
scores.append({'avg': score, 'name': engine.name})
errors.append({'avg': engine.stats['errors'], 'name': engine.name})
for engine in pageloads:
engine['percentage'] = int(engine['avg']/max_pageload*100)
for engine in results:
engine['percentage'] = int(engine['avg']/max_results*100)
for engine in scores:
engine['percentage'] = int(engine['avg']/max_score*100)
for engine in errors:
if max_errors:
engine['percentage'] = int(engine['avg']/max_errors*100)
else:
engine['percentage'] = 0
return [('Page loads (sec)', sorted(pageloads, key=itemgetter('avg')))
,('Number of results', sorted(results, key=itemgetter('avg'), reverse=True))
,('Scores', sorted(scores, key=itemgetter('avg'), reverse=True))
,('Errors', sorted(errors, key=itemgetter('avg'), reverse=True))
]
|
