|
@@ -0,0 +1,127 @@
|
|
1
|
+#!/usr/bin/env python
|
|
2
|
+
|
|
3
|
+'''
|
|
4
|
+searx is free software: you can redistribute it and/or modify
|
|
5
|
+it under the terms of the GNU Affero General Public License as published by
|
|
6
|
+the Free Software Foundation, either version 3 of the License, or
|
|
7
|
+(at your option) any later version.
|
|
8
|
+
|
|
9
|
+searx is distributed in the hope that it will be useful,
|
|
10
|
+but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
11
|
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
12
|
+GNU Affero General Public License for more details.
|
|
13
|
+
|
|
14
|
+You should have received a copy of the GNU Affero General Public License
|
|
15
|
+along with searx. If not, see < http://www.gnu.org/licenses/ >.
|
|
16
|
+
|
|
17
|
+(C) 2014 by Thomas Pointhuber, <thomas.pointhuber@gmx.at>
|
|
18
|
+'''
|
|
19
|
+
|
|
20
|
+from searx.languages import language_codes
|
|
21
|
+from searx.engines import (
|
|
22
|
+ categories, engines, engine_shortcuts
|
|
23
|
+)
|
|
24
|
+import string
|
|
25
|
+import re
|
|
26
|
+
|
|
27
|
+
|
|
28
|
+class Query(object):
|
|
29
|
+ """parse query"""
|
|
30
|
+
|
|
31
|
+ def __init__(self, query, blocked_engines):
|
|
32
|
+ self.query = query
|
|
33
|
+ self.blocked_engines = []
|
|
34
|
+
|
|
35
|
+ if blocked_engines:
|
|
36
|
+ self.blocked_engines = blocked_engines
|
|
37
|
+
|
|
38
|
+ self.query_parts = []
|
|
39
|
+ self.engines = []
|
|
40
|
+ self.languages = []
|
|
41
|
+
|
|
42
|
+ # parse query, if tags are set, which change the serch engine or search-language
|
|
43
|
+ def parse_query(self):
|
|
44
|
+ self.query_parts = []
|
|
45
|
+
|
|
46
|
+ # split query, including whitespaces
|
|
47
|
+ raw_query_parts = re.split(r'(\s+)', self.query)
|
|
48
|
+
|
|
49
|
+ parse_next = True
|
|
50
|
+
|
|
51
|
+ for query_part in raw_query_parts:
|
|
52
|
+ if not parse_next:
|
|
53
|
+ self.query_parts[-1] += query_part
|
|
54
|
+ continue
|
|
55
|
+
|
|
56
|
+ parse_next = False
|
|
57
|
+
|
|
58
|
+ # part does only contain spaces, skip
|
|
59
|
+ if query_part.isspace()\
|
|
60
|
+ or query_part == '':
|
|
61
|
+ parse_next = True
|
|
62
|
+ self.query_parts.append(query_part)
|
|
63
|
+ continue
|
|
64
|
+
|
|
65
|
+ # this force a language
|
|
66
|
+ if query_part[0] == ':':
|
|
67
|
+ lang = query_part[1:].lower()
|
|
68
|
+
|
|
69
|
+ # check if any language-code is equal with declared language-codes
|
|
70
|
+ for lc in language_codes:
|
|
71
|
+ lang_id, lang_name, country = map(str.lower, lc)
|
|
72
|
+
|
|
73
|
+ # if correct language-code is found, set it as new search-language
|
|
74
|
+ if lang == lang_id\
|
|
75
|
+ or lang_id.startswith(lang)\
|
|
76
|
+ or lang == lang_name\
|
|
77
|
+ or lang == country:
|
|
78
|
+ parse_next = True
|
|
79
|
+ self.languages.append(lang)
|
|
80
|
+ break
|
|
81
|
+
|
|
82
|
+ # this force a engine or category
|
|
83
|
+ if query_part[0] == '!':
|
|
84
|
+ prefix = query_part[1:].replace('_', ' ')
|
|
85
|
+
|
|
86
|
+ # check if prefix is equal with engine shortcut
|
|
87
|
+ if prefix in engine_shortcuts\
|
|
88
|
+ and not engine_shortcuts[prefix] in self.blocked_engines:
|
|
89
|
+ parse_next = True
|
|
90
|
+ self.engines.append({'category': 'none',
|
|
91
|
+ 'name': engine_shortcuts[prefix]})
|
|
92
|
+
|
|
93
|
+ # check if prefix is equal with engine name
|
|
94
|
+ elif prefix in engines\
|
|
95
|
+ and not prefix in self.blocked_engines:
|
|
96
|
+ parse_next = True
|
|
97
|
+ self.engines.append({'category': 'none',
|
|
98
|
+ 'name': prefix})
|
|
99
|
+
|
|
100
|
+ # check if prefix is equal with categorie name
|
|
101
|
+ elif prefix in categories:
|
|
102
|
+ # using all engines for that search, which are declared under that categorie name
|
|
103
|
+ parse_next = True
|
|
104
|
+ self.engines.extend({'category': prefix,
|
|
105
|
+ 'name': engine.name}
|
|
106
|
+ for engine in categories[prefix]
|
|
107
|
+ if not engine in self.blocked_engines)
|
|
108
|
+
|
|
109
|
+ # append query part to query_part list
|
|
110
|
+ self.query_parts.append(query_part)
|
|
111
|
+
|
|
112
|
+ def changeSearchQuery(self, search_query):
|
|
113
|
+ if len(self.query_parts):
|
|
114
|
+ self.query_parts[-1] = search_query
|
|
115
|
+ else:
|
|
116
|
+ self.query_parts.append(search_query)
|
|
117
|
+
|
|
118
|
+ def getSearchQuery(self):
|
|
119
|
+ if len(self.query_parts):
|
|
120
|
+ return self.query_parts[-1]
|
|
121
|
+ else:
|
|
122
|
+ return ''
|
|
123
|
+
|
|
124
|
+ def getFullQuery(self):
|
|
125
|
+ # get full querry including whitespaces
|
|
126
|
+ return string.join(self.query_parts, '')
|
|
127
|
+
|