on master help sign in
using general mediawiki-engine

* writing general mediawiki-engine * using this engine for wikipedia * using this engine for uncyclopedia

commit52ad49ccba389346e6216dc708891cbea2b4941d authorThomas Pointhuber <thomas.pointhuber@gmx.at> dateWed 3 Sep 2014, 11:40 (12 years ago) parent629a05e
3 files changed+59−80
searx/engines/mediawiki.py+54−9 searx/engines/wikipedia.py−67 searx/settings.yml+5−4
searx/engines/mediawiki.py+54−9
@@ -1,22 +1,67 @@
1+## Wikipedia (Web)
2+#
3+# @website http://www.wikipedia.org
4+# @provide-api yes (http://www.mediawiki.org/wiki/API:Search)
5+#
6+# @using-api yes
7+# @results JSON
8+# @stable yes
9+# @parse url, title
10+#
11+# @todo content
12+
113from json import loads
214from urllib import urlencode, quote
315
4−url = 'https://en.wikipedia.org/'
16+# engine dependent config
17+categories = ['general']
18+language_support = True
19+paging = True
20+number_of_results = 1
21+
22+# search-url
23+base_url = 'https://{language}.wikipedia.org/'
24+search_url = base_url + 'w/api.php?action=query&list=search&{query}&srprop=timestamp&format=json&sroffset={offset}&srlimit={limit}' # noqa
525
6−search_url = url + 'w/api.php?action=query&list=search&{query}&srprop=timestamp&format=json&sroffset={offset}' # noqa
726
8−number_of_results = 10
27+# do search-request
28+def request(query, params):
29+ offset = (params['pageno'] - 1) * number_of_results
930
31+ if params['language'] == 'all':
32+ language = 'en'
33+ else:
34+ language = params['language'].split('_')[0]
35+
36+ # write search-language back to params, required in response
37+ params['language'] = language
1038
11−def request(query, params):
12− offset = (params['pageno'] - 1) * 10
1339 params['url'] = search_url.format(query=urlencode({'srsearch': query}),
14− offset=offset)
40+ offset=offset,
41+ limit=number_of_results,
42+ language=language)
43+
1544 return params
1645
1746
47+# get response from search-request
1848def response(resp):
49+ results = []
50+
1951 search_results = loads(resp.text)
20− res = search_results.get('query', {}).get('search', [])
21− return [{'url': url + 'wiki/' + quote(result['title'].replace(' ', '_').encode('utf-8')), # noqa
22− 'title': result['title']} for result in res[:int(number_of_results)]]
52+
53+ # return empty array if there are no results
54+ if not search_results.get('query', {}).get('search'):
55+ return []
56+
57+ # parse results
58+ for result in search_results['query']['search']:
59+ url = base_url.format(language=resp.search_params['language']) + 'wiki/' + quote(result['title'].replace(' ', '_').encode('utf-8'))
60+
61+ # append result
62+ results.append({'url': url,
63+ 'title': result['title'],
64+ 'content': ''})
65+
66+ # return results
67+ return results
searx/engines/wikipedia.pydeleted−67
@@ -1,67 +0,0 @@
1−## Wikipedia (Web)
2−#
3−# @website http://www.wikipedia.org
4−# @provide-api yes (http://www.mediawiki.org/wiki/API:Search)
5−#
6−# @using-api yes
7−# @results JSON
8−# @stable yes
9−# @parse url, title
10−#
11−# @todo content
12−
13−from json import loads
14−from urllib import urlencode, quote
15−
16−# engine dependent config
17−categories = ['general']
18−language_support = True
19−paging = True
20−number_of_results = 1
21−
22−# search-url
23−url = 'https://{language}.wikipedia.org/'
24−search_url = url + 'w/api.php?action=query&list=search&{query}&srprop=timestamp&format=json&sroffset={offset}&srlimit={limit}' # noqa
25−
26−
27−# do search-request
28−def request(query, params):
29− offset = (params['pageno'] - 1) * number_of_results
30−
31− if params['language'] == 'all':
32− language = 'en'
33− else:
34− language = params['language'].split('_')[0]
35−
36− # write search-language back to params, required in response
37− params['language'] = language
38−
39− params['url'] = search_url.format(query=urlencode({'srsearch': query}),
40− offset=offset,
41− limit=number_of_results,
42− language=language)
43−
44− return params
45−
46−
47−# get response from search-request
48−def response(resp):
49− results = []
50−
51− search_results = loads(resp.text)
52−
53− # return empty array if there are no results
54− if not search_results.get('query', {}).get('search'):
55− return []
56−
57− # parse results
58− for result in search_results['query']['search']:
59− res_url = url.format(language=resp.search_params['language']) + 'wiki/' + quote(result['title'].replace(' ', '_').encode('utf-8'))
60−
61− # append result
62− results.append({'url': res_url,
63− 'title': result['title'],
64− 'content': ''})
65−
66− # return results
67− return results
searx/settings.yml+5−4
@@ -10,9 +10,10 @@ server:
1010
1111engines:
1212 - name : wikipedia
13− engine : wikipedia
13+ engine : mediawiki
1414 shortcut : wp
15−# number_of_results : 1 # default is 1
15+ base_url : 'https://{language}.wikipedia.org/'
16+ number_of_results : 1
1617
1718 - name : bing
1819 engine : bing
@@ -108,9 +109,9 @@ engines:
108109# maybe in a fun category
109110# - name : uncyclopedia
110111# engine : mediawiki
111−# categories : general
112112# shortcut : unc
113−# url : https://uncyclopedia.wikia.com/
113+# base_url : https://uncyclopedia.wikia.com/
114+# number_of_results : 5
114115
115116# tmp suspended - too slow, too many errors
116117# - name : urbandictionary