on master help sign in
fix wikipedia engine and add comments

* add paging support * make number_of_results changable * make result calculation more clear * add comments

commitbb628469d31d9ce61b2188aae3f570441eec8803 authorThomas Pointhuber <thomas.pointhuber@gmx.at> dateTue 2 Sep 2014, 21:01 (12 years ago) parent8eb064d
2 files changed+48−12
searx/engines/wikipedia.py+47−10 searx/settings.yml+1−2
searx/engines/wikipedia.py+47−10
@@ -1,30 +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://{language}.wikipedia.org/'
5−
6−search_url = url + 'w/api.php?action=query&list=search&{query}&srprop=timestamp&format=json&sroffset={offset}' # noqa
7−
8−number_of_results = 10
9−
16+# engine dependent config
17+categories = ['general']
1018language_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
1125
1226
27+# do search-request
1328def request(query, params):
14− offset = (params['pageno'] - 1) * 10
29+ offset = (params['pageno'] - 1) * number_of_results
30+
1531 if params['language'] == 'all':
1632 language = 'en'
1733 else:
1834 language = params['language'].split('_')[0]
35+
36+ # write search-language back to params, required in response
1937 params['language'] = language
38+
2039 params['url'] = search_url.format(query=urlencode({'srsearch': query}),
2140 offset=offset,
41+ limit=number_of_results,
2242 language=language)
43+
2344 return params
2445
2546
47+# get response from search-request
2648def response(resp):
49+ results = []
50+
2751 search_results = loads(resp.text)
28− res = search_results.get('query', {}).get('search', [])
29− return [{'url': url.format(language=resp.search_params['language']) + 'wiki/' + quote(result['title'].replace(' ', '_').encode('utf-8')), # noqa
30− '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+ 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+1−2
@@ -11,9 +11,8 @@ server:
1111engines:
1212 - name : wikipedia
1313 engine : wikipedia
14− number_of_results : 1
15− paging : False
1614 shortcut : wp
15+# number_of_results : 1 # default is 1
1716
1817 - name : bing
1918 engine : bing