117d36b · searx · git.iko.soy

searx/engines/json_engine.py+15−1

compare with master

unifiedsplit

@@ -8,6 +8,14 @@ content_query = None

8title_query = None

9# suggestion_xpath = ''

10

11+# parameters for engines with paging support

12+#

13+# number of results on each page

14+# (only needed if the site requires not a page number, but an offset)

15+page_size = 1

16+# number of the first page (usually 0 or 1)

17+first_page_num = 1

18+

11

19

12def iterate(iterable):

20def iterate(iterable):

13 if type(iterable) == dict:

21 if type(iterable) == dict:

@@ -69,8 +77,14 @@ def query(data, query_string):

69

77

70def request(query, params):

78def request(query, params):

71 query = urlencode({'q': query})[2:]

79 query = urlencode({'q': query})[2:]

80+

81+ fp = {'query': query}

82+ if paging and search_url.find('{pageno}') >= 0:

83+ fp['pageno'] = (params['pageno'] + first_page_num - 1) * page_size

84+

72− params['url'] = search_url.format(query=query)

85+ params['url'] = search_url.format(**fp)

73 params['query'] = query

86 params['query'] = query

87+

74 return params

88 return params

75

89

76

90

searx/plugins/__init__.py+3−1

@@ -19,7 +19,8 @@ from searx import logger

19

20logger = logger.getChild('plugins')

21

22−from searx.plugins import (https_rewrite,

22+from searx.plugins import (doai_rewrite,

23+ https_rewrite,

23 open_results_on_new_tab,

24 open_results_on_new_tab,

24 self_info,

25 self_info,

25 search_on_category_select,

26 search_on_category_select,

@@ -73,6 +74,7 @@ class PluginStore():

73

74

75

75plugins = PluginStore()

76plugins = PluginStore()

77+plugins.register(doai_rewrite)

76plugins.register(https_rewrite)

78plugins.register(https_rewrite)

77plugins.register(open_results_on_new_tab)

79plugins.register(open_results_on_new_tab)

78plugins.register(self_info)

80plugins.register(self_info)

searx/plugins/doai_rewrite.pynew file+31

@@ -0,0 +1,31 @@

1+from flask_babel import gettext

2+import re

3+from urlparse import urlparse, parse_qsl

4+

5+regex = re.compile(r'10\.\d{4,9}/[^\s]+')

6+

7+name = gettext('DOAI rewrite')

8+description = gettext('Avoid paywalls by redirecting to open-access versions of publications when available')

9+default_on = False

10+

11+

12+def extract_doi(url):

13+ match = regex.search(url.path)

14+ if match:

15+ return match.group(0)

16+ for _, v in parse_qsl(url.query):

17+ match = regex.search(v)

18+ if match:

19+ return match.group(0)

20+ return None

21+

22+

23+def on_result(request, ctx):

24+ doi = extract_doi(ctx['result']['parsed_url'])

25+ if doi and len(doi) < 50:

26+ for suffix in ('/', '.pdf', '/full', '/meta', '/abstract'):

27+ if doi.endswith(suffix):

28+ doi = doi[:-len(suffix)]

29+ ctx['result']['url'] = 'http://doai.io/' + doi

30+ ctx['result']['parsed_url'] = urlparse(ctx['result']['url'])

31+ return True

searx/settings.yml+10

@@ -87,6 +87,16 @@ engines:

87 - name : btdigg

88 engine : btdigg

89 shortcut : bt

90+

91+ - name : crossref

92+ engine : json_engine

93+ paging : True

94+ search_url : http://search.crossref.org/dois?q={query}&page={pageno}

95+ url_query : doi

96+ title_query : title

97+ content_query : fullCitation

98+ categories : science

99+ shortcut : cr

90

100

91 - name : currency

101 - name : currency

92 engine : currency_convert

102 engine : currency_convert