searx

commit: 96d623436c9ed9cbd7a8f7dbd8852e9100a9b7b7
parent: 53c9185cbeb05026909adec164b859e09850c74e
Author: Adam Tauber <asciimoo@gmail.com>
Date:   Wed, 10 Aug 2016 11:39:00 +0200

Merge pull request #630 from davidar/doi

Add Crossref search engine and DOAI rewrite plugin
Diffstat:
M
searx/engines/json_engine.py
16
+++++++++++++++-
M
searx/plugins/__init__.py
4
+++-
A
searx/plugins/doai_rewrite.py
31
+++++++++++++++++++++++++++++++
M
searx/settings.yml
10
++++++++++

4 files changed, 59 insertions(+), 2 deletions(-)
diff --git a/searx/engines/json_engine.py b/searx/engines/json_engine.py
@@ -8,6 +8,14 @@ content_query = None
 title_query = None
 # suggestion_xpath = ''
 
+# parameters for engines with paging support
+#
+# number of results on each page
+# (only needed if the site requires not a page number, but an offset)
+page_size = 1
+# number of the first page (usually 0 or 1)
+first_page_num = 1
+
 
 def iterate(iterable):
     if type(iterable) == dict:
@@ -69,8 +77,14 @@ def query(data, query_string):
 
 def request(query, params):
     query = urlencode({'q': query})[2:]
-    params['url'] = search_url.format(query=query)
+
+    fp = {'query': query}
+    if paging and search_url.find('{pageno}') >= 0:
+        fp['pageno'] = (params['pageno'] + first_page_num - 1) * page_size
+
+    params['url'] = search_url.format(**fp)
     params['query'] = query
+
     return params
 
 
diff --git a/searx/plugins/__init__.py b/searx/plugins/__init__.py
@@ -19,7 +19,8 @@ from searx import logger
 
 logger = logger.getChild('plugins')
 
-from searx.plugins import (https_rewrite,
+from searx.plugins import (doai_rewrite,
+                           https_rewrite,
                            open_results_on_new_tab,
                            self_info,
                            search_on_category_select,
@@ -73,6 +74,7 @@ class PluginStore():
 
 
 plugins = PluginStore()
+plugins.register(doai_rewrite)
 plugins.register(https_rewrite)
 plugins.register(open_results_on_new_tab)
 plugins.register(self_info)
diff --git a/searx/plugins/doai_rewrite.py b/searx/plugins/doai_rewrite.py
@@ -0,0 +1,31 @@
+from flask_babel import gettext
+import re
+from urlparse import urlparse, parse_qsl
+
+regex = re.compile(r'10\.\d{4,9}/[^\s]+')
+
+name = gettext('DOAI rewrite')
+description = gettext('Avoid paywalls by redirecting to open-access versions of publications when available')
+default_on = False
+
+
+def extract_doi(url):
+    match = regex.search(url.path)
+    if match:
+        return match.group(0)
+    for _, v in parse_qsl(url.query):
+        match = regex.search(v)
+        if match:
+            return match.group(0)
+    return None
+
+
+def on_result(request, ctx):
+    doi = extract_doi(ctx['result']['parsed_url'])
+    if doi and len(doi) < 50:
+        for suffix in ('/', '.pdf', '/full', '/meta', '/abstract'):
+            if doi.endswith(suffix):
+                doi = doi[:-len(suffix)]
+        ctx['result']['url'] = 'http://doai.io/' + doi
+        ctx['result']['parsed_url'] = urlparse(ctx['result']['url'])
+    return True
diff --git a/searx/settings.yml b/searx/settings.yml
@@ -87,6 +87,16 @@ engines:
   - name : btdigg
     engine : btdigg
     shortcut : bt
+    
+  - name : crossref
+    engine : json_engine
+    paging : True
+    search_url : http://search.crossref.org/dois?q={query}&page={pageno}
+    url_query : doi
+    title_query : title
+    content_query : fullCitation
+    categories : science
+    shortcut : cr
 
   - name : currency
     engine : currency_convert