summaryrefslogtreecommitdiff
path: root/searx/engines/semantic_scholar.py
diff options
context:
space:
mode:
Diffstat (limited to 'searx/engines/semantic_scholar.py')
-rw-r--r--searx/engines/semantic_scholar.py57
1 files changed, 41 insertions, 16 deletions
diff --git a/searx/engines/semantic_scholar.py b/searx/engines/semantic_scholar.py
index bda731047..b2701c333 100644
--- a/searx/engines/semantic_scholar.py
+++ b/searx/engines/semantic_scholar.py
@@ -6,6 +6,8 @@
from json import dumps, loads
from datetime import datetime
+from flask_babel import gettext
+
about = {
"website": 'https://www.semanticscholar.org/',
"wikidata_id": 'Q22908627',
@@ -15,6 +17,7 @@ about = {
"results": 'JSON',
}
+categories = ['science', 'scientific publications']
paging = True
search_url = 'https://www.semanticscholar.org/api/1/search'
paper_url = 'https://www.semanticscholar.org/paper'
@@ -47,9 +50,6 @@ def response(resp):
results = []
for result in res['results']:
- item = {}
- metadata = []
-
url = result.get('primaryPaperLink', {}).get('url')
if not url and result.get('links'):
url = result.get('links')[0]
@@ -60,22 +60,47 @@ def response(resp):
if not url:
url = paper_url + '/%s' % result['id']
- item['url'] = url
+ # publishedDate
+ if 'pubDate' in result:
+ publishedDate = datetime.strptime(result['pubDate'], "%Y-%m-%d")
+ else:
+ publishedDate = None
- item['title'] = result['title']['text']
- item['content'] = result['paperAbstract']['text']
+ # authors
+ authors = [author[0]['name'] for author in result.get('authors', [])]
- metadata = result.get('fieldsOfStudy') or []
- venue = result.get('venue', {}).get('text')
- if venue:
- metadata.append(venue)
- if metadata:
- item['metadata'] = ', '.join(metadata)
+ # pick for the first alternate link, but not from the crawler
+ pdf_url = None
+ for doc in result.get('alternatePaperLinks', []):
+ if doc['linkType'] != 'crawler':
+ pdf_url = doc['url']
+ break
- pubDate = result.get('pubDate')
- if pubDate:
- item['publishedDate'] = datetime.strptime(pubDate, "%Y-%m-%d")
+ # comments
+ comments = None
+ if 'citationStats' in result:
+ comments = gettext(
+ '{numCitations} citations from the year {firstCitationVelocityYear} to {lastCitationVelocityYear}'
+ ).format(
+ numCitations=result['citationStats']['numCitations'],
+ firstCitationVelocityYear=result['citationStats']['firstCitationVelocityYear'],
+ lastCitationVelocityYear=result['citationStats']['lastCitationVelocityYear'],
+ )
- results.append(item)
+ results.append(
+ {
+ 'template': 'paper.html',
+ 'url': url,
+ 'title': result['title']['text'],
+ 'content': result['paperAbstract']['text'],
+ 'journal': result.get('venue', {}).get('text') or result.get('journal', {}).get('name'),
+ 'doi': result.get('doiInfo', {}).get('doi'),
+ 'tags': result.get('fieldsOfStudy'),
+ 'authors': authors,
+ 'pdf_url': pdf_url,
+ 'publishedDate': publishedDate,
+ 'comments': comments,
+ }
+ )
return results