def find_article(canidates):
"""
A function that attempts to find a good match
from scholar.py for a set of canidate paper
titles, it returns the best result.
"""
canidate_scores = [5000 for dummy in canidates]
canidate_best_match = ['' for dummy in canidates]
querier = ScholarQuerier()
delay = 0
for ii, canidate in enumerate(canidates):
sleep(delay)
querier.UA = random.choice(USER_AGENTS)
querier.query(canidate)
for art in querier.articles:
title = art['title'].encode('ascii', 'ignore')
score = str_metric(canidate,
title)/max(len(title),len(canidate))
if score < canidate_scores[ii]:
canidate_scores[ii] = score
canidate_best_match[ii] = art
print '----------------------'
print 'Canidate: '+canidate
print 'Match: '+ \
canidate_best_match[ii]['title'].encode('ascii', 'ignore')
print 'Score '+str(canidate_scores[ii])
querier.clear_articles()
delay = max(random.gauss(30, 30), 5)
print('See skip_documents variable in mendeley_add_citations.py to change this.\n')
print('Tags are added immediately. You can interrupt the script and continue later.\n')
print('citations\tyear\tMendeley library title')
num_skipped = 0
documents = mendeley.library(items=-1)
scholar = ScholarQuerier(count=1)
for docid in documents['document_ids']:
document = mendeley.document_details(docid)
if skip_documents and has_citation_tag(document['tags'], ['citations_.*']):
num_skipped = num_skipped+1
continue
try:
scholar.query(document['title'])
scholar_articles = scholar.articles
if len(scholar_articles) == 0:
print('No scholar articles found for ' + document['title'])
continue
except urllib2.HTTPError as e:
print e.msg
print e.reason
sys.exit(-1)
if 'year' in document:
year = document['year']
else:
year = -1
print('%s\t\t%s\t%s' %