def test_is_suitable_match(self):
self.assertTrue(author_match.is_suitable_match(self.A1,self.A1))
self.assertTrue(author_match.is_suitable_match(self.C1,self.C1))
self.assertTrue(author_match.is_suitable_match(self.M1,self.M1))
self.assertTrue(author_match.is_suitable_match(self.A1,self.A2))
self.assertTrue(author_match.is_suitable_match(self.C1,self.C2))
self.assertTrue(author_match.is_suitable_match(self.M1,self.M2))
self.assertFalse(author_match.is_suitable_match(self.A1,self.C1))
self.assertFalse(author_match.is_suitable_match(self.A1,self.M1))
self.assertFalse(author_match.is_suitable_match(self.A1,self.C2))
self.assertFalse(author_match.is_suitable_match(self.A1,self.M1))
def test_match_ads_author06(self):
# case where 1987Ap&SS.138...61H was incorrectly matched to 1986tnra.book.....D
# (no match should be take place here)
a1 = {"name": {"western": "Hadrava, P."}, "affiliations": ["Astronomical Institute of the Czechoslovak Academy of Sciences"]}
a2 = {"name": {'western': 'de Waard, Gerrit Jan'}, "affiliations": []}
matches = author_match.match_ads_author_fields([a1],[a2])
self.assertEqual(matches,[(a1,None)])
self.assertFalse(author_match.is_suitable_match(a1,a2))
def authorMerger(self,field='authors'):
data = [ [i[field],i['tempdata']] for i in self.blocks if field in i]
result = None
while len(data) > 0:
f1 = data.pop()
f2 = result if result else data.pop()
result = self._getBestOrigin(f1,f2,'authors')
other = f2 if result == f1 else f1
#Only do the matching if at least one of the the bestOrigin authors lacks an affiliation
#AND the other author field has at least one
if not all( [i['affiliations'] for i in result[0]] ) and\
any( [i['affiliations'] for i in other[0]] ):
best_matches = author_match.match_ads_author_fields(result[0],other[0])
for match in best_matches:
if not author_match.is_suitable_match(*match):
continue
if not match[0]['affiliations'] and match[1]['affiliations']:
match[0]['affiliations'] = match[1]['affiliations']
result = [[i[0] for i in best_matches],result[1]]
return result[0]