def test_simple(self):
self.assertEqual(parse_comma_name('Claire Mathieu'),
('Claire', 'Mathieu'))
self.assertEqual(parse_comma_name('Mathieu, Claire'),
('Claire', 'Mathieu'))
self.assertEqual(parse_comma_name('Kenyon-Mathieu, Claire'),
('Claire', 'Kenyon-Mathieu'))
self.assertEqual(parse_comma_name('Arvind'), ('', 'Arvind'))
def test_simple(self):
self.assertEqual(parse_comma_name(
'Claire Mathieu'), ('Claire', 'Mathieu'))
self.assertEqual(parse_comma_name(
'Mathieu, Claire'), ('Claire', 'Mathieu'))
self.assertEqual(parse_comma_name(
'Kenyon-Mathieu, Claire'), ('Claire', 'Kenyon-Mathieu'))
self.assertEqual(parse_comma_name('Arvind'), ('', 'Arvind'))
def test_hard_cases(self):
# TODO ?
self.assertEqual(parse_comma_name('W. Timothy Gowers'),
('W. Timothy', 'Gowers'))
self.assertEqual(parse_comma_name('Guido van Rossum'),
('Guido', 'van Rossum'))
self.assertEqual(parse_comma_name('Éric Colin de Verdière'),
('Éric', 'Colin de Verdière'))
def test_hard_cases(self):
# TODO ?
self.assertEqual(parse_comma_name('W. Timothy Gowers'),
('W. Timothy', 'Gowers'))
self.assertEqual(parse_comma_name('Guido van Rossum'),
('Guido', 'van Rossum'))
self.assertEqual(parse_comma_name(
'Éric Colin de Verdière'), ('Éric', 'Colin de Verdière'))
def convert_to_name_pair(dct):
""" Converts a dictionary {'family':'Last','given':'First'} to ('First','Last') """
result = None
if 'family' in dct and 'given' in dct:
result = (dct['given'],dct['family'])
elif 'family' in dct: # The 'Arvind' case
result = ('',dct['family'])
elif 'literal' in dct:
result = parse_comma_name(dct['literal'])
if result:
result = (normalize_name_words(result[0]), normalize_name_words(result[1]))
return result
def name(self):
"""
Returns a parsed version of the "credit name" in the ORCID profile.
If there is no such name, returns the given and family names on the profile
(they should exist)
"""
name_item = jpath('orcid-profile/orcid-bio/personal-details', self.json)
name = jpath('credit-name/value', name_item)
if name is not None:
return parse_comma_name(name)
return (normalize_name_words(jpath('given-names/value', name_item, '')),
normalize_name_words(jpath('family-name/value', name_item, '')))
def convert_to_name_pair(dct):
""" Converts a dictionary {'family':'Last','given':'First'} to ('First','Last') """
result = None
if 'family' in dct and 'given' in dct:
result = (dct['given'],dct['family'])
elif 'family' in dct: # The 'Arvind' case
result = ('',dct['family'])
elif 'literal' in dct:
result = parse_comma_name(dct['literal'])
if result:
result = (normalize_name_words(result[0]), normalize_name_words(result[1]))
return result
def name(self):
"""
Returns a parsed version of the "credit name" in the ORCID profile.
If there is no such name, returns the given and family names on the profile
(they should exist)
"""
name_item = jpath('person/name', self.json)
name = jpath('credit-name/value', name_item)
if name:
return parse_comma_name(name)
return (normalize_name_words(jpath('given-names/value', name_item,
'')),
normalize_name_words(jpath('family-name/value', name_item,
'')))
def other_names(self):
"""
Returns the list of other names listed on the ORCiD profile.
This includes the (given,family) name if a credit name was defined.
"""
name_item = jpath('orcid-profile/orcid-bio/personal-details', self.json)
names = []
credit_name = jpath('credit-name/value', name_item)
if credit_name is not None:
names.append((normalize_name_words(jpath('given-names/value', name_item)),
normalize_name_words(jpath('family-name/value', name_item))))
other_names = jpath('other-names/other-name', name_item, default=[])
for name in other_names:
val = name.get('value')
if val is not None:
names.append(parse_comma_name(val))
return names
def other_names(self):
"""
Returns the list of other names listed on the ORCiD profile.
This includes the (given,family) name if a credit name was defined.
"""
person = jpath('person', self.json)
names = []
credit_name = jpath('name/credit-name/value', person)
if credit_name is not None:
names.append((normalize_name_words(
jpath('name/given-names/value', person, '')),
normalize_name_words(
jpath('name/family-name/value', person, ''))))
other_names = jpath('other-names/other-name', person, default=[])
for name in other_names:
val = name.get('content')
if val is not None:
names.append(parse_comma_name(val))
return names
def parse_authors_list(record):
"""
Split author field into a list of (First name, Last name)
"""
if 'author' in record:
if record['author']:
# Handle "et al"
record['author'] = ET_AL_RE.sub('', record['author'])
# Handle "others"
record['author'] = OTHERS_RE.sub('', record['author'])
# Normalizations
record['author'] = record['author'].replace('\n', ' ')
# Split author field into list of first and last names
record['author'] = [
parse_comma_name(author.strip())
for author in record['author'].split(' and ')
]
else:
del record['author']
return record
def parse_authors_list(record):
"""
Split author field into a list of (First name, Last name)
"""
if 'author' in record:
if record['author']:
# Handle "et al"
record['author'] = ET_AL_RE.sub('', record['author'])
# Handle "others"
record['author'] = OTHERS_RE.sub('', record['author'])
# Normalizations
record['author'] = record['author'].replace('\n', ' ')
# Split author field into list of first and last names
record['author'] = [
parse_comma_name(author.strip())
for author in record['author'].split(' and ')
]
else:
del record['author']
return record
def other_names(self):
"""
Returns the list of other names listed on the ORCiD profile.
This includes the (given,family) name if a credit name was defined.
"""
name_item = jpath('orcid-profile/orcid-bio/personal-details',
self.json)
names = []
credit_name = jpath('credit-name/value', name_item)
if credit_name is not None:
names.append(
(normalize_name_words(jpath('given-names/value', name_item,
'')),
normalize_name_words(jpath('family-name/value', name_item,
''))))
other_names = jpath('other-names/other-name', name_item, default=[])
for name in other_names:
val = name.get('value')
if val is not None:
names.append(parse_comma_name(val))
return names
def api_paper_query(request):
try:
fields = json.loads(request.body.decode('utf-8'))
except (ValueError, UnicodeDecodeError):
raise BadRequest('Invalid JSON payload')
doi = fields.get('doi')
if doi:
p = None
try:
p = Paper.create_by_doi(doi, bare=True)
except MetadataSourceException:
pass
if p is None:
raise BadRequest('Could not find a paper with this DOI')
return {'status': 'ok', 'paper': p.json()}
title = fields.get('title')
if not isinstance(title, unicode) or not title or len(title) > 512:
raise BadRequest(
'Invalid title, has to be a non-empty string shorter than 512 characters'
)
date = fields.get('date')
if not isinstance(date, unicode):
raise BadRequest('A date is required')
try:
date = tolerant_datestamp_to_datetime(date)
except ValueError as e:
raise BadRequest(unicode(e))
authors = fields.get('authors')
if not isinstance(authors, list):
raise BadRequest('A list of authors is expected')
parsed_authors = []
for a in authors:
author = None
if not isinstance(a, dict):
raise BadRequest('Invalid author')
if 'first' in a and 'last' in a:
if not isinstance(a['first'], unicode) or not isinstance(
a['last'], unicode) or not a['last']:
raise BadRequest('Invalid (first,last) name provided')
else:
author = (a['first'], a['last'])
elif 'plain' in a:
if not isinstance(a['plain'], unicode) or not a['plain']:
raise BadRequest('Invalid plain name provided')
else:
author = parse_comma_name(a['plain'])
if author is None:
raise BadRequest('Invalid author')
parsed_authors.append(BareName.create(author[0], author[1]))
if not authors:
raise BadRequest('No authors provided')
try:
p = BarePaper.create(title, parsed_authors, date)
except ValueError:
raise BadRequest('Invalid paper')
return {'status': 'ok', 'paper': p.json()}
def test_initial_capitalized(self):
self.assertEqual(parse_comma_name(
'MATHIEU Claire'), ('Claire', 'Mathieu'))
self.assertEqual(parse_comma_name('MATHIEU C.'), ('C.', 'Mathieu'))
def test_collapsed_initials(self):
self.assertEqual(parse_comma_name('Badiou CS'), ('C. S.', 'Badiou'))
self.assertEqual(parse_comma_name('Tony LI'), ('Tony', 'Li'))
def test_middle_initials(self):
self.assertEqual(parse_comma_name('Neal E. Young'),
('Neal E.', 'Young'))
def test_final_initials(self):
self.assertEqual(parse_comma_name('Mathieu C.'), ('C.', 'Mathieu'))
self.assertEqual(parse_comma_name('Gowers W. T..'),
('W. T.', 'Gowers'))
def test_initial_initials(self):
self.assertEqual(parse_comma_name('C. Mathieu'), ('C.', 'Mathieu'))
self.assertEqual(parse_comma_name('N. E. Young'), ('N. E.', 'Young'))
def test_middle_initials(self):
self.assertEqual(parse_comma_name(
'Neal E. Young'), ('Neal E.', 'Young'))
def test_initial_capitalized(self):
self.assertEqual(parse_comma_name('MATHIEU Claire'),
('Claire', 'Mathieu'))
self.assertEqual(parse_comma_name('MATHIEU C.'), ('C.', 'Mathieu'))
def test_final_capitalized(self):
self.assertEqual(parse_comma_name(
'Claire MATHIEU'), ('Claire', 'Mathieu'))
self.assertEqual(parse_comma_name('C. MATHIEU'), ('C.', 'Mathieu'))
def test_initial_initials(self):
self.assertEqual(parse_comma_name('C. Mathieu'), ('C.', 'Mathieu'))
self.assertEqual(parse_comma_name('N. E. Young'), ('N. E.', 'Young'))
def test_final_initials(self):
self.assertEqual(parse_comma_name('Mathieu C.'), ('C.', 'Mathieu'))
self.assertEqual(parse_comma_name('Gowers W. T..'), ('W. T.', 'Gowers'))
def api_paper_query(request):
try:
fields = json.loads(request.body.decode('utf-8'))
except (ValueError, UnicodeDecodeError):
raise BadRequest('Invalid JSON payload')
doi = fields.get('doi')
if doi:
p = None
try:
p = Paper.create_by_doi(doi, bare=True)
except MetadataSourceException:
pass
if p is None:
raise BadRequest('Could not find a paper with this DOI')
return {'status':'ok','paper':p.json()}
title = fields.get('title')
if type(title) != unicode or not title or len(title) > 512:
raise BadRequest('Invalid title, has to be a non-empty string shorter than 512 characters')
pubdate = fields.get('date')
date = fields.get('date')
if type(date) != unicode:
raise BadRequest('A date is required')
try:
date = tolerant_datestamp_to_datetime(date)
except ValueError as e:
raise BadRequest(unicode(e))
authors = fields.get('authors')
if type(authors) != list:
raise BadRequest('A list of authors is expected')
parsed_authors = []
for a in authors:
author = None
if type(a) != dict:
raise BadRequest('Invalid author')
if 'first' in a and 'last' in a:
if type(a['first']) != unicode or type(a['last']) != unicode or not a['last']:
raise BadRequest('Invalid (first,last) name provided')
else:
author = (a['first'],a['last'])
elif 'plain' in a:
if type(a['plain']) != unicode or not a['plain']:
raise BadRequest('Invalid plain name provided')
else:
author = parse_comma_name(a['plain'])
if author is None:
raise BadRequest('Invalid author')
parsed_authors.append(BareName.create(author[0],author[1]))
if not authors:
raise BadRequest('No authors provided')
try:
p = BarePaper.create(title, parsed_authors, date)
except ValueError:
raise BadRequest('Invalid paper')
import backend.oai as oai
oaisource = oai.OaiPaperSource(max_results=10)
p = oaisource.fetch_accessibility(p)
return {'status':'ok','paper':p.json()}
def api_paper_query(request):
try:
fields = json.loads(request.body.decode('utf-8'))
except (ValueError, UnicodeDecodeError):
raise BadRequest('Invalid JSON payload')
doi = fields.get('doi')
if doi:
p = None
try:
p = Paper.get_by_doi(doi)
if not p:
p = Paper.create_by_doi(doi)
except MetadataSourceException:
pass
if p is None:
raise BadRequest('Could not find a paper with this DOI')
return {'status': 'ok', 'paper': p.json()}
title = fields.get('title')
if not isinstance(title, str) or not title or len(title) > 512:
raise BadRequest(
'Invalid title, has to be a non-empty string shorter than 512 characters'
)
date = fields.get('date')
if not isinstance(date, str):
raise BadRequest('A date is required')
try:
date = tolerant_datestamp_to_datetime(date)
except ValueError as e:
raise BadRequest(str(e))
authors = fields.get('authors')
if not isinstance(authors, list):
raise BadRequest('A list of authors is expected')
parsed_authors = []
for a in authors:
author = None
if not isinstance(a, dict):
raise BadRequest('Invalid author')
if 'first' in a and 'last' in a:
if not isinstance(a['first'], str) or not isinstance(
a['last'], str) or not a['last']:
raise BadRequest('Invalid (first,last) name provided')
else:
author = (a['first'], a['last'])
elif 'plain' in a:
if not isinstance(a['plain'], str) or not a['plain']:
raise BadRequest('Invalid plain name provided')
else:
author = parse_comma_name(a['plain'])
if author is None:
raise BadRequest('Invalid author')
parsed_authors.append(BareName.create(author[0], author[1]))
if not authors:
raise BadRequest('No authors provided')
try:
# Validate the metadata against our data model,
# and compute the fingerprint to look up the paper in the DB.
# This does NOT create a paper in the database - we do not want
# to create papers for every search query we get!
p = BarePaper.create(title, parsed_authors, date)
except ValueError as e:
raise BadRequest('Invalid paper: {}'.format(e))
try:
model_paper = Paper.objects.get(fingerprint=p.fingerprint)
return {'status': 'ok', 'paper': model_paper.json()}
except Paper.DoesNotExist:
return {'status': 'not found'}, 404
def test_final_capitalized(self):
self.assertEqual(parse_comma_name('Claire MATHIEU'),
('Claire', 'Mathieu'))
self.assertEqual(parse_comma_name('C. MATHIEU'), ('C.', 'Mathieu'))
def api_paper_query(request):
try:
fields = json.loads(request.body.decode('utf-8'))
except (ValueError, UnicodeDecodeError):
raise BadRequest('Invalid JSON payload')
doi = fields.get('doi')
if doi:
p = None
try:
p = Paper.get_by_doi(doi)
if not p:
p = Paper.create_by_doi(doi)
except MetadataSourceException:
pass
if p is None:
raise BadRequest('Could not find a paper with this DOI')
return {'status': 'ok', 'paper': p.json()}
title = fields.get('title')
if not isinstance(title, str) or not title or len(title) > 512:
raise BadRequest(
'Invalid title, has to be a non-empty string shorter than 512 characters')
date = fields.get('date')
if not isinstance(date, str):
raise BadRequest('A date is required')
try:
date = tolerant_datestamp_to_datetime(date)
except ValueError as e:
raise BadRequest(str(e))
authors = fields.get('authors')
if not isinstance(authors, list):
raise BadRequest('A list of authors is expected')
parsed_authors = []
for a in authors:
author = None
if not isinstance(a, dict):
raise BadRequest('Invalid author')
if 'first' in a and 'last' in a:
if not isinstance(a['first'], str) or not isinstance(a['last'], str) or not a['last']:
raise BadRequest('Invalid (first,last) name provided')
else:
author = (a['first'], a['last'])
elif 'plain' in a:
if not isinstance(a['plain'], str) or not a['plain']:
raise BadRequest('Invalid plain name provided')
else:
author = parse_comma_name(a['plain'])
if author is None:
raise BadRequest('Invalid author')
parsed_authors.append(BareName.create(author[0], author[1]))
if not authors:
raise BadRequest('No authors provided')
try:
# Validate the metadata against our data model,
# and compute the fingerprint to look up the paper in the DB.
# This does NOT create a paper in the database - we do not want
# to create papers for every search query we get!
p = BarePaper.create(title, parsed_authors, date)
except ValueError as e:
raise BadRequest('Invalid paper: {}'.format(e))
try:
model_paper = Paper.objects.get(fingerprint=p.fingerprint)
return {'status': 'ok', 'paper': model_paper.json()}
except Paper.DoesNotExist:
return {'status': 'not found'}, 404
def test_collapsed_initials(self):
self.assertEqual(parse_comma_name('Badiou CS'), ('C. S.', 'Badiou'))
self.assertEqual(parse_comma_name('Tony LI'), ('Tony', 'Li'))
