def standardize_citekey(citekey, warn_if_changed=False):
"""
Standardize citation keys based on their source
"""
source, identifier = citekey.split(':', 1)
if source == 'doi':
if identifier.startswith('10/'):
from manubot.cite.doi import expand_short_doi
try:
identifier = expand_short_doi(identifier)
except Exception as error:
# If DOI shortening fails, return the unshortened DOI.
# DOI metadata lookup will eventually fail somewhere with
# appropriate error handling, as opposed to here.
logging.error(f'Error in expand_short_doi for {identifier} '
f'due to a {error.__class__.__name__}:\n{error}')
logging.info(error, exc_info=True)
identifier = identifier.lower()
if source == 'isbn':
from isbnlib import to_isbn13
identifier = to_isbn13(identifier)
standard_citekey = f'{source}:{identifier}'
if warn_if_changed and citekey != standard_citekey:
logging.warning(
f'standardize_citekey expected citekey to already be standardized.\n'
f'Instead citekey was changed from {citekey!r} to {standard_citekey!r}'
)
return standard_citekey
def test_expand_short_doi():
doi = expand_short_doi('10/b6vnmd')
assert doi == "10.1016/s0933-3657(96)00367-3"
def test_expand_short_doi_invalid():
with pytest.raises(ValueError,
match='Handle not found. Double check short_doi'):
expand_short_doi('10/b6vnmdxxxxxx')
def test_expand_short_doi_not_short():
with pytest.raises(ValueError, match='shortDOIs start with `10/`'):
expand_short_doi('10.1016/S0933-3657(96)00367-3')
def canonic(self):
from manubot.cite.doi import expand_short_doi
if self.identifier.startswith('10/'):
self.identifier = expand_short_doi(self.identifier)
return self