def normalize_field_categories(sender, *args, **kwargs):
"""Normalize field_categories."""
for idx, field in enumerate(sender.get('field_categories', [])):
if field.get('scheme') == "INSPIRE" or '_scheme' in field or '_term' in field:
# Already normalized form
continue
original_term = field.get('term')
normalized_term = classify_field(original_term)
scheme = 'INSPIRE' if normalized_term else None
original_scheme = field.get('scheme')
if isinstance(original_scheme, (list, tuple)):
original_scheme = original_scheme[0]
updated_field = {
'_scheme': original_scheme,
'scheme': scheme,
'_term': original_term,
'term': normalized_term,
}
source = field.get('source')
if source:
if 'automatically' in source:
source = 'INSPIRE'
updated_field['source'] = source
sender['field_categories'][idx].update(updated_field)
def normalize_field_categories(sender, *args, **kwargs):
"""Normalize the content of the `field_categories` key.
We use the heuristic that a field is normalized if its scheme is 'INSPIRE'
or if it contains either the `_scheme` key or the `_term` key.
If the field wasn't normalized we use some mapping defined in the
configuration to output a `term` belonging to a known set of values.
We also use the heuristic that the source is 'INSPIRE' if it contains the
word 'automatically', otherwise we preserve it.
"""
def _is_normalized(field):
scheme_is_inspire = field.get('scheme') == 'INSPIRE'
return scheme_is_inspire or '_scheme' in field or '_term' in field
def _is_from_inspire(term):
return term and term != 'Other'
for i, field in enumerate(sender.get('field_categories', [])):
if _is_normalized(field):
continue
original_term = field.get('term')
normalized_term = classify_field(original_term)
scheme = 'INSPIRE' if _is_from_inspire(normalized_term) else None
original_scheme = field.get('scheme')
if isinstance(original_scheme, (list, tuple)):
original_scheme = original_scheme[0]
updated_field = {
'_scheme': original_scheme,
'scheme': scheme,
'_term': original_term,
'term': normalized_term,
}
source = field.get('source')
if source:
if 'automatically' in source:
source = 'INSPIRE'
updated_field['source'] = source
sender['field_categories'][i].update(updated_field)
def test_classify_field_falls_back_on_other():
expected = 'Other'
result = classify_field('FOO')
assert expected == result
def test_classify_field_ignores_case():
expected = 'Astrophysics'
result = classify_field('ASTRO-PH.CO')
assert expected == result
def test_classify_field_returns_category_if_found_among_values():
expected = 'Astrophysics'
result = classify_field('Astrophysics')
assert expected == result
def test_classify_field_returns_category_if_found_among_keys():
expected = 'Math and Math Physics'
result = classify_field('alg-geom')
assert expected == result
def test_classify_field_returns_none_on_non_string_value():
assert classify_field(0) is None
def test_classify_field_returns_none_on_falsy_value():
assert classify_field('') is None
def test_classify_field_returns_none_on_unknown_values():
assert classify_field('FOO') is None
def test_classify_field_ignores_case():
expected = 'Astrophysics'
result = classify_field('ASTRO-PH.CO')
assert expected == result
def test_classify_field_returns_category_if_found_among_values():
expected = 'Astrophysics'
result = classify_field('Astrophysics')
assert expected == result
def test_classify_field_returns_category_if_found_among_keys():
expected = 'Math and Math Physics'
result = classify_field('alg-geom')
assert expected == result
def test_classify_field_returns_none_on_non_string_value():
assert classify_field(0) is None
def test_classify_field_returns_none_on_falsy_value():
assert classify_field('') is None
def test_classify_field_returns_none_on_unknown_values():
assert classify_field('FOO') is None