def test_anonymize_keeps_free_text_untouched(patient_report):
header, free_text, final_text = patient_report()
text = TextAdapter(final_text)
text.anonymize()
assert free_text in text.data
def test_text_adapter_can_anonymize_data(patient_report):
header, free_text, final_text = patient_report()
text = TextAdapter(final_text)
text.anonymize()
for key in header.keys():
assert key not in text.data
def test_original_id_property_is_set_after_a_call_to_anonymize(patient_report):
header, free_text, final_text = patient_report()
text = TextAdapter(final_text)
text.anonymize()
assert text.original_id
def test_digest_code_only_depends_on_patient_data():
text1 = TextAdapter('no patient data')
text2 = TextAdapter('but I do not have patient data either')
text1.anonymize()
text2.anonymize()
assert text1.digest_code == text2.digest_code
def test_original_id_cannot_be_accesed_without_anonymize_call(patient_report):
header, free_text, final_text = patient_report()
text = TextAdapter(final_text)
with pytest.raises(RuntimeError, match=r'Text has not been anonymized'):
text.original_id
def test_different_headers_create_different_digest_codes(patient_report):
header, free_text, final_text = patient_report()
header2, free_text2, final_text2 = patient_report()
text1 = TextAdapter(final_text)
text2 = TextAdapter(final_text2)
text1.anonymize()
text2.anonymize()
assert text1.digest_code != text2.digest_code
def test_two_equal_patients_have_the_same_digest_code(patient_report):
header, free_text, final_text = patient_report()
header2, free_text2, final_text2 = patient_report(header=header)
text1 = TextAdapter(final_text)
text2 = TextAdapter(final_text2)
text1.anonymize()
text2.anonymize()
assert text1.digest_code == text2.digest_code
def test_patient_data_order_is_irrelevant_on_digest_codes(patient_report):
header, free_text, final_text = patient_report()
header2, free_text2, final_text2 = patient_report(header=header,
ordered=False)
text1 = TextAdapter(final_text)
text2 = TextAdapter(final_text2)
text1.anonymize()
text2.anonymize()
assert text1.digest_code == text2.digest_code
def test_multiple_calls_to_anonymize_do_not_change_data(patient_report):
header, free_text, final_text = patient_report()
text = TextAdapter(final_text)
text.anonymize()
previous_anonymous_data = text.data
previous_digest_code = text.digest_code
previous_original_id = text.original_id
text.anonymize()
assert text.data == previous_anonymous_data
assert text.digest_code == previous_digest_code
assert text.original_id == previous_original_id
def test_text_adapter_is_a_modified_string():
text = TextAdapter('lololo')
assert isinstance(text, UserString)
def test_digest_code_is_initialized_after_anonymize_call():
text = TextAdapter('some random data')
text.anonymize()
assert text.digest_code
def test_text_cannot_access_digest_code_property_without_anonymize_call():
text = TextAdapter('some random data here')
with pytest.raises(RuntimeError, match=r'Text has not been anonymized'):
text.digest_code
def test_text_adapter_can_return_strings():
text = TextAdapter('my nice and beautiful string')
assert text.data == 'my nice and beautiful string'
assert text.to_unicode() == 'my nice and beautiful string'
def test_text_adapter_can_be_initialized_with_none():
text = TextAdapter()
assert text.data == ''