def test_letters_count_none():
drop_caches()
with settings.file_override("TOKENS"):
with open(settings.TOKENS, "wb") as f:
pickle.dump({"these", "are", "tokens"}, f)
assert SubstitutionFeaturesMixin._letters_count() == {"these", "are", "tokens"}
def test_letters_count_none_with_computed():
drop_caches()
# Lemmas are all lowercase.
for word in SubstitutionFeaturesMixin._letters_count():
assert word.islower() or is_int(word[0]) or is_int(word[-1])
def test_letters_count():
drop_caches()
assert SubstitutionFeaturesMixin._letters_count("hello") == 5
assert SubstitutionFeaturesMixin._letters_count("mountain") == 8
assert SubstitutionFeaturesMixin._letters_count("makakiki") == 8
