def _transform(self, document):
lines_from_section = section_extract(self.section_regex,
document['description'])
return [
word_tokenize(clean_str(strip_bullets_from_line(line.text)))
for line in lines_from_section
]
def _clean(self, document):
return self.join_spaces([
clean_str(document[field])
for field in self.document_schema_fields
])
def test_deep_wrapper(self):
assert clean_str([["macy's engineer / apply now", "engineer/apply now"], ["engineer.", "python!"]]) == \
[["macy s engineer apply now", "engineer apply now"], ["engineer ", "python "]]