def test_clean_string(self):
input_string = " legitimate-.,;:_·<>+\\|/'#@()\"\t\n\r!%&=?¡¿ text "
expected_string = "legitimate text"
p = Processing()
output_string = Tools.clean_string(input_string,
StringProcessor.unwanted_chars)
output_string = Tools.clean_spaces(output_string)
self.assertEqual(expected_string, output_string)
def process(self, input_string):
# Transliterate to ascii
final_str = self.trans.to_ascii(input_string)
# Clean useless chars and trim spaces:
final_str = Tools.clean_string(final_str,
StringProcessor.unwanted_chars)
final_str = Tools.clean_spaces(final_str)
# Uppercase the string
final_str = final_str.upper()
return final_str
def process(self, input_string):
# Trim spaces and remove unwanted chars:
final_str = Tools.clean_string(input_string,
NumberProcessor.unwanted_chars)
final_str = Tools.clean_spaces(final_str)
try:
# Parse the number to the most general one (float)
number = float(final_str)
# Turn the number to a string again (to homogenise representation)
final_str = str(number)
except ValueError:
logging.warning("Number string could not be parsed: " +
input_string + " (original)" + final_str +
" (after cleaning)")
# If number cannot be parsed, return original string
final_str = input_string
return final_str
def process(self, input_string):
# Trim spaces and remove unwanted chars:
final_str = Tools.clean_string(input_string,
DateProcessor.unwanted_chars)
final_str = Tools.clean_spaces(final_str)
try:
# Parse the date with multiple approaches
date_object = dateparser.parse(final_str)
if date_object is None:
raise TypeError
# Turn the date to a string again (use ISO format)
final_str = str(date_object.date().isoformat())
except TypeError:
logging.warning("Date string could not be parsed: <" +
input_string + "> (original) - <" + final_str +
"> (after cleaning)")
# If date cannot be parsed, return original string
final_str = input_string
return final_str