def get_datetime(self, line: str):
line_clean = clean_string(line, charset=self.ALLOWED_CHARS_ALPHANUM)
date_items = re.findall(r"\d{1,8}", line_clean)
date_value = date_items[0] if len(date_items) >= 2 else ""
if date_value.isdigit() and len(date_value) >= 8:
return datetime.strptime(date_items[0], '%d%m%Y')
return None
def _is_words_and_pay_line(self, first_desire_word: str,
line: str) -> bool:
cleaned_line = clean_string(
self._remove_spaces_and_to_lower(line),
charset="abcdefghijklmnopqrstuvwxyz0123456789.")
# Does have the right body?
if ((not (re.search(r"[a-z]+\d+\.\d{2}", cleaned_line)
or re.search(r"[a-z]+\d{3,}", cleaned_line)))
or re.search(r"^[a-z]+$", cleaned_line)
or re.search(r"^\d+$", cleaned_line)
or re.search(r"^\d+\.\d+$", cleaned_line)):
return False
# Try to guess if first chars are: first_desire_word
firs_input_word = ""
# Does it has dot?
if "." not in cleaned_line:
firs_input_word = re.split(r"\d{3,}", cleaned_line)[0]
else:
firs_input_word = re.split(r"\d+\.\d{2}", cleaned_line)[0]
if not firs_input_word:
return False
return self._is_it_similar(
firs_input_word,
first_desire_word) or self._is_it_phonetic_similar(
firs_input_word, first_desire_word)
def clean_line(self, line: str, charset: str = None) -> str:
"""
Remove not desire chars from a string
"""
return clean_string(
line.upper(),
charset=charset or "ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789",
)
def get_amount_format(self, line):
value = clean_string(line, charset=self.ALLOWED_CHARS_ALPHANUM)
amount = self.get_numbers(value)
if not amount:
return None
amount_str = str(amount[0])
insert_point = len(amount_str) - 2
return amount_str[:insert_point] + "." + amount_str[insert_point:]
def get_next_line(self, list_data: list):
if len(list_data) >= 1:
line_data = list_data.pop(0)
text = line_data.get("text")
#print(f"get_next_line: {text}")
return Line(
uuid=line_data.get("uuid"),
text=text,
text_clean=clean_string(
text, charset=self.ALLOWED_CHARS_ALPHANUM).upper(),
text_len=len(text),
)
return Line(uuid="-1")
def get_date(self, line: str) -> str:
"""
Get the date value of FECHA 26/11/2020 09:35:53
"""
cleaned_line = clean_string(line.upper(), charset="0123456789 :-/+.")
if line == "":
return None
date_values = {"year": 2020, "month": 1, "day": 1}
date = re.split(r'-|/|\s|\+', cleaned_line)
cleaned_line = self._translate_line(cleaned_line)
time_items = re.split(r":|\.", cleaned_line)
try:
if len(date) >= 4:
year = re.findall(r"\d{4}|\d{2}", date[-2])[0]
date_values.update({
"year": int(year),
"month": int(date[-2]),
"day": int(date[-4]),
"hour": int(time_items[-3]),
"minute": int(time_items[-2]),
"second": int(time_items[-1])
})
else:
year = re.findall(r"\d{4}|\d{2}", date[-1])[0]
date_values.update({
"year": int(year),
"month": int(date[-2]),
"day": int(date[-3]),
"hour": int(time_items[-3]),
"minute": int(time_items[-2]),
"second": int(time_items[-1])
})
return datetime(**date_values)
except ValueError:
return None
except IndexError:
return None
return None
def get_quantity_without_text(self, text: str, number_zero: int):
quantitys = clean_string(text, "0123456789. ")
return self.format_quantity(quantitys, number_zero)
def clean_headers(headers):
headers = [clean_string(h) for h in headers]
headers = [title_to_snake(h) for h in headers]
return headers