87 lines
3.2 KiB
Python
87 lines
3.2 KiB
Python
import re
|
||
from datetime import datetime
|
||
from pathlib import Path
|
||
from typing import List
|
||
|
||
import pdfplumber
|
||
|
||
from backend.parsers.base import BaseBankParser, ParsedTransaction
|
||
|
||
|
||
def _normalize_amount(s: str) -> float:
|
||
return float(s.replace("\u00a0", " ").replace(" ", "").replace(",", "."))
|
||
|
||
|
||
def _parse_date(d: str) -> str:
|
||
"""DD.MM.YYYY -> YYYY-MM-DD"""
|
||
try:
|
||
dt = datetime.strptime(d.strip(), "%d.%m.%Y")
|
||
return dt.strftime("%Y-%m-%d")
|
||
except ValueError:
|
||
return d.strip()
|
||
|
||
|
||
def _parse_datetime(d: str, time_str: str) -> str:
|
||
"""Date DD.MM.YYYY + time HH:MM -> ISO"""
|
||
try:
|
||
part = d.strip() + " " + (time_str or "00:00").strip()
|
||
dt = datetime.strptime(part, "%d.%m.%Y %H:%M")
|
||
return dt.strftime("%Y-%m-%dT%H:%M:%S")
|
||
except ValueError:
|
||
return _parse_date(d) + "T00:00:00"
|
||
|
||
|
||
class BankTParser(BaseBankParser):
|
||
"""Парсер выписок Т-банка. Файлы Т-MM-YY.pdf."""
|
||
|
||
# Первая строка операции: дата дата сумма ₽ сумма ₽ описание 4цифры (сумма может быть + или -)
|
||
ROW_RE = re.compile(
|
||
r"^(\d{2}\.\d{2}\.\d{4})\s+(\d{2}\.\d{2}\.\d{4})\s+([-+]?[\d\s,]+\.\d{2})\s*₽\s+([-+]?[\d\s,]+\.\d{2})\s*₽\s+(.+?)\s+(\d{4})\s*$",
|
||
re.UNICODE,
|
||
)
|
||
# Вторая строка (время): HH:MM HH:MM остаток текста
|
||
TIME_RE = re.compile(r"^(\d{1,2}:\d{2})\s+(\d{1,2}:\d{2})\s*(.*)$")
|
||
|
||
def can_parse(self, filename: str) -> bool:
|
||
name = Path(filename).name
|
||
return name.startswith("Т-") and name.lower().endswith(".pdf")
|
||
|
||
def parse(self, file_path: str) -> List[ParsedTransaction]:
|
||
result: List[ParsedTransaction] = []
|
||
with pdfplumber.open(file_path) as pdf:
|
||
lines: List[str] = []
|
||
for page in pdf.pages:
|
||
text = page.extract_text()
|
||
if text:
|
||
lines.extend(text.split("\n"))
|
||
|
||
i = 0
|
||
while i < len(lines):
|
||
line = lines[i]
|
||
m = self.ROW_RE.match(line.strip())
|
||
if m:
|
||
date_op, date_debit, amt_op, amt_card, desc, card_tail = m.groups()
|
||
op_time, debit_time = "00:00", "00:00"
|
||
if i + 1 < len(lines):
|
||
tm = self.TIME_RE.match(lines[i + 1].strip())
|
||
if tm:
|
||
op_time, debit_time, rest = tm.groups()
|
||
if rest:
|
||
desc = (desc + " " + rest).strip()
|
||
i += 1
|
||
|
||
amount = _normalize_amount(amt_op)
|
||
amount_card = _normalize_amount(amt_card) if amt_card else None
|
||
result.append(
|
||
ParsedTransaction(
|
||
operation_date=_parse_datetime(date_op, op_time),
|
||
debit_date=_parse_datetime(date_debit, debit_time),
|
||
amount=amount,
|
||
amount_card_currency=amount_card,
|
||
description=(desc or "").strip(),
|
||
card_tail=card_tail,
|
||
)
|
||
)
|
||
i += 1
|
||
return result
|