#!/usr/bin/env python3 """ SWIFT FIN Parser Parses SWIFT FIN messages (MT103, MT202, MT940, etc.) """ import re from typing import Dict, Any, List, Optional from datetime import datetime import logging logger = logging.getLogger(__name__) class SWIFTFINParser: """Parser for SWIFT FIN messages""" SUPPORTED_MESSAGE_TYPES = [ 'MT103', # Single Customer Credit Transfer 'MT202', # General Financial Institution Transfer 'MT940', # Customer Statement Message 'MT942', # Interim Transaction Report 'MT950', # Statement Message ] def parse(self, swift_message: str) -> Dict[str, Any]: """Parse SWIFT FIN message""" lines = swift_message.strip().split('\n') message = { 'type': 'SWIFT_FIN', 'basicHeader': self._parse_block(lines, 1), 'applicationHeader': self._parse_block(lines, 2), 'userHeader': self._parse_block(lines, 3), 'text': self._parse_text_block(lines), 'trailer': self._parse_block(lines, 5), } # Parse message type app_header = message['applicationHeader'].get('raw', '') message_type = self._extract_message_type(app_header) message['messageType'] = message_type # Parse based on message type if message_type == 'MT103': message['parsed'] = self._parse_mt103(message['text']) elif message_type == 'MT202': message['parsed'] = self._parse_mt202(message['text']) elif message_type == 'MT940': message['parsed'] = self._parse_mt940(message['text']) elif message_type == 'MT942': message['parsed'] = self._parse_mt942(message['text']) elif message_type == 'MT950': message['parsed'] = self._parse_mt950(message['text']) return message def _parse_block(self, lines: List[str], block_num: int) -> Dict[str, Any]: """Parse SWIFT block""" block_start = f"{block_num}:" block_end = f"-{block_num}" block_lines = [] in_block = False for line in lines: if line.startswith(block_start): in_block = True block_lines.append(line[len(block_start):]) continue if line.startswith(block_end): break if in_block: block_lines.append(line) return {'raw': '\n'.join(block_lines)} def _parse_text_block(self, lines: List[str]) -> str: """Parse SWIFT text block (block 4)""" text_start = "4:" text_end = "-" text_lines = [] in_text = False for line in lines: if line.startswith(text_start): in_text = True text_lines.append(line[2:]) continue if line.startswith(text_end) and not line.startswith("-4"): break if in_text: text_lines.append(line) return '\n'.join(text_lines) def _extract_message_type(self, app_header: str) -> str: """Extract message type from application header""" # MT103, MT202, etc. are in the application header match = re.search(r'MT\d{3}', app_header) return match.group(0) if match else 'UNKNOWN' def _parse_mt103(self, text: str) -> Dict[str, Any]: """Parse MT103 (Single Customer Credit Transfer)""" fields = self._parse_swift_fields(text) return { 'senderReference': fields.get('20', ''), 'bankOperationCode': fields.get('23B', ''), 'valueDate': fields.get('32A', '').split()[0] if fields.get('32A') else '', 'currency': fields.get('32A', '').split()[1] if fields.get('32A') and len(fields.get('32A', '').split()) > 1 else '', 'amount': fields.get('32A', '').split()[2] if fields.get('32A') and len(fields.get('32A', '').split()) > 2 else '', 'orderingCustomer': fields.get('50A', '') or fields.get('50K', ''), 'sendingInstitution': fields.get('52A', '') or fields.get('52D', ''), 'orderingInstitution': fields.get('56A', '') or fields.get('56C', '') or fields.get('56D', ''), 'beneficiaryCustomer': fields.get('59', '') or fields.get('59A', ''), 'remittanceInfo': fields.get('70', ''), 'senderToReceiverInfo': fields.get('72', ''), } def _parse_mt202(self, text: str) -> Dict[str, Any]: """Parse MT202 (General Financial Institution Transfer)""" fields = self._parse_swift_fields(text) return { 'senderReference': fields.get('20', ''), 'transactionReference': fields.get('21', ''), 'valueDate': fields.get('32A', '').split()[0] if fields.get('32A') else '', 'currency': fields.get('32A', '').split()[1] if fields.get('32A') and len(fields.get('32A', '').split()) > 1 else '', 'amount': fields.get('32A', '').split()[2] if fields.get('32A') and len(fields.get('32A', '').split()) > 2 else '', 'sendingInstitution': fields.get('52A', '') or fields.get('52D', ''), 'orderingInstitution': fields.get('56A', '') or fields.get('56C', '') or fields.get('56D', ''), 'intermediary': fields.get('56A', '') or fields.get('56C', '') or fields.get('56D', ''), 'accountWithInstitution': fields.get('57A', '') or fields.get('57C', '') or fields.get('57D', ''), 'beneficiaryInstitution': fields.get('58A', '') or fields.get('58D', ''), 'remittanceInfo': fields.get('70', ''), 'senderToReceiverInfo': fields.get('72', ''), } def _parse_mt940(self, text: str) -> Dict[str, Any]: """Parse MT940 (Customer Statement Message)""" fields = self._parse_swift_fields(text) return { 'statementNumber': fields.get('28C', ''), 'accountIdentification': fields.get('25', ''), 'statementNumberSequence': fields.get('28C', '').split('/')[0] if fields.get('28C') else '', 'statementNumberPage': fields.get('28C', '').split('/')[1] if fields.get('28C') and '/' in fields.get('28C', '') else '', 'openingBalance': self._parse_balance(fields.get('60F', '') or fields.get('60M', '')), 'closingBalance': self._parse_balance(fields.get('62F', '') or fields.get('62M', '')), 'transactions': self._parse_transactions(text), } def _parse_mt942(self, text: str) -> Dict[str, Any]: """Parse MT942 (Interim Transaction Report)""" fields = self._parse_swift_fields(text) return { 'statementNumber': fields.get('28C', ''), 'accountIdentification': fields.get('25', ''), 'openingBalance': self._parse_balance(fields.get('60F', '') or fields.get('60M', '')), 'transactions': self._parse_transactions(text), } def _parse_mt950(self, text: str) -> Dict[str, Any]: """Parse MT950 (Statement Message)""" fields = self._parse_swift_fields(text) return { 'statementNumber': fields.get('28C', ''), 'accountIdentification': fields.get('25', ''), 'openingBalance': self._parse_balance(fields.get('60F', '')), 'closingBalance': self._parse_balance(fields.get('62F', '')), 'transactions': self._parse_transactions(text), } def _parse_swift_fields(self, text: str) -> Dict[str, str]: """Parse SWIFT fields from text block""" fields = {} current_field = None current_value = [] for line in text.split('\n'): # Field starts with colon (e.g., ":20:") if re.match(r'^:\d{2}[A-Z]?:', line): if current_field: fields[current_field] = '\n'.join(current_value).strip() match = re.match(r'^:(\d{2}[A-Z]?):', line) current_field = match.group(1) if match else None current_value = [line.split(':', 2)[2] if ':' in line[3:] else ''] elif current_field: current_value.append(line) if current_field: fields[current_field] = '\n'.join(current_value).strip() return fields def _parse_balance(self, balance_str: str) -> Dict[str, Any]: """Parse balance string (e.g., "D230101USD1000,00")""" if not balance_str: return {} match = re.match(r'^([DC])(\d{6})([A-Z]{3})([\d,]+)$', balance_str) if match: return { 'creditDebitIndicator': 'Credit' if match.group(1) == 'C' else 'Debit', 'date': match.group(2), 'currency': match.group(3), 'amount': match.group(4).replace(',', '.'), } return {'raw': balance_str} def _parse_transactions(self, text: str) -> List[Dict[str, Any]]: """Parse transaction entries from text""" transactions = [] fields = self._parse_swift_fields(text) # MT940/942/950 transactions are in field 61 and 86 transaction_text = fields.get('61', '') if transaction_text: for line in transaction_text.split('\n'): if line.strip(): transactions.append(self._parse_transaction_line(line)) return transactions def _parse_transaction_line(self, line: str) -> Dict[str, Any]: """Parse a single transaction line (field 61)""" # Format: ValueDate EntryDate D/C Mark Currency Amount TransactionType Reference match = re.match(r'^(\d{6})(\d{4})?([DC])([A-Z])([N]?)(\d{1,2})([A-Z]{3})([\d,]+)(.*)$', line) if match: return { 'valueDate': match.group(1), 'entryDate': match.group(2), 'creditDebitIndicator': 'Credit' if match.group(3) == 'C' else 'Debit', 'fundsCode': match.group(4), 'supplementaryDetails': match.group(5), 'transactionType': match.group(6), 'currency': match.group(7), 'amount': match.group(8).replace(',', '.'), 'reference': match.group(9).strip(), } return {'raw': line}