UPI Transaction Parser

A lightweight Dart package for parsing UPI transaction SMS messages into structured transaction objects.

The package extracts transaction amount, merchant information, transaction type (income/expense), and spending category from raw UPI SMS messages using pattern matching and keyword detection.

Perfect for expense trackers, personal finance applications, transaction analyzers, and financial dashboards.

---

Features

Transaction amount extraction

Income / Expense detection

Merchant identification

Automatic spending categorization

UPI ID fallback detection

JSON serialization support

Pure Dart implementation

---

Getting Started

1. Add the dependency

dependencies:
  upi_transaction_parser: ^0.0.1

2. Install dependencies

flutter pub get

---

Usage

Import the package

import 'package:upi_transaction_parser/upi_transaction_parser.dart';

Parse a transaction message

final transaction = UpiParser.parse(
  'Paid Rs. 250 to Swiggy via UPI',
  DateTime.now(),
);

print(transaction);

Output

UpiTransaction(
  amount: 250.0,
  merchant: Swiggy,
  type: expense,
  category: Food & Dining
)

---

Expense Transaction Example

final transaction = UpiParser.parse(
  'Paid Rs. 499 to Amazon via UPI',
  DateTime.now(),
);

print(transaction?.merchant);
print(transaction?.category);

Output:

Amazon
Shopping

---

Income Transaction Example

final transaction = UpiParser.parse(
  'Received Rs. 5000 from Rahul',
  DateTime.now(),
);

print(transaction?.type);
print(transaction?.category);

Output:

TransactionType.income
Income

---

Supported Message Formats

The parser supports common UPI and banking transaction formats such as:

Paid Rs. 250 to Swiggy via UPI

Received Rs. 500 from Rahul

INR 1,200 debited from account

₹999 transferred to Amazon

Rs. 350 credited to your account

Sent Rs. 150 to abc@ybl

---

Transaction Model

The parser returns a structured UpiTransaction object.

class UpiTransaction {
  final double amount;
  final String merchant;
  final TransactionType type;
  final DateTime date;
  final String category;
  final String rawMessage;
}

---

Transaction Type

enum TransactionType {
  income,
  expense,
}

---

JSON Serialization

Convert transaction to JSON

final json = transaction.toJson();

Example Output:

{
  "amount": 250.0,
  "merchant": "Swiggy",
  "type": "expense",
  "date": "2025-06-06T12:00:00.000",
  "category": "Food & Dining",
  "rawMessage": "Paid Rs. 250 to Swiggy via UPI"
}

Restore transaction from JSON

final transaction =
    UpiTransaction.fromJson(json);

---

Components

UpiParser

Main parser engine responsible for:

• Amount extraction
• Merchant extraction
• Transaction type detection
• Category detection

UpiTransaction

Structured model representing a parsed transaction.

TransactionType

Enum representing transaction type.

TransactionType.income
TransactionType.expense

---

Package Structure

lib/
├── upi_transaction_parser.dart
└── src/
    ├── parser_engine.dart
    ├── transaction_model.dart
    └── transaction_type.dart

---

License

MIT License