Testing-framework

UML Sequence

Main Workflow Steps

1. User Sample Generation:

   • Generate a user sample using the configuration.

2. Merchant Configuration Matching:

   • Match the generated sample with merchant configurations to identify suitable connectors.

3. Connector Selection (Based on Algorithm):

   • Use the routing algorithm (StraightThroughRouting) to select the appropriate connector.

4. Transaction Simulation with PSP Config:

   • Simulate the transaction based on PSP configuration and the generated sample.

5. Transaction Result Feedback goes to the Algorithm

   • Transaction outcomes are used to provide feedback to the connector selection algorithm.

6. Recording Transaction Data:

   • Send transaction data to the recorder, which writes to CSV and updates metrics.

7. Metrics Calculation and Analysis:

   • Calculate metrics from the recorder data to evaluate the performance of different algorithms.

Detailed Code Snippets Explanation

1. User Sample Generation & Connector Matching

// Excerpt from main.rs: generate_user_sample function
fn generate_user_sample(config: &Config) -> Result<(String, Vec<Key>)> {
    let output = config.user.generate_sample()?;
    let connectors = find_suitable_connectors(&output, &config.merchant);
    // ...existing code...
}

This snippet generates a sample based on user configurations and finds connectors that match merchant settings.

2. Transaction Simulation & Recording

// Excerpt from main.rs: call_script function
match config.psp.call_evaluator(&connector, &user_sample)? {
    Status::Success => {
        // ...existing code...
        let record_data = PaymentRecorderData::set_values(connector.clone(), Status::Success, Key(user_sample.clone()));
        record_data.record_transaction(metrics)?;
    },
    Status::Failure => {
        // ...existing code...
        let record_data = PaymentRecorderData::set_values(connector.clone(), Status::Failure, Key(user_sample.clone()));
        record_data.record_transaction(metrics)?;
    },
}

Here the PSP evaluator simulates the transaction outcome, and the result is recorded using the recorder trait.

3. Metrics Calculation & Aggregation

// Excerpt from record.rs: updating metrics data
metrics.key.entry(connector).or_insert_with(HashMap::new)
    .entry(pm).or_insert_with(HashMap::new)
    .entry(pmt).or_insert_with(HashMap::new)
    .entry(verdict).and_modify(|e| *e += 1).or_insert(1);

This code maintains a nested map that tracks transaction counts per connector, payment method, and status, allowing later calculation of success rates.

Data Format in record.csv

The CSV file stores transaction records with the following columns:

1. Connector Name (e.g., stripe, paypal)
2. Transaction Status (e.g., Success, Failure)
3. Payment Method (e.g., card, wallet, bnpl)
4. Payment Method Type (e.g., debit, credit)

Example entry:

stripe,Success,card,debit
paypal,Success,card,credit
adyen,Success,bnpl,debit
stripe,Failure,bnpl,debit

OutPut:

Connector: "adyen", Success Rate: 38.46%
Connector: "adyen", Payment Method: "card", Success Rate: 40.00%
Connector: "adyen", Payment Method: "card", Payment Method Type: "credit", Success Rate: 40.00%
Connector: "adyen", Payment Method: "card", Payment Method Type: "debit", Success Rate: 40.00%
Connector: "adyen", Payment Method: "wallet", Success Rate: 37.68%
Connector: "adyen", Payment Method: "wallet", Payment Method Type: "N/A", Success Rate: 37.68%
Connector: "stripe", Success Rate: 100.00%
Connector: "stripe", Payment Method: "card", Success Rate: 100.00%
Connector: "stripe", Payment Method: "card", Payment Method Type: "debit", Success Rate: 100.00%
Connector: "stripe", Payment Method: "card", Payment Method Type: "credit", Success Rate: 100.00%
Connector: "paypal", Success Rate: 35.14%
Connector: "paypal", Payment Method: "bnpl", Success Rate: 31.65%
Connector: "paypal", Payment Method: "bnpl", Payment Method Type: "N/A", Success Rate: 31.65%
Connector: "paypal", Payment Method: "card", Success Rate: 43.75%
Connector: "paypal", Payment Method: "card", Payment Method Type: "debit", Success Rate: 52.38%
Connector: "paypal", Payment Method: "card", Payment Method Type: "credit", Success Rate: 27.27%


Total transactions: 1500
Total transactions from "adyen" connector: 6.93%
Total transactions from "stripe" connector: 85.67%
Total transactions from "paypal" connector: 7.40%
Total Success Rate: 90.93%

How to Run

1. Manually edit the iteration count in the main loop in main.rs (e.g., change the for _ in 0..1500 loop to your desired number of runs).
2. Save your changes.
3. Run the project using:

   cargo run
   

TODO

• Integrate hyperswitch-dynamo

• Implement feedback loop to algorithm.

• Output format

  Label	Dynamo suggested connector	payment_status	conclusive_classification
  Stripe	Stripe	Success	True Positive
  Stripe	Adyen	Success	False Negative
  Stripe	Stripe	Failure	False Positive
  Stripe	Adyen	Failure	True Negative