datalogic-rs

A high-performance, production-ready Rust implementation of JSONLogic.

Evaluate complex rules and dynamic expressions with a powerful, memory-efficient, and developer-friendly engine.

🏢 Organization • 📖 Docs • 🐛 Report a Bug

---

datalogic-rs is a Rust implementation of JSONLogic designed for performance, safety, and ease of use. Whether you're building a feature flagging system, a dynamic pricing engine, or a complex validation pipeline, datalogic-rs provides the power and flexibility you need.

Our secret sauce? An arena-based memory model that delivers zero-copy parsing and evaluation, making it up to 30% faster than other implementations.

🚀 What Makes datalogic-rs Awesome?

• Blazing Fast: Our arena-based allocator minimizes memory overhead and maximizes speed.
• 100% JSONLogic Compliant: Full compatibility with the official JSONLogic test suite.
• Extensible: Easily add your own custom operators with our simple or advanced APIs.
• Powerful Templating: Preserve object structures for dynamic, structured output.
• Ready for Production: Thread-safe, statically dispatched, and optimized for real-world workloads.
• Batteries Included: Over 50 built-in operators, including datetime and regex support.
• WASM Support: Compile to WebAssembly for use in browser environments.

🏗️ How It Works: The Arena Advantage

datalogic-rs uses an arena allocator for memory-efficient evaluation. This means we can parse a rule once and evaluate it many times with minimal overhead.

use datalogic_rs::DataLogic;

// Create an evaluator with a custom arena chunk size.
let dl = DataLogic::with_chunk_size(8192);

// Parse your rule and data once.
let rule = dl.parse_logic(r#"{ ">": [{ "var": "temp" }, 100] }"#).unwrap();
let data = dl.parse_data(r#"{ "temp": 110 }"#).unwrap();

// Evaluate efficiently.
let result = dl.evaluate(&rule, &data).unwrap();
assert!(result.to_json().as_bool().unwrap());

// Reset the arena to reuse memory for the next batch of evaluations.
dl.reset_arena();

Structured Object Preservation

Generate powerful, structured outputs by preserving non-operator keys in your JSON.

use datalogic_rs::DataLogic;

// Enable structure preservation for templating.
let dl = DataLogic::with_preserve_structure();

let result = dl.evaluate_str(
    r#"{
        "status": { "if": [{ ">=": [{ "var": "score" }, 90] }, "pass", "fail"] },
        "grade": { "+": [{ "var": "score" }, { "var": "bonus" }] },
        "timestamp": { "now": [] }
    }"#,
    r#"{ "score": 85, "bonus": 10 }"#,
).unwrap();

// The result is a structured object with evaluated fields.
// { "status": "pass", "grade": 95, "timestamp": "2024-01-15T10:30:00Z" }

🎯 Getting Started: Core API

1. evaluate - For Reusable Rules & Data

Ideal for scenarios where you need to run the same rule against different data.

use datalogic_rs::DataLogic;

let dl = DataLogic::new();

// Parse the rule and data once.
let rule = dl.parse_logic(r#"{ ">": [{ "var": "temp" }, 100] }"#).unwrap();
let data = dl.parse_data(r#"{ "temp": 110 }"#).unwrap();

// Evaluate as many times as you need.
let result = dl.evaluate(&rule, &data).unwrap();
assert!(result.to_json().as_bool().unwrap());

2. evaluate_str - For Quick, One-Shot Evaluations

Perfect for when your rules are dynamic or you only need a single evaluation.

use datalogic_rs::DataLogic;

let dl = DataLogic::new();

let result = dl.evaluate_str(
    r#"{ "abs": -42 }"#,
    r#"{}"#, // No data needed for this rule.
).unwrap();

assert_eq!(result.as_i64().unwrap(), 42);

3. evaluate_json - For Seamless JSON Integration

Works directly with serde_json::Value for easy integration into your existing JSON workflows.

use datalogic_rs::DataLogic;
use serde_json::json;

let dl = DataLogic::new();

let logic = json!({
    "if": [
        { ">": [{ "var": "cart.total" }, 100] },
        "Premium",
        "Standard"
    ]
});

let data = json!({ "cart": { "total": 120 } });

let result = dl.evaluate_json(&logic, &data).unwrap();
assert_eq!(result.as_str().unwrap(), "Premium");

🔧 Installation

Add datalogic-rs to your Cargo.toml:

[dependencies]
datalogic-rs = "3.0"

📖 Real-World Examples

Complex Business Rules

Check for multi-condition eligibility with nested logic.

use datalogic_rs::DataLogic;

let dl = DataLogic::new();

let result = dl.evaluate_str(
    r#"{
        "and": [
            { ">=": [{ "var": "age" }, 18] },
            { "<": [{ "var": "age" }, 65] },
            { "or": [
                { "==": [{ "var": "subscription" }, "premium"] },
                { ">=": [{ "var": "purchases" }, 5] }
            ] }
        ]
    }"#,
    r#"{ "age": 25, "subscription": "basic", "purchases": 7 }"#,
).unwrap();

assert!(result.as_bool().unwrap());

Array Processing & Filtering

Filter and map user data with ease.

use datalogic_rs::DataLogic;

let dl = DataLogic::new();

let result = dl.evaluate_str(
    r#"{
        "map": [
            {
                "filter": [
                    { "var": "users" },
                    { ">=": [{ "var": "age" }, 18] }
                ]
            },
            { "var": "name" }
        ]
    }"#,
    r#"{
        "users": [
            { "name": "Alice", "age": 20 },
            { "name": "Bob", "age": 15 },
            { "name": "Charlie", "age": 25 }
        ]
    }"#,
).unwrap();

// Returns ["Alice", "Charlie"]
assert_eq!(result.as_array().unwrap().len(), 2);

DateTime Operations

Perform date arithmetic with timezone support.

use datalogic_rs::DataLogic;

let dl = DataLogic::new();

let result = dl.evaluate_str(
    r#"{
        ">": [
            { "+": [
                { "datetime": "2023-07-15T08:30:00Z" },
                { "duration": "2d" }
            ] },
            { "datetime": "2023-07-16T08:30:00Z" }
        ]
    }"#,
    r#"{}"#,
).unwrap();

assert!(result.as_bool().unwrap());

Regex Data Extraction

Extract structured data from strings using named capture groups.

use datalogic_rs::DataLogic;

let dl = DataLogic::new();

let result = dl.evaluate_str(
    r#"{
        "split": [
            "SBININBB101",
            "^(?P<bank>[A-Z]{4})(?P<country>[A-Z]{2})(?P<location>[A-Z0-9]{2})(?P<branch>[A-Z0-9]{3})?$"
        ]
    }"#,
    r#"{}"#,
).unwrap();

// Returns: { "bank": "SBIN", "country": "IN", "location": "BB", "branch": "101" }
let obj = result.as_object().unwrap();
assert_eq!(obj.get("bank").unwrap().as_str().unwrap(), "SBIN");

🔌 Custom Operators

Extend datalogic-rs with your own domain-specific logic.

use datalogic_rs::{DataLogic, DataValue};
use datalogic_rs::value::NumberValue;

// Define a custom operator that doubles a number.
fn double(args: Vec<DataValue>, _data: DataValue) -> Result<DataValue, String> {
    if let Some(n) = args.first().and_then(|v| v.as_f64()) {
        return Ok(DataValue::Number(NumberValue::from_f64(n * 2.0)));
    }
    Err("Argument must be a number".to_string())
}

let mut dl = DataLogic::new();
dl.register_simple_operator("double", double);

// Use your custom operator in a rule.
let result = dl.evaluate_str(r#"{ "double": 21 }"#, r#"{}"#).unwrap();

assert_eq!(result.as_f64().unwrap(), 42.0);

📊 Performance

datalogic-rs is fast. Here's how it stacks up against other implementations on an Apple M2 Pro:

Implementation	Execution Time	Relative Performance
datalogic-rs	380ms	1.0x (baseline)
json-logic-engine (pre-compiled)	417ms	1.1x slower
json-logic-engine (interpreted)	986ms	2.6x slower
json-logic-js	5,755ms	15.1x slower

Benchmarks run on the standard JSONLogic test suite.

🎯 Use Cases

Feature Flagging

{
    "and": [
        { "==": [{ "var": "user.country" }, "US"] },
        { "or": [
            { "==": [{ "var": "user.role" }, "beta_tester"] },
            { ">=": [{ "var": "user.account_age_days" }, 30] }
        ] }
    ]
}

Dynamic Pricing

{
    "if": [
        { ">=": [{ "var": "cart.total" }, 100] },
        { "-": [{ "var": "cart.total" }, { "*": [{ "var": "cart.total" }, 0.1] }] },
        { "var": "cart.total" }
    ]
}

Fraud Detection

{
    "or": [
        { "and": [
            { "!=": [{ "var": "transaction.billing_country" }, { "var": "user.country" }] },
            { ">=": [{ "var": "transaction.amount" }, 1000] }
        ] },
        { ">=": [{ "var": "transaction.attempts_last_hour" }, 5] }
    ]
}

📋 Supported Operations

Category	Operators
Comparison	==, ===, !=, !==, >, >=, <, <=
Logic	and, or, !, !!
Arithmetic	+, -, *, /, %, min, max, abs, ceil, floor
Control Flow	if, ?:, ??
Arrays	map, filter, reduce, all, some, none, merge, in, length, slice, sort
Strings	cat, substr, starts_with, ends_with, upper, lower, trim, replace, split
Data Access	var, val, exists, missing, missing_some
DateTime	datetime, timestamp, now, parse_date, format_date, date_diff
Error Handling	throw, try
Custom	User-defined operators

🤝 Contributing

We welcome contributions! Please see CONTRIBUTING.md for details on how to get started.

🏢 About Plasmatic

datalogic-rs is developed by Plasmatic, an organization dedicated to building open-source tools for financial infrastructure and data processing.

Check out our other projects:

• SwiftMTMessage: A library for parsing SWIFT MT messages with CBPR+ compliance.
• Reframe: A transformation engine for converting SWIFT MT messages to ISO 20022.
• MXMessage: A library for parsing ISO 20022 (MX) messages.

📄 License

datalogic-rs is licensed under the Apache License, Version 2.0. See LICENSE for details.

---

Built with ❤️ by the Plasmatic team