MathParser for Arduino & ESP32

Evaluate math expressions from strings at runtime no dependencies, no heap beyond the parser itself.

---

What is MathParser?

MathParser lets your Arduino evaluate a math expression stored as a string at runtime. Instead of hardcoding y = x * 1.8 + 32, you store "x * 1.8 + 32" in EEPROM and evaluate it dynamically. Change the formula without reflashing.

Real-world use cases:

• 🌡️ Sensor calibration curves stored in EEPROM/SPIFFS
• ⚙️ Config-driven thresholds and formulas
• 🎮 User-entered expressions via Serial or touchscreen UI
• 🔧 Dynamic PID tuning formulas
• 🔄 Unit conversions defined at runtime
• 🤖 Joystick / motor input shaping with custom functions

---

Supported syntax

Category	Examples
Operators	+ - * / ^ %
Grouping	( )
Functions	sin cos tan asin acos atan atan2 sinh cosh tanh
	sqrt cbrt pow exp log log2 log10
	abs ceil floor round fmod
Constants	pi e
Variables	any name you bind via setVariable()
Custom functions	0, 1, 2, or 3 argument function pointers

---

Installation

Via Arduino IDE

1. Download MathParser.zip from Releases
2. Sketch → Include Library → Add .ZIP Library...

Via Arduino Library Manager

Search MathParser in Tools → Manage Libraries.

Manual

cd ~/Arduino/libraries
git clone https://github.com/meerzafarnoohani/MathParser.git

---

Quick Start

#include <MathParser_Arduino.h>

One-shot evaluation (no variables)

double r = MathParser::evaluate("2^10 + sqrt(144)");  // 1036.0

One-shot with a single variable

double r = MathParser::evaluate("x * sin(x)", "x", 1.5708);  // ≈ 1.5708

Compiled expression with bound variables

MathParser te;
double x = 0.0, y = 0.0;

te.setVariable("x", &x);
te.setVariable("y", &y);
te.compile("sqrt(x^2 + y^2)");   // compile once

x = 3.0; y = 4.0;
double dist = te.eval();          // 5.0   variables read at eval time

x = 5.0; y = 12.0;
dist = te.eval();                 // 13.0

Custom functions

double myClamp(double val, double limit) {
    return val > limit ? limit : val;
}

MathParser te;
te.addFunction2("clamp", myClamp);

double v = 0.0;
te.setVariable("v", &v);
te.compile("clamp(v * 2.5, 100)");

v = 30.0;
double out = te.eval();   // 75.0

v = 50.0;
out = te.eval();          // 100.0 (clamped)

---

API Reference

Static helpers (no object needed)

double MathParser::evaluate(const char* expr, int* error = nullptr);
double MathParser::evaluate(const char* expr, const char* varName, double value, int* error = nullptr);

Instance methods

MathParser te;

// Bind variables (up to 8)
te.setVariable("x", &x);           // double* pointer

// Bind custom functions
te.addFunction0("myConst", fn);    // double fn()
te.addFunction1("myFn",    fn);    // double fn(double)
te.addFunction2("myFn2",   fn);    // double fn(double, double)

// Compile and evaluate
te.compile("x^2 + 1");             // returns bool
te.eval();                          // returns double

// Error reporting
te.isValid();                       // bool last compile succeeded?
te.errorPosition();                 // int  0 = no error
te.printError("x^2 + 1");          // print arrow pointing at error

// Reset
te.reset();                         // clear expression + all bindings

---

Examples

Example	Description
BasicEval	One-shot evaluation of 12 test expressions with pass/fail
WithVariables	Sensor calibration formula with live ADC simulation
CustomFunctions	Register clamp, deadband, unit conversion, motor shaping

---

Memory

• Stack usage per MathParser instance: ~200 bytes (8 variable slots)
• Parser uses heap transiently during compile(), then frees it
• eval() uses no heap walks the compiled expression tree

---

Supported Platforms

Platform	Status
ESP32	✅ Primary target
ESP8266	✅ Supported
Arduino Uno / Nano	✅ Use short expressions
Arduino Mega	✅ Supported
SAMD / RP2040	✅ Supported

---

Credits

Original MathParser by Lewis Van Winkle github.com/codeplea/MathParser (zlib licence). Arduino/ESP32 port and C++ wrapper by Meer Zafarullah Noohani github.com/meerzafarnoohani.

---

License

MIT see LICENSE.