Merge pull request #1760 from knoxiboy/1673-feat-expression-parser

feat: Build a Mathematical Expression Parser and Evaluator from Scratch (#1673)

Author: steam-bell-92

expression_parser/README.md

@@ -0,0 +1,38 @@
+# Mathematical Expression Parser and Evaluator
+
+A zero-external-dependency mathematical expression parser, tokenizer, AST builder, and evaluator built from scratch in Python.
+
+## Features
+
+- **Tokenizer**: Converts input text strings into tokens (Numbers, Operators, Identifiers, Functions, Parentheses). Supports implicit multiplication (e.g. `2(3+4)` or `3pi`).
+- **AST Parser**: Recursive-descent parser building Abstract Syntax Trees respecting operator precedence and associativity (`^`, `*`, `/`, `%`, `+`, `-`).
+- **Evaluator**: Walks AST nodes safely to calculate results with variable binding support and builtin math functions (`sin`, `cos`, `tan`, `sqrt`, `abs`, `log`, `exp`, `min`, `max`, `floor`, `ceil`).
+- **Interactive REPL & CLI**: Support for interactive session and CLI expression evaluation.
+
+## Usage
+
+### CLI REPL
+
+```bash
+python expression_parser/main.py
+```
+
+### Direct Expression Evaluation
+
+```bash
+python expression_parser/main.py "2 + 3 * (4 - 1)^2"
+# Output: 29.0
+```
+
+### Python API
+
+```python
+from expression_parser import evaluate_expression
+
+result = evaluate_expression("sin(pi / 2) + cos(0)")
+print(result) # 2.0
+
+vars_dict = {"x": 5, "y": 2}
+res = evaluate_expression("x^2 + 2*x*y + y^2", vars_dict)
+print(res) # 49.0
+```

expression_parser/__init__.py

@@ -0,0 +1,13 @@
+"""
+__init__.py for expression_parser module.
+"""
+
+from expression_parser.tokenizer import Tokenizer, Token, TokenType, TokenizerError
+from expression_parser.parser import Parser, ParseError, ASTNode
+from expression_parser.evaluator import Evaluator, EvaluationError, evaluate_expression
+
+__all__ = [
+    "Tokenizer", "Token", "TokenType", "TokenizerError",
+    "Parser", "ParseError", "ASTNode",
+    "Evaluator", "EvaluationError", "evaluate_expression"
+]

expression_parser/evaluator.py

@@ -0,0 +1,119 @@
+"""
+Evaluator for AST nodes generated by Parser.
+Evaluates math expressions with variable and function support.
+"""
+
+import math
+from typing import Dict, Any, Union
+from expression_parser.parser import ASTNode, NumberNode, VariableNode, UnaryOpNode, BinaryOpNode, FunctionCallNode
+
+
+class EvaluationError(Exception):
+    pass
+
+
+class Evaluator:
+    BUILTIN_FUNCTIONS = {
+        "sin": math.sin,
+        "cos": math.cos,
+        "tan": math.tan,
+        "asin": math.asin,
+        "acos": math.acos,
+        "atan": math.atan,
+        "sqrt": math.sqrt,
+        "abs": abs,
+        "log": math.log,
+        "exp": math.exp,
+        "floor": math.floor,
+        "ceil": math.ceil,
+        "min": min,
+        "max": max,
+    }
+
+    BUILTIN_CONSTANTS = {
+        "pi": math.pi,
+        "e": math.e,
+        "tau": math.tau,
+    }
+
+    def __init__(self, variables: Dict[str, Union[int, float]] = None):
+        self.variables = self.BUILTIN_CONSTANTS.copy()
+        if variables:
+            self.variables.update(variables)
+
+    def evaluate(self, node: ASTNode) -> Union[int, float]:
+        if isinstance(node, NumberNode):
+            return node.value
+
+        if isinstance(node, VariableNode):
+            if node.name in self.variables:
+                return self.variables[node.name]
+            raise EvaluationError(f"Undefined variable or constant: '{node.name}'")
+
+        if isinstance(node, UnaryOpNode):
+            val = self.evaluate(node.operand)
+            if node.operator == '+':
+                return +val
+            elif node.operator == '-':
+                return -val
+            raise EvaluationError(f"Unsupported unary operator: '{node.operator}'")
+
+        if isinstance(node, BinaryOpNode):
+            left_val = self.evaluate(node.left)
+            right_val = self.evaluate(node.right)
+
+            if node.operator == '+':
+                return left_val + right_val
+            elif node.operator == '-':
+                return left_val - right_val
+            elif node.operator == '*':
+                return left_val * right_val
+            elif node.operator == '/':
+                if right_val == 0:
+                    raise EvaluationError("Division by zero")
+                return left_val / right_val
+            elif node.operator == '%':
+                if right_val == 0:
+                    raise EvaluationError("Modulo by zero")
+                return left_val % right_val
+            elif node.operator in ('^', '**'):
+                try:
+                    res = left_val ** right_val
+                    if isinstance(res, complex):
+                        raise EvaluationError("Complex numbers result not supported in real domain")
+                    return res
+                except OverflowError:
+                    raise EvaluationError("Numerical overflow in exponentiation")
+
+            raise EvaluationError(f"Unsupported binary operator: '{node.operator}'")
+
+        if isinstance(node, FunctionCallNode):
+            fn_name = node.name.lower()
+            if fn_name not in self.BUILTIN_FUNCTIONS:
+                raise EvaluationError(f"Unknown or unsupported function: '{node.name}'")
+
+            args = [self.evaluate(arg) for arg in node.args]
+            func = self.BUILTIN_FUNCTIONS[fn_name]
+
+            try:
+                result = func(*args)
+                return result
+            except TypeError as e:
+                raise EvaluationError(f"Invalid arguments for function '{node.name}': {e}")
+            except ValueError as e:
+                raise EvaluationError(f"Mathematical domain error in function '{node.name}': {e}")
+
+        raise EvaluationError(f"Unknown AST node type: {type(node).__name__}")
+
+
+def evaluate_expression(expression: str, variables: Dict[str, Union[int, float]] = None) -> Union[int, float]:
+    """Helper function to tokenize, parse, and evaluate a math expression string."""
+    from expression_parser.tokenizer import Tokenizer
+    from expression_parser.parser import Parser
+
+    tokenizer = Tokenizer(expression)
+    tokens = tokenizer.tokenize()
+    parser = Parser(tokens)
+    ast = parser.parse()
+    evaluator = Evaluator(variables)
+    return evaluator.evaluate(ast)

expression_parser/main.py

@@ -0,0 +1,98 @@
+"""
+Interactive CLI REPL and runner for Mathematical Expression Parser & Evaluator.
+"""
+
+import sys
+import os
+
+# Ensure project root is in sys.path
+sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), "..")))
+
+from typing import Dict, Union
+from expression_parser import evaluate_expression, TokenizerError, ParseError, EvaluationError
+
+
+def run_repl():
+    print("=" * 60)
+    print("  Mathematical Expression Parser & Evaluator (CLI REPL)")
+    print("=" * 60)
+    print("Commands:")
+    print("  exit / quit : Exit REPL")
+    print("  vars        : Show current active variables")
+    print("  let x = val : Set variable value (e.g. let x = 10)")
+    print("  clear       : Reset custom variables")
+    print("-" * 60)
+
+    user_vars: Dict[str, Union[int, float]] = {}
+
+    while True:
+        try:
+            inp = input("math> ").strip()
+        except (KeyboardInterrupt, EOFError):
+            print("\nExiting REPL. Goodbye!")
+            break
+
+        if not inp:
+            continue
+
+        if inp.lower() in ("exit", "quit"):
+            print("Goodbye!")
+            break
+
+        if inp.lower() == "vars":
+            if not user_vars:
+                print("No custom variables set. Builtins available: pi, e, tau.")
+            else:
+                print("Active variables:", user_vars)
+            continue
+
+        if inp.lower() == "clear":
+            user_vars.clear()
+            print("Custom variables cleared.")
+            continue
+
+        if inp.lower().startswith("let "):
+            parts = inp[4:].split("=", 1)
+            if len(parts) == 2:
+                var_name = parts[0].strip()
+                val_expr = parts[1].strip()
+
+                if not var_name.isidentifier():
+                    print(f"Error: Invalid variable name '{var_name}'")
+                    continue
+
+                try:
+                    val = evaluate_expression(val_expr, user_vars)
+                    user_vars[var_name] = val
+                    print(f"Set {var_name} = {val}")
+                except Exception as e:
+                    print(f"Error evaluating variable value: {e}")
+                continue
+            else:
+                print("Usage for assignment: let <var_name> = <expression>")
+                continue
+
+        try:
+            result = evaluate_expression(inp, user_vars)
+            print(f"Result: {result}")
+        except (TokenizerError, ParseError, EvaluationError) as e:
+            print(f"Error: {e}")
+        except Exception as e:
+            print(f"Unexpected error: {e}")
+
+
+def main():
+    if len(sys.argv) > 1:
+        expr = " ".join(sys.argv[1:])
+        try:
+            res = evaluate_expression(expr)
+            print(res)
+        except Exception as e:
+            print(f"Error: {e}", file=sys.stderr)
+            sys.exit(1)
+    else:
+        run_repl()
+
+
+if __name__ == "__main__":
+    main()