API Reference

Command Line Interface

flow run <script.flow>

Arguments:

• <script.flow>: Path to the FlowLang script file

Examples:

flow run examples/basic.flow
flow run workflow.flow

Programmatic Usage

FlowLang can be embedded in Python applications:

from flowlang.lexer import Lexer
from flowlang.parser import Parser
from flowlang.interpreter import Interpreter

# Your FlowLang-Script code
code = """
let x = 10
let y = 20
print x + y
"""

# Execute
tokens = Lexer(code).tokens
ast = Parser(tokens).parse()
Interpreter().eval(ast)

Custom Built-in Functions

Extend FlowLang with custom built-in functions:

from flowlang.interpreter import Interpreter

interp = Interpreter()

# Add custom function
interp.builtins['custom_func'] = lambda args: args[0] * 2

# Execute code that uses it
code = "print custom_func(21)"
tokens = Lexer(code).tokens
ast = Parser(tokens).parse()
interp.eval(ast)

Error Handling

FlowLang provides error messages for common issues:

• Syntax errors during parsing
• Undefined variables or functions
• Type errors in operations
• HTTP request failures
• File I/O errors

Module Structure

FlowLang consists of three main components:

• Lexer: Tokenizes the source code
• Parser: Builds an abstract syntax tree (AST)
• Interpreter: Evaluates the AST and executes the code

Return Values

When used programmatically, the interpreter returns the result of the last evaluated expression. This can be captured and used in your Python code.

Environment Variables

FlowLang scripts run in an isolated environment with access to:

• Local file system (via read_file and write_file)
• Network access (via HTTP functions)
• Standard input/output (via input and print)