PoPL/04_semantics_and_running/main.py

#!/bin/env python3

import argparse
from calendar import timegm
from copy import deepcopy
from datetime import date, timedelta
import subprocess
import tree_print
from build_ast import ASTnode, syntax_check_file

class SemData:
    def __init__(self):
        self.scope = None
        self.root = None
        self.callables = {}
        self.global_symbol_table = {}
        self.local_symbol_table = {}

def semantic_error(msg: str, node: ASTnode) -> None:
    print(f'\033[31mSemantic Error: {msg} at line {node.lineno}\033[m')
    raise SystemExit(1)

def print_todo(msg: str, node: ASTnode) -> None:
    print(f'\033[33mTODO: {msg} at line {node.lineno}\033[m')
    raise SystemExit(2)

def semantic_check(node: ASTnode, sem_data: SemData) -> None | ASTnode:
    if sem_data.root is None:
        sem_data.root = node
    match node.nodetype:
        case 'program':
            # Collect function and procedure definitions first,
            # since they can be called before they are defined
            for child in node.children_definitions:
                if child.nodetype in ['function_definition', 'procedure_definition']:
                    if child.value in sem_data.callables:
                        semantic_error(f'Redefinition of {child.nodetype.split("_")[0]} \'{child.value}\'', child)
                    sem_data.callables[child.value] = child

            # Then do the actual semantic checking
            for child in node.children_definitions:
                semantic_check(child, sem_data)
            for child in node.children_statements:
                if semantic_check(child, sem_data) is not None:
                    semantic_error(f'Expression return value is not handled', child)

            return None
        case 'variable_definition':
            # Check if variable is already defined
            symbol_table = sem_data.global_symbol_table
            if sem_data.scope is not None:
                symbol_table = sem_data.local_symbol_table
            if node.value in symbol_table:
                semantic_error(f'Redefinition of variable \'{node.value}\'', node)

            # Check if expression is valid and store it in symbol table
            variable = semantic_check(node.child_expression, sem_data)
            if variable is None or variable.type not in ['int', 'string', 'date']:
                semantic_error(f'Invalid variable type \'{variable.type if variable is not None else None}\'', node)
            symbol_table[node.value] = variable

            return None
        case 'function_definition' | 'procedure_definition':
            # Function and procedures are added to global symbol table
            # as the first step, so they can be called before they are defined
            assert node.value in sem_data.callables

            # Local symbols table should be empty while doing checking,
            # since functions and procedures can only be defined in global scope
            assert len(sem_data.local_symbol_table) == 0 and sem_data.scope is None
            sem_data.scope = node

            # Collect local arguments
            for formal in node.children_formals:
                if formal.value in sem_data.local_symbol_table:
                    semantic_error(f'Redefinition of variable \'{formal.value}\' in {node.nodetype.split("_")[0]} \'{node.value}\' arguments', node)
                sem_data.local_symbol_table[formal.value] = formal

            # Collect local variables
            for variable_definition in node.children_variable_definitions:
                semantic_check(variable_definition, sem_data)

            # Check return type
            if node.nodetype == 'function_definition':
                expression = semantic_check(node.child_expression, sem_data)
                if expression is None:
                    semantic_error(f'Function \'{node.value}\' must return a value', node)
                if node.child_return_type == 'auto':
                    node.child_return_type = expression.type
                if expression.type != node.child_return_type:
                    semantic_error(f'Function \'{node.value}\' return type is {node.child_return_type} but returns {expression.type}', node)
            elif node.nodetype == 'procedure_definition':
                returns = None
                for statement in node.children_statements:
                    returns = None
                    value = semantic_check(statement, sem_data)
                    if value is None:
                        continue
                    if value.nodetype != 'return':
                        semantic_error(f'Expression return value is not handled', statement)
                    if node.child_return_type is None:
                        semantic_error(f'Procedure \'{node.value}\' does not have a return type', node)
                    if node.child_return_type == 'auto':
                        node.child_return_type = value.type
                    if value.type != node.child_return_type:
                        semantic_error(f'Procedure \'{node.value}\' return type is {node.child_return_type} but returns {value.type}', node)
                    returns = value.type
                if returns is None and node.child_return_type is not None:
                    if node.child_return_type != 'void':
                        semantic_error(f'Procedure \'{node.value}\' must return a value when scope exits', node)
            else:
                assert False

            node.type = node.child_return_type

            sem_data.scope = None
            sem_data.local_symbol_table = {}

            return None
        case 'return':
            if sem_data.scope is None or sem_data.scope.nodetype != 'procedure_definition':
                semantic_error(f'Keyword \'return\' can only appear in procefure_definition')
            result = semantic_check(node.child_expression, sem_data)
            if result is None:
                semantic_error(f'Procedure \'{sem_data.scope.value}\' must return a value', node)
            node.type = result.type
            return node
        case 'date_literal' | 'int_literal' | 'string_literal':
            node.type = node.nodetype.split('_')[0]
            return node
        case 'assignment':
            lhs = semantic_check(node.child_lhs, sem_data)
            rhs = semantic_check(node.child_rhs, sem_data)
            if lhs is None or rhs is None or lhs.type != rhs.type:
                semantic_error(f'Invalid assignment of \'{rhs.type if rhs is not None else None}\' to \'{lhs.type if lhs is not None else None}\'', node)
            return None
        case 'binary_op':
            lhs = semantic_check(node.child_lhs, sem_data)
            rhs = semantic_check(node.child_rhs, sem_data)
            if lhs is None or rhs is None:
                semantic_error(f'Invalid operands \'{lhs.type if lhs is not None else None}\' and \'{rhs.type if rhs is not None else None}\' for binary operation {node.value}', node)

            # Validate operands and result type
            if node.value in ['*', '/']:
                if lhs.type == 'int' and rhs.type == 'int':
                    node.type = 'int'
                    return node
            elif node.value == '+':
                if lhs.type == 'date' and rhs.type == 'int':
                    node.type = 'date'
                    return node
                if lhs.type == 'int' and rhs.type == 'int':
                    node.type = 'int'
                    return node
            elif node.value == '-':
                if lhs.type == 'date' and rhs.type == 'int':
                    node.type = 'date'
                    return node
                if lhs.type == 'date' and rhs.type == 'date':
                    node.type = 'int'
                    return node
                if lhs.type == 'int' and rhs.type == 'int':
                    node.type = 'int'
                    return node
            elif node.value in ['<', '=']:
                if lhs.type == rhs.type:
                    node.type = 'bool'
                    return node

            semantic_error(f'Invalid operands \'{lhs.type}\' and \'{rhs.type}\' for operation {node.value}', node)
        case 'identifier':
            # Check if variable is defined
            symbol = None
            if node.value in sem_data.local_symbol_table:
                symbol = sem_data.local_symbol_table[node.value]
            if node.value in sem_data.global_symbol_table:
                symbol = sem_data.global_symbol_table[node.value]
            if symbol is not None:
                node.type = symbol.type
                return symbol

            semantic_error(f'Variable \'{node.value}\' not defined', node)
        case 'function_call' | 'procedure_call':
            # Handle built in functions
            if node.nodetype == 'function_call' and node.value == 'Today':
                if len(node.children_arguments) != 0:
                    semantic_error(f'Builtin function \'Today\' takes no arguments', node)
                node.type = 'date'
                return node

            # Check if function/procedure is defined
            if node.value not in sem_data.callables:
                semantic_error(f'{node.nodetype.split("_")[0]} \'{node.value}\' not defined', node)
            func = sem_data.callables[node.value]

            # Check if arguments match (count and types)
            if len(node.children_arguments) != len(func.children_formals):
                semantic_error(f'Argument count mismatch for {node.nodetype.split("_")[0]} \'{node.value}\', expected {len(func.children_formals)} but got {len(node.children_arguments)}', node)
            for formal, actual in zip(func.children_formals, node.children_arguments):
                resolved = semantic_check(actual, sem_data)
                if resolved is None or formal.type != resolved.type:
                    semantic_error(f'Argument type mismatch for {node.nodetype.split("_")[0]} \'{node.value}\', expected \'{formal.type}\' but got \'{resolved.type if resolved is not None else None}\'', node)

            # Set return type and return node if func has a return type
            node.type = func.child_return_type
            return node if node.type is not None else None
        case 'do_unless':
            # Validate condition
            condition = semantic_check(node.child_condition, sem_data)
            if condition is None or condition.type != 'bool':
                semantic_error('Condition must be of type \'bool\'', node)

            # Validate both branches
            for statement in node.children_statements_true:
                if semantic_check(statement, sem_data) is not None:
                    semantic_error(f'Expression return value is not handled', statement)
            for statement in node.children_statements_false:
                if semantic_check(statement, sem_data) is not None:
                    semantic_error(f'Expression return value is not handled', statement)

            return None
        case 'do_until':
            # Validate condition
            condition = semantic_check(node.child_condition, sem_data)
            if condition is None or condition.type != 'bool':
                semantic_error('Condition must be of type bool', node)

            # Validate body
            for statement in node.children_statements:
                if semantic_check(statement, sem_data) is not None:
                    semantic_error(f'Expression return value is not handled', statement)

            return None
        case 'unless_expression':
            # Validate condition
            condition = semantic_check(node.child_condition, sem_data)
            if condition is None or condition.type != 'bool':
                semantic_error('Condition must be of type bool', node)

            # Validate both branches
            expression_true = semantic_check(node.child_expression_true, sem_data)
            expression_false = semantic_check(node.child_expression_false, sem_data)
            if expression_true is None or expression_false is None or expression_true.type != expression_false.type:
                semantic_error(f'Branches must return the same type, got \'{expression_false.type}\' and \'{expression_true.type}\'', node)

            node.type = expression_true.type
            return node
        case 'attribute_read' | 'attribute_write':
            # Check if variable is defined
            symbol = None
            if node.child_identifier.value in sem_data.local_symbol_table:
                symbol = sem_data.local_symbol_table[node.child_identifier.value]
            elif node.child_identifier.value in sem_data.global_symbol_table:
                symbol = sem_data.global_symbol_table[node.child_identifier.value]
            else:
                semantic_error(f'Variable \'{node.child_identifier.value}\' not defined', node.child_identifier)

            # Validate attribute
            assert node.child_attribute.nodetype == 'identifier'
            if symbol.type != 'date':
                semantic_error(f'Cannot access attribute of non-date variable', node.child_attribute)
            valid_attributes = ['day', 'month', 'year']
            if node.nodetype == 'attribute_read':
                valid_attributes += ['weekday', 'weeknum']
            if node.child_attribute.value not in valid_attributes:
                semantic_error(f'Invalid attribute \'{node.child_attribute.value}\' for {node.nodetype.split("_")[0]}, allowed values {valid_attributes}', node.child_attribute)

            node.type = 'int'
            return node
        case 'print':
            for item in node.children_items:
                value = semantic_check(item, sem_data)
                if value is None or value.type not in ['int', 'string', 'date']:
                    semantic_error('Print argument can only be \'int\', \'date\' or \'string\'', node)
            return None
        case _:
            print_todo(f'Semantic check type \'{node.nodetype}\'', node)

class Instruction:
    def __init__(self, opcode: str, operands: list[str] = []):
        self.opcode = opcode
        self.operands = operands

    def __str__(self):
        return f'{self.opcode} {', '.join(self.operands)}'

class CompileData:
    def __init__(self, sem_data: SemData):
        self.sem_data = sem_data
        self.date_buffer_size = 128
        self.label_counter = 0
        self.string_literals: list[str] = []
        self.callables: dict[str, list[Instruction]] = {}
        self.scope: ASTnode = None
        self.code: list[Instruction] = []
        self.add_builtin_functions()

    def get_label(self) -> str:
        self.label_counter += 1
        return f'.L{self.label_counter - 1}'

    def insert_label(self, label) -> None:
        self.code.append(Instruction('<label>', [label]))

    def add_string_literal(self, value: str) -> str:
        for index, string in enumerate(self.string_literals):
            if string == value:
                return f'S{index}'
        self.string_literals.append(value)
        return f'S{len(self.string_literals) - 1}'

    def symbol_address(self, symbol: str) -> str:
        if self.scope is not None:
            for index, formal in enumerate(self.scope.children_formals):
                if formal.value == symbol:
                    offset = 8 * index + 16
                    return f'{offset}(%rbp)'
            for index, variable in enumerate(self.scope.children_variable_definitions):
                if variable.value == symbol:
                    offset = 8 * index + 8
                    return f'-{offset}(%rbp)'
        if symbol in self.sem_data.global_symbol_table:
            offset = 8 * list(self.sem_data.global_symbol_table.keys()).index(symbol)
            return f'(.globals + {offset})'
        assert False

    def optimize_assembly(self) -> str:
        for name, instructions in self.callables.items():
            if name.startswith('__builtin_'):
                continue

            changed = True
            while changed:
                changed = False

                i = 0
                # Remove redundant movq instructions
                #   movq %rax, %rax
                while i < len(instructions):
                    if instructions[i].opcode != 'movq' or instructions[i].operands[0] != instructions[i].operands[1]:
                        i += 1
                        continue
                    instructions.pop(i)
                    changed = True
                if changed: continue

                i = 0
                # Optimize movq to register followed by pushq register
                #   movq $1, %rax
                #   pushq %rax
                # becomes
                #   pushq $1
                while i < len(instructions) - 1:
                    if instructions[i].opcode != 'movq' or instructions[i + 1].opcode != 'pushq':
                        i += 1
                        continue
                    # push of 64 bit immediate is not possible
                    if instructions[i].operands[0][0] == '$' and int(instructions[i].operands[0][1:]) > 0xFFFFFFFF:
                        continue
                    instructions[i] = Instruction('pushq', [instructions[i].operands[0]])
                    instructions.pop(i + 1)
                    i -= 1
                    changed = True
                if changed: continue

                i = 0
                # Optimize movq to rax followed by movq from rax
                #   movq $1, %rax
                #   movq %rax, %rcx
                # becomes
                #   movq $1, %rcx
                while i < len(instructions) - 1:
                    if instructions[i].opcode != 'movq' or instructions[i + 1].opcode != 'movq':
                        i += 1
                        continue
                    if instructions[i].operands[1] != '%rax' or instructions[i + 1].operands[0] != '%rax':
                        i += 1
                        continue
                    if instructions[i].operands[0][0] not in ['$', '%'] and instructions[i + 1].operands[1][0] != '%':
                        i += 1
                        continue
                    # move of 64 bit immediate to memory is not possible
                    if instructions[i].operands[0][0] == '$' and instructions[i + 1].operands[0] != '%' and int(instructions[i].operands[0][1:]) > 0xFFFFFFFF:
                        continue
                    instructions[i] = Instruction('movq', [instructions[i].operands[0], instructions[i + 1].operands[1]])
                    instructions.pop(i + 1)
                    i -= 1
                    changed = True
                if changed: continue

                i = 0
                # Replace negative immediate in addq/subq with positive immediate
                # This is not a real optimization, but it makes the code easier to optimize
                #   subq $-1, %rax
                # becomes
                #   addq $1, %rax
                while i < len(instructions):
                    if instructions[i].opcode not in ['addq', 'subq']:
                        i += 1
                        continue
                    if instructions[i].operands[1][0] != '$':
                        i += 1
                        continue
                    value = int(instructions[i].operands[1][1:])
                    if value >= 0:
                        i += 1
                        continue
                    new_opcode = 'subq' if instructions[i].opcode == 'addq' else 'addq'
                    instructions[i] = Instruction(new_opcode, [instructions[i].operands[0], f'${-value}'])
                    changed = True
                if changed: continue

                i = 0
                # Optimize repeated addq/subq instructions
                #   addq $1, %rax
                #   addq $2, %rax
                # becomes
                #   addq $3, %rax
                while i < len(instructions) - 1:
                    if instructions[i].opcode not in ['addq', 'subq']:
                        i += 1
                        continue
                    if instructions[i].operands[1][0] != '$' or instructions[i + 1].operands[1][0] != '$':
                        i += 1
                        continue
                    if instructions[i].operands[1] != instructions[i + 1].operands[1]:
                        i += 1
                        continue
                    lhs = int(instructions[i].operands[0][1:])
                    if instructions[i].opcode == 'subq': lhs = -lhs
                    rhs = int(instructions[i + 1].operands[0][1:])
                    if instructions[i + 1].opcode == 'subq': rhs = -rhs
                    new_value = lhs + rhs
                    if abs(new_value) > 0xFFFFFFFF:
                        i += 1
                        continue
                    new_opcode = 'addq' if new_value >= 0 else 'subq'
                    instructions[i] = Instruction(new_opcode, [f'${abs(new_value)}', instructions[i].operands[1]])
                    instructions.pop(i + 1)
                    i -= 1
                    changed = True
                if changed: continue

                i = 0
                # Optimize movq immediate to register followed addq/subq with immediate
                #   movq $1, %rax
                #   addq $2, %rax
                # becomes
                #   movq $3, %rax
                while i < len(instructions) - 1:
                    if instructions[i].opcode != 'movq' or instructions[i + 1].opcode not in ['addq', 'subq']:
                        i += 1
                        continue
                    if instructions[i].operands[1] != instructions[i + 1].operands[1]:
                        i += 1
                        continue
                    if instructions[i].operands[0][0] != '$' or instructions[i + 1].operands[0][0] != '$':
                        i += 1
                        continue
                    lhs = int(instructions[i].operands[0][1:])
                    rhs = int(instructions[i + 1].operands[0][1:])
                    if instructions[i + 1].opcode == 'subq': rhs = -rhs
                    new_value = lhs + rhs
                    instructions[i] = Instruction('movq', [f'${new_value}', instructions[i].operands[1]])
                    instructions.pop(i + 1)
                    i -= 1
                    changed = True
                if changed: continue

                i = 0
                # Optimize addq/subq for immediate 1
                #   addq $1, %rax
                # becomes
                #   incq %rax
                while i < len(instructions):
                    if instructions[i].opcode not in ['addq', 'subq']:
                        i += 1
                        continue
                    if instructions[i].operands[0] != '$1':
                        i += 1
                        continue
                    new_opcode = 'incq' if instructions[i].opcode == 'addq' else 'decq'
                    instructions[i] = Instruction(new_opcode, [instructions[i].operands[1]])
                    changed = True
                if changed: continue

                i = 0
                # Optimize zeroing of register
                #   movq $0, %rax
                # becomes
                #   xorq %rax, %rax
                while i < len(instructions):
                    if instructions[i].opcode != 'movq':
                        i += 1
                        continue
                    if instructions[i].operands[0] != '$0' or instructions[i].operands[1][0] != '%':
                        i += 1
                        continue
                    instructions[i] = Instruction('xorq', [instructions[i].operands[1], instructions[i].operands[1]])
                    changed = True
                if changed: continue

    def add_builtin_functions(self) -> None:
        today = []
        today.append(Instruction('xorq', ['%rdi', '%rdi']))
        today.append(Instruction('call', ['time']))
        today.append(Instruction('movq', ['%rax', '%rdi']))
        today.append(Instruction('movq', ['$86400', '%rcx']))
        today.append(Instruction('xorq', ['%rdx', '%rdx']))
        today.append(Instruction('divq', ['%rcx']))
        today.append(Instruction('movq', ['%rdi', '%rax']))
        today.append(Instruction('subq', ['%rdx', '%rax']))
        self.callables['__builtin_today'] = today

        print_date = []
        print_date.append(Instruction('subq', ['$16', '%rsp']))
        print_date.append(Instruction('movq', ['%rdi', '0(%rsp)']))
        print_date.append(Instruction('leaq', ['0(%rsp)', '%rdi']))
        print_date.append(Instruction('call', ['localtime']))
        print_date.append(Instruction('movq', ['$.date_buffer', '%rdi']))
        print_date.append(Instruction('movq', [f'${self.date_buffer_size}', '%rsi']))
        print_date.append(Instruction('movq', ['$.date_format', '%rdx']))
        print_date.append(Instruction('movq', ['%rax', '%rcx']))
        print_date.append(Instruction('call', ['strftime']))
        print_date.append(Instruction('movq', ['$.str_format', '%rdi']))
        print_date.append(Instruction('movq', ['$.date_buffer', '%rsi']))
        print_date.append(Instruction('call', ['printf']))
        self.callables['__builtin_print_date'] = print_date

        get_day_attr = []
        get_day_attr.append(Instruction('subq', ['$16', '%rsp']))
        get_day_attr.append(Instruction('movq', ['%rdi', '0(%rsp)']))
        get_day_attr.append(Instruction('movq', ['%rsi', '8(%rsp)']))
        get_day_attr.append(Instruction('leaq', ['0(%rsp)', '%rdi']))
        get_day_attr.append(Instruction('call', ['localtime']))
        get_day_attr.append(Instruction('movq', ['$.date_buffer', '%rdi']))
        get_day_attr.append(Instruction('movq', [f'${self.date_buffer_size}', '%rsi']))
        get_day_attr.append(Instruction('movq', ['8(%rsp)', '%rdx']))
        get_day_attr.append(Instruction('movq', ['%rax', '%rcx']))
        get_day_attr.append(Instruction('call', ['strftime']))
        get_day_attr.append(Instruction('movq', ['$.date_buffer', '%rdi']))
        get_day_attr.append(Instruction('call', ['atoi']))
        self.callables['__builtin_get_day_attr'] = get_day_attr

    def get_full_code(self) -> str:
        # Data section with string literals
        code_str = '.section .data\n'
        code_str += '.int_format:     .asciz "%lld"\n'
        code_str += '.str_format:     .asciz "%s"\n'
        code_str += '.date_format:    .asciz "%Y-%m-%d"\n'
        code_str += '.day_format:     .asciz "%d"\n'
        code_str += '.month_format:   .asciz "%m"\n'
        code_str += '.year_format:    .asciz "%Y"\n'
        code_str += '.weekday_format: .asciz "%u"\n'
        code_str += '.weeknum_format: .asciz "%W"\n'
        for index, string in enumerate(self.string_literals):
            code_str += f'S{index}: .asciz "{string}"\n'
        code_str += '\n'

        # BSS section for uninitialized data
        code_str += f'.section .bss\n'
        code_str += f'.date_buffer:\n'
        code_str += f'  .skip {self.date_buffer_size}\n'
        if len(self.sem_data.global_symbol_table) != 0:
            code_str += '.globals:\n'
            code_str += f'  .skip {len(sem_data.global_symbol_table) * 8}\n'
        code_str += '\n'

        # Text section with code
        code_str += '.section .text\n'
        code_str += '\n'

        # Add function and procedure definitions
        for name, code in self.callables.items():

            if name == 'main':
                code_str += '.global main\n'
                saved_registers = []
            else:
                caller_saved = ['%r8', '%r9', '%r10', '%r11']
                saved_registers = set()
                for instruction in code:
                    for reg in caller_saved:
                        if reg in instruction.operands:
                            saved_registers.add(reg)
                            break
            saved_registers = list(saved_registers)

            does_call = False
            for instruction in code:
                if instruction.opcode == 'call':
                    does_call = True
                    break

            code_str += name + ':\n'
            if does_call:
                code_str += '  pushq %rbp\n'
                code_str += '  movq %rsp, %rbp\n'
            for reg in saved_registers:
                code_str += f'  pushq {reg}\n'
            if len(saved_registers) % 2 != 0:
                code_str += '  subq $8, %rsp\n'
            for instruction in code:
                if instruction.opcode == '<label>':
                    code_str += f'{instruction.operands[0]}:\n'
                else:
                    code_str += f'  {instruction}\n'
            if len(saved_registers) % 2 != 0:
                code_str += '  addq $8, %rsp\n'
            for reg in reversed(saved_registers):
                code_str += f'  popq {reg}\n'
            if does_call:
                code_str += '  leave\n'
            code_str += '  ret\n'
            code_str += '\n'

        return code_str

def compile_ast(node: ASTnode, compile_data: CompileData) -> None:
    match node.nodetype:
        case 'program':
            # Compile function and procedure definitions
            for definition in node.children_definitions:
                if definition.nodetype not in ['function_definition', 'procedure_definition']:
                    continue
                assert len(compile_data.code) == 0
                assert compile_data.scope is None
                compile_data.scope = definition

                # initialize local variables
                stack_size = 8 * len(definition.children_variable_definitions)
                if stack_size % 16 != 0:
                    stack_size += 8
                if stack_size != 0:
                    compile_data.code.append(Instruction('subq', [f'${stack_size}', '%rsp']))
                for variable in definition.children_variable_definitions:
                    address = compile_data.symbol_address(variable.value)
                    compile_ast(variable.child_expression, compile_data)
                    compile_data.code.append(Instruction('movq', ['%rax', address]))

                # compile statements
                if definition.nodetype == 'function_definition':
                    compile_ast(definition.child_expression, compile_data)
                elif definition.nodetype == 'procedure_definition':
                    for statement in definition.children_statements:
                        compile_ast(statement, compile_data)
                else: assert False

                # Add function/procedure to callables
                compile_data.callables[definition.value] = compile_data.code
                compile_data.code = []
                compile_data.scope = None

            # Initialize global variables
            for index, (name, variable) in enumerate(compile_data.sem_data.global_symbol_table.items()):
                address = compile_data.symbol_address(name)
                compile_ast(variable, compile_data)
                compile_data.code.append(Instruction('movq', ['%rax', address]))

            # Compile program statements
            for statement in node.children_statements:
                compile_ast(statement, compile_data)
            compile_data.code.append(Instruction('xorq', ['%rax', '%rax']))

            # Add main function
            compile_data.callables['main'] = compile_data.code
            compile_data.code = []
        case 'variable_definition' | 'function_definition' | 'procedure_definition':
            assert False
        case 'identifier':
            address = compile_data.symbol_address(node.value)
            compile_data.code.append(Instruction('movq', [address, '%rax']))
        case 'assignment':
            if node.child_lhs.nodetype == 'attribute_write':
                print_todo('Attribute write', node)
            elif node.child_lhs.nodetype == 'identifier':
                address = compile_data.symbol_address(node.child_lhs.value)
                compile_ast(node.child_rhs, compile_data)
                compile_data.code.append(Instruction('movq', ['%rax', address]))
            else: assert False
        case 'binary_op':
            assert node.value in ['+', '-', '*', '/', '<', '=']

            if node.value in ['*', '/']:
                assert node.child_lhs.type == 'int'
            else:
                assert node.child_lhs.type in ['int', 'date']

            if node.value == '-' and node.child_lhs.type == 'date':
                assert node.child_rhs.type in ['int', 'date']
            else:
                assert node.child_rhs.type == 'int'


            old_code = compile_data.code

            # compile LHS
            compile_data.code = []
            compile_ast(node.child_lhs, compile_data)
            lhs_code = compile_data.code

            # compile RHS
            compile_data.code = []
            compile_ast(node.child_rhs, compile_data)
            rhs_code = compile_data.code

            compile_data.code = old_code

            # If RHS is an 32 bit integer literal, we can use it as an immediate value
            register = None
            if node.child_rhs.nodetype == 'int_literal' and node.child_rhs.value <= 0x7FFFFFFF:
                if node.child_lhs.type != 'date':
                    register = f'${node.child_rhs.value}'
                elif node.child_rhs.value * 86400 <= 0x7FFFFFFF:
                    register = f'${node.child_rhs.value * 86400}'

            # Otherwise, we need to use a register
            usable_registers = ['%r8', '%r9', '%r10', '%r11']
            if register is None:
                register = '%rcx'
                for reg in usable_registers:
                    valid = True
                    for instruction in lhs_code:
                        if reg in instruction.operands:
                            valid = False
                            break
                    if valid:
                        register = reg
                        break

            # check if lhs uses call, this determines whether we need to align stack
            align_stack = False
            for instruction in lhs_code:
                if instruction.opcode == 'call':
                    align_stack = True
                    break

            # Add code for RHS calculation
            if register[0] == '$':
                pass
            else:
                compile_data.code += rhs_code
                if register != '%rcx':
                    compile_data.code.append(Instruction('movq', ['%rax', register]))
                elif not align_stack:
                    compile_data.code.append(Instruction('pushq', ['%rax']))
                else:
                    compile_data.code.append(Instruction('subq', ['$16', '%rsp']))
                    compile_data.code.append(Instruction('movq', ['%rax', '0(%rsp)']))

            # Add code for LHS calculation
            compile_data.code += lhs_code
            if register[0] == '$' or register != '%rcx':
                pass
            elif not align_stack:
                compile_data.code.append(Instruction('popq', [register]))
            else:
                compile_data.code.append(Instruction('movq', ['0(%rsp)', register]))
                compile_data.code.append(Instruction('addq', ['$16', '%rsp']))

            # If we are adding or subtracting dates with integers, multiply the integer by number of seconds in a day
            # If register is immediate, this has already been done
            if register[0] != '$' and node.child_lhs.type == 'date' and node.child_rhs.type == 'int':
                compile_data.code.append(Instruction('imulq', ['$86400', register]))

            # Perform operation
            if node.value == '+':
                compile_data.code.append(Instruction('addq', [register, '%rax']))
            elif node.value == '-':
                compile_data.code.append(Instruction('subq', [register, '%rax']))
            elif node.value == '*':
                compile_data.code.append(Instruction('imulq', [register, '%rax']))
            elif node.value == '/':
                # Division by immediate is not possible
                if register[0] == '$':
                    compile_data.code.append(Instruction('movq', [register, '%rcx']))
                    register = '%rcx'
                compile_data.code.append(Instruction('cqo'))
                compile_data.code.append(Instruction('idivq', [register]))
            elif node.value == '<':
                compile_data.code.append(Instruction('cmpq', [register, '%rax']))
                compile_data.code.append(Instruction('setl', ['%al']))
                compile_data.code.append(Instruction('movzbq', ['%al', '%rax']))
            elif node.value == '=':
                compile_data.code.append(Instruction('cmpq', [register, '%rax']))
                compile_data.code.append(Instruction('sete', ['%al']))
                compile_data.code.append(Instruction('movzbq', ['%al', '%rax']))
            else: assert False

            # If both operands are dates, divide result by number of seconds in a day
            if node.child_lhs.type == 'date' and node.child_rhs.type == 'date':
                assert node.value == '-'
                compile_data.code.append(Instruction('movq', ['$86400', register]))
                compile_data.code.append(Instruction('cqo'))
                compile_data.code.append(Instruction('idivq', [register]))
        case 'function_call' | 'procedure_call':
            if node.value == 'Today':
                compile_data.code.append(Instruction('call', ['__builtin_today']))
            else:
                # align stack
                stack_needed = len(node.children_arguments) * 8
                if stack_needed % 16 != 0:
                    stack_needed += 8
                if stack_needed != 0:
                    compile_data.code.append(Instruction('subq', [f'${stack_needed}', '%rsp']))

                # push arguments to the stack
                offset = 0
                for argument in node.children_arguments:
                    compile_ast(argument, compile_data)
                    compile_data.code.append(Instruction('movq', ['%rax', f'{offset}(%rsp)']))
                    offset += 8

                # call function and restore stack
                compile_data.code.append(Instruction('call', [node.value]))
                if stack_needed != 0:
                    compile_data.code.append(Instruction('addq', [f'${stack_needed}', '%rsp']))
        case 'return':
            compile_ast(node.child_expression, compile_data)
            compile_data.code.append(Instruction('leave'))
            compile_data.code.append(Instruction('ret'))
        case 'int_literal':
            compile_data.code.append(Instruction('movq', [f'${node.value}', '%rax']))
        case 'string_literal':
            label = compile_data.add_string_literal(node.value)
            compile_data.code.append(Instruction('movq', [f'${label}', '%rax']))
        case 'date_literal':
            compile_data.code.append(Instruction('movq', [f'${timegm(node.value.timetuple())}', '%rax']))
        case 'attribute_read':
            compile_ast(node.child_identifier, compile_data)
            compile_data.code.append(Instruction('movq', ['%rax', '%rdi']))
            compile_data.code.append(Instruction('movq', [f'$.{node.child_attribute.value}_format', '%rsi']))
            compile_data.code.append(Instruction('call', ['__builtin_get_day_attr']))
        case 'do_until':
            label_loop = compile_data.get_label()
            compile_data.insert_label(label_loop)

            # compile statements
            for statement in node.children_statements:
                compile_ast(statement, compile_data)

            # compile condition
            compile_ast(node.child_condition, compile_data)
            compile_data.code.append(Instruction('testq', ['%rax', '%rax']))
            compile_data.code.append(Instruction('jz', [label_loop]))
        case 'do_unless' | 'unless_expression':
            label_true = compile_data.get_label()
            label_done = compile_data.get_label()

            # compile condition
            compile_ast(node.child_condition, compile_data)
            compile_data.code.append(Instruction('testq', ['%rax', '%rax']))
            compile_data.code.append(Instruction('jnz', [label_true]))

            # compile false statements
            if node.nodetype == 'unless_expression':
                compile_ast(node.child_expression_false, compile_data)
            elif node.nodetype == 'do_unless':
                for statement in node.children_statements_false:
                    compile_ast(statement, compile_data)
            else: assert False
            compile_data.code.append(Instruction('jmp', [label_done]))

            # compile true statements
            compile_data.insert_label(label_true)
            if node.nodetype == 'unless_expression':
                compile_ast(node.child_expression_true, compile_data)
            elif node.nodetype == 'do_unless':
                for statement in node.children_statements_true:
                    compile_ast(statement, compile_data)
            else: assert False

            # add label for done
            compile_data.insert_label(label_done)
        case 'print':
            for i, item in enumerate(node.children_items):
                assert item.type in ['int', 'string', 'date']
                compile_ast(item, compile_data)

                match item.type:
                    case 'int':
                        compile_data.code.append(Instruction('movq', ['$.int_format', '%rdi']))
                        compile_data.code.append(Instruction('movq', ['%rax', '%rsi']))
                        compile_data.code.append(Instruction('call', ['printf']))
                    case 'string':
                        compile_data.code.append(Instruction('movq', ['$.str_format', '%rdi']))
                        compile_data.code.append(Instruction('movq', ['%rax', '%rsi']))
                        compile_data.code.append(Instruction('call', ['printf']))
                    case 'date':
                        compile_data.code.append(Instruction('movq', ['%rax', '%rdi']))
                        compile_data.code.append(Instruction('call', ['__builtin_print_date']))
                    case _:
                        assert False

                # Print space if there are more items
                if i < len(node.children_items) - 1:
                    compile_data.code.append(Instruction('movl', ["$' '", '%edi']))
                    compile_data.code.append(Instruction('call', ['putchar']))

            # Print newline
            compile_data.code.append(Instruction('movl', ["$'\\n'", '%edi']))
            compile_data.code.append(Instruction('call', ['putchar']))
        case _:
            print_todo(f'Compile type \'{node.nodetype}\'', node)

if __name__ == '__main__':
    parser = argparse.ArgumentParser()
    parser.add_argument('-d', '--debug', action='store_true', help='debug?')
    group = parser.add_mutually_exclusive_group(required=True)
    group.add_argument('--who', action='store_true', help='print out student IDs and NAMEs of authors')
    group.add_argument('-f', '--file', help='filename to process')
    parser.add_argument('-o', '--output', help='output filename for compiled code. default (a.out)', default='a.out')
    parser.add_argument('-a', '--assembly', help='output filename for generated assembly code')
    parser.add_argument('-O', '--optimize', action='store_true', help='run simple optimization steps on the generated assembly code')
    parser.add_argument('-r', '--run', action='store_true', help='run the compiled code after compilation')

    args = parser.parse_args()

    if args.who:
        print('Author')
        print('  Student ID: 150189237')
        print('  Name:       Oskari Alaranta')
    else:
        ast = syntax_check_file(args.file, args.debug)

        sem_data = SemData()
        semantic_check(ast, sem_data)

        if args.debug:
            tree_print.treeprint(ast, 'unicode')

        compile_data = CompileData(sem_data)
        compile_ast(ast, compile_data)

        if args.optimize:
            compile_data.optimize_assembly()

        assembly = compile_data.get_full_code()

        if args.assembly is not None:
            with open(args.assembly, 'w', encoding='utf-8') as file:
                file.write(assembly)

        subprocess.run(['gcc', '-x', 'assembler', '-o', args.output, '-static', '-'], input=assembly, encoding='utf-8')

        if args.run:
            if args.output.startswith('/'):
                subprocess.run([args.output])
            else:
                subprocess.run([f'./{args.output}'])