82 lines
2.8 KiB
Python
82 lines
2.8 KiB
Python
import re
|
|
|
|
class Lexer:
|
|
def __init__(self, code):
|
|
self.code = code
|
|
self.tokens = []
|
|
self.pos = 0
|
|
self.rules = [
|
|
('COMMENT', r'--\[\[.*?\].*?\]\]|--.*'),
|
|
('STRING', r'"(?:\\.|[^"\\])*"|\'(?:\\.|[^\'\])*\'|\[\[.*?\].*?\]\]'),
|
|
('NUMBER', r'\d+\.?\d*'),
|
|
('KEYWORD', r'\b(and|break|do|else|elseif|end|false|for|function|if|in|local|nil|not|or|repeat|return|then|true|until|while)\b'),
|
|
('IDENT', r'[a-zA-Z_][a-zA-Z0-9_]*'),
|
|
('OP', r'==|~=|<=|>=|\.\.\.|\.\.|>>|<<|[\+\-\*/%^#=\<>\(\)\{\}\[\];:,\.]'),
|
|
('SPACE', r'\s+')
|
|
]
|
|
|
|
def tokenize(self):
|
|
while self.pos < len(self.code):
|
|
match = None
|
|
for name, pattern in self.rules:
|
|
regex = re.compile(pattern, re.DOTALL)
|
|
match = regex.match(self.code, self.pos)
|
|
if match:
|
|
if name != 'SPACE' and name != 'COMMENT':
|
|
self.tokens.append((name, match.group(0)))
|
|
self.pos = match.end()
|
|
break
|
|
if not match:
|
|
self.pos += 1 # Skip unknown
|
|
return self.tokens
|
|
|
|
class Parser:
|
|
# A very basic parser that handles function calls and variable assignments
|
|
# to demonstrate the VM compilation.
|
|
def __init__(self, tokens):
|
|
self.tokens = tokens
|
|
self.pos = 0
|
|
|
|
def peek(self):
|
|
return self.tokens[self.pos] if self.pos < len(self.tokens) else (None, None)
|
|
|
|
def consume(self, expected_type=None):
|
|
token = self.peek()
|
|
if expected_type and token[0] != expected_type:
|
|
return None
|
|
self.pos += 1
|
|
return token
|
|
|
|
def parse(self):
|
|
nodes = []
|
|
while self.pos < len(self.tokens):
|
|
node = self.parse_statement()
|
|
if node:
|
|
nodes.append(node)
|
|
else:
|
|
self.pos += 1
|
|
return nodes
|
|
|
|
def parse_statement(self):
|
|
token = self.peek()
|
|
if token[0] == 'IDENT':
|
|
ident = self.consume()[1]
|
|
next_token = self.peek()
|
|
if next_token[1] == '(':
|
|
# Function call
|
|
self.consume() # (
|
|
args = []
|
|
while self.peek()[1] != ')':
|
|
args.append(self.peek()[1]) # Simplified: only strings/numbers/idents
|
|
self.consume()
|
|
if self.peek()[1] == ',':
|
|
self.consume()
|
|
self.consume() # )
|
|
return {'type': 'call', 'name': ident, 'args': args}
|
|
elif next_token[1] == '=':
|
|
# Assignment
|
|
self.consume() # =
|
|
value = self.consume()[1]
|
|
return {'type': 'assign', 'name': ident, 'value': value}
|
|
return None
|