import collections
+def consume_newlines(index, tokens):
+ while index < len(tokens) and tokens[index].type == 'newline':
+ index += 1
+
+ return True, index, None
+
def _or_parser(*parsers):
def result_parser(index, tokens):
failure = (False, index, None)
],
)
-FurParenthesizedExpression = collections.namedtuple(
- 'FurParenthesizedExpression',
- [
- 'internal',
- ],
-)
-
FurInfixExpression = collections.namedtuple(
'FurInfixExpression',
[
tokens[index].match,
))
- return True, index, FurParenthesizedExpression(internal=internal)
+ return True, index, internal
def _negation_expression_parser(index, tokens):
failure = (False, index, None)
_symbol_expression_parser,
)(index, tokens)
-def _left_recursive_infix_operator_parser(token_type, operand_parser, order):
+def _left_recursive_infix_operator_parser(operator_token_matcher, operand_parser, order):
def result_parser(index, tokens):
failure = (False, index, None)
if not success:
return failure
- while success and index < len(tokens) and tokens[index].type == token_type:
+ while success and index < len(tokens) and operator_token_matcher(tokens[index]):
success = False
if index + 1 < len(tokens):
def _multiplication_level_expression_parser(index, tokens):
return _left_recursive_infix_operator_parser(
- 'multiplication_level_operator',
+ lambda token: token.type == 'multiplication_level_operator',
_literal_level_expression_parser,
'multiplication_level',
)(index, tokens)
def _addition_level_expression_parser(index, tokens):
return _left_recursive_infix_operator_parser(
- 'addition_level_operator',
+ lambda token: token.type == 'addition_level_operator',
_multiplication_level_expression_parser,
'addition_level',
)(index, tokens)
-def _equality_level_expression_parser(index, tokens):
+def _comparison_level_expression_parser(index, tokens):
return _left_recursive_infix_operator_parser(
- 'equality_level_operator',
+ lambda token: token.type == 'comparison_level_operator',
_addition_level_expression_parser,
- 'equality_level',
+ 'comparison_level',
)(index, tokens)
-def _comma_separated_list_parser(index, tokens):
- failure = (False, index, None)
-
- expressions = []
+def _and_level_expression_parser(index, tokens):
+ return _left_recursive_infix_operator_parser(
+ lambda token: token.type == 'symbol' and token.match == 'and',
+ _comparison_level_expression_parser,
+ 'and_level',
+ )(index, tokens)
- success, index, expression = _expression_parser(index, tokens)
+def _or_level_expression_parser(index, tokens):
+ return _left_recursive_infix_operator_parser(
+ lambda token: token.type == 'symbol' and token.match == 'or',
+ _and_level_expression_parser,
+ 'or_level',
+ )(index, tokens)
- if success:
- expressions.append(expression)
- else:
- return failure
+def _comma_separated_list_parser(subparser):
+ def result_parser(index, tokens):
+ start_index = index
- while success and index < len(tokens) and tokens[index].type == 'comma':
- success = False
+ items = []
- if index + 1 < len(tokens):
- success, try_index, expression = _expression_parser(index + 1, tokens)
+ success, index, item = subparser(index, tokens)
if success:
- expressions.append(expression)
- index = try_index
+ items.append(item)
+ else:
+ return (True, start_index, ())
- return True, index, tuple(expressions)
+ while success and index < len(tokens) and tokens[index].type == 'comma':
+ success = False
+
+ if index + 1 < len(tokens):
+ success, try_index, item = subparser(index + 1, tokens)
+ if success:
+ items.append(item)
+ index = try_index
+
+ return True, index, tuple(items)
+
+ return result_parser
+
+def _comma_separated_expression_list_parser(index, tokens):
+ return _comma_separated_list_parser(_expression_parser)(index, tokens)
FurFunctionCallExpression = collections.namedtuple(
'FurFunctionCallExpression',
],
)
+FurExpressionStatement = collections.namedtuple(
+ 'FurExpressionStatement',
+ [
+ 'expression',
+ ],
+)
+
FurAssignmentStatement = collections.namedtuple(
'FurAssignmentStatement',
[
],
)
+FurFunctionDefinitionStatement = collections.namedtuple(
+ 'FurFunctionDefinitionStatement',
+ [
+ 'name',
+ 'argument_name_list',
+ 'statement_list',
+ ],
+)
+
FurProgram = collections.namedtuple(
'FurProgram',
[
)
def _function_call_expression_parser(index, tokens):
- # TODO Use a FurSymbolExpression for the name
+ # TODO Allow function calls as the source of the function. This requires a
+ # left-recursive parser, however.
failure = (False, index, None)
- success, index, function = _symbol_expression_parser(index, tokens)
+ # We have to be careful what expressions we add here. Otherwise expressions
+ # like "a + b()" become ambiguous to the parser.
+ success, index, function = _or_parser(
+ _symbol_expression_parser,
+ _parenthesized_expression_parser,
+ )(index, tokens)
if not success:
return failure
return failure
index += 1
- success, index, arguments = _comma_separated_list_parser(index, tokens)
+ success, index, arguments = _comma_separated_expression_list_parser(index, tokens)
if not success:
return failure
return True, index, FurFunctionCallExpression(function=function, arguments=arguments)
-_expression_parser = _equality_level_expression_parser
+_expression_parser = _or_level_expression_parser
+
+def _expression_statement_parser(index, tokens):
+ failure = (False, index, None)
+
+ success, index, expression = _expression_parser(index, tokens)
+
+ if not success:
+ return failure
+
+ return (True, index, FurExpressionStatement(expression=expression))
def _assignment_statement_parser(index, tokens):
# TODO Use a FurSymbolExpression for the target? Maybe this is actually not a good idea
return True, index, FurAssignmentStatement(target=target, expression=expression)
+def _function_definition_statement_parser(index, tokens):
+ failure = (False, index, None)
+
+ if tokens[index].type == 'keyword' and tokens[index].match == 'def':
+ index += 1
+ else:
+ return failure
+
+ if tokens[index].type == 'symbol':
+ name = tokens[index].match
+ index += 1
+ else:
+ raise Exception('Expected function name, found "{}" on line {}'.format(
+ tokens[index].match,
+ tokens[index].line,
+ ))
+
+ if tokens[index].type == 'open_parenthese':
+ index += 1
+ else:
+ raise Exception('Expected "(", found "{}" on line {}'.format(
+ tokens[index].match,
+ tokens[index].line,
+ ))
+
+ success, index, argument_name_list = _comma_separated_list_parser(_symbol_expression_parser)(
+ index,
+ tokens,
+ )
+
+ if tokens[index].type == 'close_parenthese':
+ index += 1
+ else:
+ raise Exception('Expected ")", found "{}" on line {}'.format(
+ tokens[index].match,
+ tokens[index].line,
+ ))
+
+ if tokens[index].type == 'symbol' and tokens[index].match == 'do':
+ index += 1
+ else:
+ return failure
+
+ success, index, statement_list = _zero_or_more_parser(tuple, _statement_parser)(index, tokens)
+
+ _, index, _ = consume_newlines(index, tokens)
+
+ if tokens[index].type == 'keyword' and tokens[index].match == 'end':
+ index += 1
+ else:
+ return failure
+
+ return True, index, FurFunctionDefinitionStatement(
+ name=name,
+ argument_name_list=tuple(an.value for an in argument_name_list),
+ statement_list=statement_list,
+ )
+
def _statement_parser(index, tokens):
- # TODO It would be good to include newlines in the parsing of this because it removes the ambiguity between "function(argument)" (one statement) and "function\n(argument)" (two statements)
+ _, index, _ = consume_newlines(index, tokens)
+
+ if index == len(tokens):
+ return (False, index, None)
+
return _or_parser(
_assignment_statement_parser,
- _expression_parser,
+ _expression_statement_parser,
+ _function_definition_statement_parser,
)(index, tokens)
def _program_formatter(statement_list):