def test_make_optional():
grammar = Grammar()
name_id = grammar.add_token('Name')
comb = make_optional(name_id)
assert isinstance(comb, OptionalCombinator)
assert isinstance(comb.combinator, TokenCombinator)
assert comb.result_type == Optional[SyntaxToken]
def test_make_token():
grammar = Grammar()
name_id = grammar.add_token('Name')
comb = make_token(name_id)
assert isinstance(comb, TokenCombinator)
assert comb.token_id == name_id
assert comb.result_type == SyntaxToken
assert comb.variables == {}
def test_extend_packrat_grammar():
grammar1 = Grammar()
grammar1.add_token('Number')
grammar1.add_token('String')
expr_id = grammar1.add_parselet('expr', result_type=object)
grammar1.add_parser(expr_id, 'Number')
grammar1.add_parser(expr_id, 'String')
grammar2 = Grammar()
grammar2.add_token('Number')
grammar2.add_token('String')
expr_id = grammar2.add_parselet('expr', result_type=object)
grammar2.add_parser(expr_id, 'Number')
grammar2.add_parser(expr_id, 'String')
result = Grammar.merge(grammar1, grammar2)
expr_id = result.parselets['expr']
assert expr_id in result.tables
assert len(cast(PackratTable, result.tables[expr_id]).parselets) == 4
def test_add_token():
grammar = Grammar()
token_id = grammar.add_token('Name')
assert 'Name' in grammar.tokens
assert len(grammar.patterns) == 0
assert token_id == token_id
assert token_id.name == 'Name'
assert token_id.description == 'name'
assert not token_id.is_implicit
def test_make_sequence():
grammar = Grammar()
name_id = grammar.add_token('Name')
expr_id = grammar.add_parselet('expr')
comb = make_sequence(name_id, expr_id)
assert isinstance(comb, SequenceCombinator)
assert len(comb) == 2
assert isinstance(comb[0], TokenCombinator)
assert isinstance(comb[1], ParseletCombinator)
assert comb.result_type == SyntaxNode
def test_add_pattern():
grammar = Grammar()
token_id = grammar.add_token('Name')
result_id = grammar.add_pattern(token_id, r'[a-zA-Z]*')
assert result_id is token_id, "add_pattern must return token id"
assert len(grammar.patterns) == 1
pattern = grammar.patterns[0]
assert pattern.token_id == token_id
assert pattern.pattern == re.compile(r'[a-zA-Z]*')
assert pattern.priority == PRIORITY_MAX
assert not pattern.is_implicit
def test_add_pratt_parser():
grammar = Grammar()
expr_id = grammar.add_parselet('expr',
kind=ParseletKind.Pratt,
result_type=SyntaxToken)
integer_id = grammar.add_token('Integer')
string_id = grammar.add_token('String')
plus_id = grammar.add_implicit('+')
star_id = grammar.add_implicit('*')
table = cast(PrattTable, grammar.tables[expr_id])
assert table.prefix_tokens == set()
assert grammar.add_parser(expr_id, integer_id)
assert integer_id in table.prefix_tokens, "Cleanup of pratt table prefix tokens is not worked"
assert grammar.add_parser(expr_id, make_named('value', string_id))
assert string_id in table.prefix_tokens, "Cleanup of pratt table prefix tokens is not worked"
assert grammar.add_parser(expr_id, make_sequence(expr_id, plus_id,
expr_id))
assert grammar.add_parser(
expr_id,
make_sequence(make_named('lhs', expr_id), make_named('op', star_id),
expr_id))
def test_flat_sequence():
grammar = Grammar()
name_id = grammar.add_token('Name')
expr_id = grammar.add_parselet('expr')
combinators = tuple(
flat_sequence(TokenCombinator(name_id),
ParseletCombinator(expr_id),
SequenceCombinator((
TokenCombinator(name_id),
ParseletCombinator(expr_id),
)),
kind=SequenceCombinator))
assert len(combinators) == 4
assert isinstance(combinators[0], TokenCombinator)
assert isinstance(combinators[1], ParseletCombinator)
assert isinstance(combinators[2], TokenCombinator)
assert isinstance(combinators[3], ParseletCombinator)
def test_flat_combinator():
grammar = Grammar()
name_id = grammar.add_token('Name')
expr_id = grammar.add_parselet('expr')
# convert token id to token combinator
comb = flat_combinator(name_id)
assert isinstance(comb, TokenCombinator)
assert comb.token_id is name_id
# convert parselet id to parselet combinator
comb = flat_combinator(expr_id)
assert isinstance(comb, ParseletCombinator)
assert comb.parser_id is expr_id
assert comb.priority is None
# don't convert combinator
comb = TokenCombinator(name_id)
result = flat_combinator(comb)
assert comb is result
def grammar() -> Grammar:
grammar = Grammar()
whitespace_id = grammar.add_pattern(grammar.add_token('Whitespace'),
r'\s+')
grammar.add_trivia(whitespace_id)
grammar.add_pattern(grammar.add_token('Name'), r'[a-zA-Z_][a-zA-Z0-9]*')
grammar.add_pattern(grammar.add_token('Number'), r'[0-9]+')
make_implicit = grammar.add_implicit
expr_id = grammar.add_parselet('expr',
kind=ParseletKind.Pratt,
result_type=object)
# expr := value:Number
grammar.add_parser(expr_id, "value:Number",
make_call(lambda value: value.value, object))
# expr := lhs:expr op:'+' rhs:expr
grammar.add_parser(expr_id,
'lhs:expr "**" rhs:expr <899>',
make_call(lambda lhs, rhs: (lhs, '**', rhs), object),
priority=900)
# expr := lhs:expr op:'+' rhs:expr
grammar.add_parser(expr_id,
'lhs:expr "+" rhs:expr <600>',
make_call(lambda lhs, rhs: (lhs, '+', rhs), object),
priority=600)
# expr := lhs:expr op:'-' rhs:expr
grammar.add_parser(expr_id,
'lhs:expr "-" rhs:expr <600>',
make_call(lambda lhs, rhs: (lhs, '-', rhs), object),
priority=600)
# expr := lhs:expr op:'*' rhs:expr
grammar.add_parser(expr_id,
'lhs:expr "*" rhs:expr <700>',
make_call(lambda lhs, rhs: (lhs, '*', rhs), object),
priority=700)
# expr := lhs:expr op:'/' rhs:expr
grammar.add_parser(expr_id,
'lhs:expr "/" rhs:expr <700>',
make_call(lambda lhs, rhs: (lhs, '/', rhs), object),
priority=700)
# expr := op:'-' value:expr
grammar.add_parser(expr_id, '"-" value:expr <800>',
make_call(lambda value: ('-', value), object))
# expr := op:'-' value:expr
grammar.add_parser(expr_id, '"+" value:expr <800>',
make_call(lambda value: ('+', value), object))
# expr := '(' value:expr ')'
grammar.add_parser(expr_id, '"(" value:expr ")"',
make_return_variable('value'))
return grammar
def test_make_sequence_with_single_element():
grammar = Grammar()
name_id = grammar.add_token('Name')
comb = make_sequence(name_id)
assert isinstance(comb, TokenCombinator)
def test_add_idempotent_trivia():
grammar = Grammar()
token_id = grammar.add_token('Whitespace')
for _ in range(3):
grammar.add_trivia(token_id)
assert grammar.trivia == {token_id}
def test_add_trivia():
grammar = Grammar()
token_id = grammar.add_token('Whitespace')
assert grammar.trivia == set()
grammar.add_trivia(token_id)
assert grammar.trivia == {token_id}
def test_add_idempotent_token():
grammar = Grammar()
t1 = grammar.add_token('Name')
t2 = grammar.add_token('Name')
assert t1 is t2 and t1 == t2
def test_add_fast_parser():
grammar = Grammar()
grammar.add_token('Number')
grammar.add_parser('expr', 'Number')
