def test_extend_grammar():
grammar1 = Grammar()
grammar1.add_pattern(grammar1.add_token('A'), 'a+')
grammar1.add_pattern(grammar1.add_token('B'), 'b+')
grammar1.add_pattern(grammar1.add_token('C'), 'c+')
grammar1.add_parselet('expr', kind=ParseletKind.Pratt)
grammar2 = Grammar()
grammar2.add_pattern(grammar2.add_token('A'), '_a+')
grammar2.add_pattern(grammar2.add_token('B'), '_b+')
grammar2.add_parselet('expr', kind=ParseletKind.Pratt)
result = Grammar()
initial_count = len(result.symbols)
result.extend(grammar1)
result.extend(grammar2)
assert 'A' in result.tokens
assert 'B' in result.tokens
assert 'C' in result.tokens
assert len(result.symbols) == initial_count + 4
assert len(result.parselets) == 1
assert len(result.patterns) == 5
assert {pattern.pattern.pattern
for pattern in result.patterns
} == {'a+', '_a+', 'b+', '_b+', 'c+'}
def test_extend_implicit_grammar():
grammar1 = Grammar()
grammar1.add_implicit('(')
result = Grammar()
result.extend(grammar1)
assert result.patterns[0].token_id == result.tokens['(']
assert result.patterns[0].priority == -len('(')
assert result.patterns[0].is_implicit
def create_combinator_grammar() -> Grammar:
"""
Create grammar for parse combinator definition
P.S. This grammar is used for bootstrap process of initial grammar, e.g. definition of combinators in grammar
"""
grammar = Grammar()
grammar.extend(create_core_grammar())
# tokens
name_id = grammar.tokens['Name']
string_id = grammar.tokens['String']
number_id = grammar.tokens['Integer']
colon_id = grammar.add_implicit(':')
parent_open_id = grammar.tokens['(']
parent_close_id = grammar.tokens[')']
square_open_id = grammar.tokens['[']
square_close_id = grammar.tokens[']']
curly_open_id = grammar.tokens['{']
curly_close_id = grammar.tokens['}']
less_id = grammar.tokens['<']
great_id = grammar.tokens['>']
# parse combinator definition
comb_id = grammar.add_parselet('combinator', result_type=CombinatorNode)
seq_id = grammar.add_parselet('combinator_sequence', result_type=SequenceNode)
# combinator := name: Name ":" combinator=combinator ; named variable
grammar.add_parser(
comb_id,
make_sequence(make_named('name', name_id), colon_id, make_named('combinator', comb_id)),
make_ctor(NamedNode)
)
# combinator := name: Name [ '<' priority: Number '>' ] ; reference to parselet or token
grammar.add_parser(
comb_id,
make_sequence(make_named('name', name_id), make_optional(less_id, make_named('priority', number_id), great_id)),
make_ctor(ReferenceNode)
)
# combinator := value: String ; reference to implicit token
grammar.add_parser(comb_id, make_named('value', string_id), make_ctor(ImplicitNode))
# combinator := '[' combinator: combinator_sequence ']' ; optional combinator
grammar.add_parser(
comb_id,
make_sequence(square_open_id, make_named('combinator', seq_id), square_close_id),
make_ctor(OptionalNode)
)
# combinator := '{' combinator: combinator_sequence '}' ; repeat combinator
grammar.add_parser(
comb_id,
make_sequence(curly_open_id, make_named('combinator', seq_id), curly_close_id),
make_ctor(RepeatNode)
)
# combinator := '(' combinator: combinator_sequence ')' ; parenthesis combinator
grammar.add_parser(
comb_id,
make_sequence(parent_open_id, make_named('combinator', seq_id), parent_close_id),
make_return_variable('combinator')
)
# combinator_sequence := combinators:combinator combinators:{ combinator } ; sequence combinator
grammar.add_parser(
seq_id,
make_sequence(make_named('combinators', comb_id), make_named('combinators', make_repeat(comb_id))),
make_ctor(SequenceNode)
)
return grammar