def test_explicit(self):
#basicConfig(level=DEBUG)
number = Token(Digit())
letter = Token(Letter())
block = Delayed()
line = Or(Line(number),
Line(letter) & block) > list
block += Block(line[1:])
program = Trace(line[1:])
text = '''1
2
a
3
b
4
5
6
'''
program.config.lines(block_policy=explicit)
parser = program.get_parse_string()
result = parser(text)
assert result == [['1'],
['2'],
['a', ['3'],
['b', ['4'],
['5']],
['6']]], result
def test_line(self):
#basicConfig(level=DEBUG)
text = Token('[^\n\r]+')
quoted = Regexp("'[^']'")
line = Line(text(quoted))
line.config.lines(block_start=0)
parser = line.get_parse_string()
assert parser("'a'") == ["'a'"]
def test_offset(self):
#basicConfig(level=DEBUG)
text = Token('[^\n\r]+')
line = Line(text(~Literal('aa') & Regexp('.*')))
line.config.lines(block_start=0)
parser = line.get_parse_string()
assert parser('aabc') == ['bc']
# what happens with an empty match?
check = ~Literal('aa') & Regexp('.*')
check.config.no_full_first_match()
assert check.parse('aa') == ['']
assert parser('aa') == ['']
def simple_grammar(self):
'''
Test a simple example: letters introduce numbers in an indented block.
'''
#basicConfig(level=DEBUG)
number = Token(Digit())
letter = Token(Letter())
# the simplest whitespace grammar i can think of - lines are either
# numbers (which are single, simple statements) or letters (which
# mark the start of a new, indented block).
block = Delayed()
line = Or(Line(number),
Line(letter) & block) > list
# and a block is simply a collection of lines, as above
block += Block(line[1:])
program = Trace(line[1:])
program.config.lines(block_policy=1)
return program