def test_word_with_or_without_leading_num(self) -> None:
exp_outputs = [
ParseResult(),
ParseResult(),
ParseResult(),
ParseResult(0, ''),
ParseResult(0, ''),
]
exp_outputs[3].add(6, '0brown')
exp_outputs[4].add(4, '4fox')
n_parser = NumParser()
zero_or_more_num_parser = combinator.KleeneStar(n_parser)
z_parser = AlphaParser()
word_parser = combinator.KleeneStar(z_parser)
leading_num_parser = combinator.AND(zero_or_more_num_parser,
word_parser)
print(leading_num_parser)
parser_output = []
for inp in self.inp_strings_word:
parser_output.append(leading_num_parser(inp))
for n, r in enumerate(parser_output):
print(n, r)
for n, (exp, out) in enumerate(zip(exp_outputs, parser_output)):
print("[%d / %d] : comparing %s -> %s" % \
(n, len(parser_output), str(exp), str(out))
)
assert exp == out
def test_ab_concat_offset(self) -> None:
a_parser = CharParser('a')
b_parser = CharParser('b')
ab_combo = combinator.AND(a_parser, b_parser)
exp_a_only = ParseResult(1, 'a')
exp_b_only = ParseResult(2, 'b')
a_only = a_parser('ab', idx=0)
assert exp_a_only == a_only
b_only = b_parser('ab', idx=1)
assert exp_b_only == b_only
exp_a_result = ParseResult()
exp_ab_result_0 = ParseResult(2, 'ab')
exp_ab_result_1 = ParseResult()
a_result = ab_combo('a', idx=0)
assert exp_a_result == a_result
ab_result_0 = ab_combo('ab', idx=0)
assert exp_ab_result_0 == ab_result_0
ab_result_1 = ab_combo('ab', idx=1)
assert exp_ab_result_1 == ab_result_1
def __call__(self,
inp: str,
parse_inp: ParseResult = None,
idx: int = 0) -> ParseResult:
prev_result = parse_inp if parse_inp is not None else ParseResult()
start_idx = prev_result.last_idx()
# keep parsing more tokens for as long as we can
# NOTE: this implementation here seems much more complicated than
# it needs to be....
while True:
new_result = self.A(inp, idx=prev_result.last_idx())
if new_result.empty(
) or new_result.last_idx() <= prev_result.last_idx():
break
prev_result = new_result
parser_out = ParseResult()
# if we dont accept partial results and haven't consumed the
# whole input then return an empty result
#if not self.accept_partial and result.last_idx() < len(inp):
# return parser_out
parser_out.add(prev_result.last_idx(),
inp[start_idx:prev_result.last_idx()])
return parser_out
def test_parse_strings(self) -> None:
exp_outputs_2 = [
ParseResult(),
ParseResult(),
ParseResult(3, "ayy"),
ParseResult(),
ParseResult(3, "ayy"),
ParseResult(),
ParseResult(),
ParseResult(3, "ayy"),
ParseResult(3, "ayy"),
]
s_parser = StringParser('ayy')
parser_outputs = []
for i in self.inp_strings:
parser_outputs.append(s_parser(i, idx=0))
# display
for n, o in enumerate(parser_outputs):
print('s_parser(%s) produced : %s' % (self.inp_strings[n], str(o)))
assert len(parser_outputs) == len(exp_outputs_2)
for exp_out, test_out in zip(exp_outputs_2, parser_outputs):
assert exp_out == test_out
def test_kleene_dot_num(self) -> None:
p = NumParser()
ks = combinator.KleeneDot(p)
exp_outputs = [
ParseResult(0, ''), # []
ParseResult(1, '1'), # '1'
ParseResult(1, '2'), # '2'
ParseResult(2, '12'), # '12'
ParseResult(3, '122'), # '122'
ParseResult(3, '122'), # '122a'
ParseResult(0, ''), # 'a112'
ParseResult(1, '1'), # '1a2'
]
# Parse the strings
results = []
for inp in self.inp_strings_num:
results.append(ks(inp))
print('%s results for each string :' % str(ks))
for n, r in enumerate(results):
print(n, r, repr(ks), self.inp_strings_num[n])
assert len(results) == len(exp_outputs)
for n, (exp, out) in enumerate(zip(exp_outputs, results)):
print("[%d / %d] : comparing %s -> %s" % \
(n, len(results), str(exp), str(out))
)
assert exp == out
def __call__(self, inp:str, parse_inp:ParseResult=None, idx:int=0) -> ParseResult:
if parse_inp is not None:
idx = parse_inp.last_idx()
else:
idx = 0
parse_result = ParseResult()
for target_idx, c in enumerate(inp[idx:]):
if not c.isspace():
break
parse_result.add(idx + target_idx + 1, inp[idx : idx + target_idx+1])
return parse_result
def test_kleene_star_char(self) -> None:
p = CharParser('a')
ks = combinator.KleeneStar(p)
# expected outputs
exp_outputs = [
ParseResult(0, ''), # ''
ParseResult(0, ''), # 'a'
ParseResult(0, ''), # 'aa'
ParseResult(0, ''), # 'aaa'
ParseResult(0, ''), # 'aaaa'
ParseResult(0, ''), # 'aaabcdefg'
]
# 'a'
exp_outputs[1].add(1, 'a')
# 'aa'
exp_outputs[2].add(2, 'aa')
# 'aaa'
exp_outputs[3].add(3, 'aaa')
# 'aaaa'
exp_outputs[4].add(4, 'aaaa')
# 'aaabcdefg'
exp_outputs[5].add(3, 'aaa')
# Parse the strings
results = []
for inp in self.inp_strings_alpha:
results.append(ks(inp))
print('%s results for each string :' % str(ks))
for n, r in enumerate(results):
print(n, r, repr(ks), self.inp_strings_alpha[n])
assert len(results) == len(exp_outputs)
for n, (exp, out) in enumerate(zip(exp_outputs, results)):
print("[%d / %d] : comparing %s -> %s" % \
(n, len(results), str(exp), str(out))
)
assert exp == out
# if we turn on partial match then we should get 'aaa' as the
# result for the input string 'aaabcdefg'
exp_partial_match = ParseResult(0, '')
exp_partial_match.add(3, 'aaa')
ks.accept_partial = True
partial_match_result = ks(self.inp_strings_alpha[5])
assert exp_partial_match == partial_match_result
def test_left_combinator(self) -> None:
tag_parser = CharParser("<")
iden_parser = Identifier()
left_combo = Left(tag_parser, iden_parser)
expected_result = ParseResult(1, "<")
parse_result = left_combo(self.test_element)
assert parse_result == expected_result
def test_right_combinator(self) -> None:
tag_parser = CharParser("<")
iden_parser = Identifier()
right_combo = Right(tag_parser, iden_parser)
#expected_result = ParseResult(2, "/>")
expected_result = ParseResult(14, "some-element/")
parse_result = right_combo(self.test_element)
assert parse_result == expected_result
def __call__(self,
inp: str,
parse_inp: ParseResult = None,
idx: int = 0) -> ParseResult:
prev_result = parse_inp if parse_inp is not None else ParseResult()
start_idx = prev_result.last_idx()
while True:
new_result = self.A(inp, idx=prev_result.last_idx())
if new_result.empty(
) or new_result.last_idx() <= prev_result.last_idx():
break
prev_result = new_result
parser_out = ParseResult()
parser_out.add(prev_result.last_idx(),
inp[start_idx:prev_result.last_idx()])
return parser_out
def test_one_or_more(self) -> None:
one_or_more = OneOrMore()
expected_results = [
ParseResult(),
ParseResult(17, "at least one char"),
ParseResult(9, "0 or more"),
ParseResult(9, "1 or more")
]
parse_results = []
for s in self.other_comb_inputs:
parse_results.append(one_or_more(s))
for n, res in enumerate(parse_results):
print(n, res)
assert len(parse_results) == len(expected_results)
for res, exp_res in zip(parse_results, expected_results):
assert res == exp_res
def test_identifier(self) -> None:
iden_parser = Identifier()
expected_results = [
ParseResult(26, "this-is-a-valid-identifier"),
ParseResult(6, "valid2"),
ParseResult()
]
parse_results = []
for iden in self.iden_input:
parse_results.append(iden_parser(iden))
# Display
for n, res in enumerate(parse_results):
print(n, res)
assert len(parse_results) == len(expected_results)
for res, exp_res in zip(parse_results, expected_results):
assert res == exp_res
def test_tag_open_combinator(self) -> None:
tag_parser = CharParser("<")
iden_parser = Identifier()
tag_open_combinator = AND(tag_parser, iden_parser)
# NOTE: this is what the combinator outputs, but I am not quite happy
# with it...
expected_result = ParseResult(14, "<some-element/")
parse_result = tag_open_combinator(self.test_element)
assert parse_result == expected_result
def __call__(self, inp:str, parse_inp:ParseResult=None, idx:int=0) -> ParseResult:
if parse_inp is not None:
idx = parse_inp.last_idx()
else:
idx = 0
parse_result = ParseResult()
for target_idx, c in enumerate(inp[idx:]):
if c.isalpha():
continue
elif target_idx > 0 and (c.isalnum() or c == '-'):
continue
else:
break
if target_idx == 0:
return parse_result
parse_result.add(idx + target_idx + 1, inp[idx : idx + target_idx+1])
return parse_result
def test_concatenation(self) -> None:
a_parser = CharParser('a')
b_parser = CharParser('b')
# this combinator can recognise 'ab'
ab_combo = combinator.AND(a_parser, b_parser)
# expected outputs
exp_outputs = [
ParseResult(), # ''
ParseResult(), # 'a'
ParseResult(2, 'ab'), # 'ab'
ParseResult(), # 'aa'
ParseResult(), # 'aaa'
ParseResult(2, 'ab'), # 'aba'
]
#from pudb import set_trace; set_trace()
results = []
for inp in self.inp_strings:
results.append(ab_combo(inp))
print('%s results for each string :' % str(ab_combo))
for n, r in enumerate(results):
print(n, r)
assert len(results) == len(exp_outputs)
for n, (exp, out) in enumerate(zip(exp_outputs, results)):
print("[%d / %d] : comparing %s -> %s" % \
(n, len(results), str(exp), str(out))
)
assert exp == out
def test_kleene_dot_char(self) -> None:
p = CharParser('a')
ks = combinator.KleeneDot(p)
# expected outputs
exp_outputs = [
ParseResult(0, ''), # ''
ParseResult(1, 'a'), # 'a'
ParseResult(2, 'aa'), # 'aa'
ParseResult(3, 'aaa'), # 'aaa'
ParseResult(4, 'aaaa'), # 'aaaa'
ParseResult(3, 'aaa'), # 'aaabcdefg'
]
# Parse the strings
results = []
for inp in self.inp_strings_alpha:
results.append(ks(inp))
print('%s results for each string :' % str(ks))
for n, r in enumerate(results):
print(n, r, repr(ks), self.inp_strings_alpha[n])
assert len(results) == len(exp_outputs)
for n, (exp, out) in enumerate(zip(exp_outputs, results)):
print("[%d / %d] : comparing %s -> %s" % \
(n, len(results), str(exp), str(out))
)
assert exp == out
def test_word_with_trailing_num(self) -> None:
exp_outputs = [
ParseResult(),
ParseResult(),
ParseResult(),
ParseResult(),
ParseResult(),
ParseResult(7, 'jumped2'),
]
z_parser = AlphaParser()
n_parser = NumParser()
word_parser = combinator.KleeneDot(z_parser)
trailing_num_parser = combinator.AND(word_parser, n_parser)
parser_output = []
for inp in self.inp_strings_word:
parser_output.append(trailing_num_parser(inp))
for n, r in enumerate(parser_output):
print(n, r)
for n, (exp, out) in enumerate(zip(exp_outputs, parser_output)):
print("[%d / %d] : comparing %s -> %s" % \
(n, len(parser_output), str(exp), str(out))
)
assert exp == out
def test_zero_or_more(self) -> None:
c = CharParser('a')
comb = combinator.ZeroOrMore(c)
# expected outputs
exp_outputs = [
ParseResult(1, ''), # ''
ParseResult(1, 'a'), # 'a'
ParseResult(2, 'aa'), # 'aa'
ParseResult(3, 'aaa'), # 'aaa'
ParseResult(4, 'aaaa'), # 'aaaa'
ParseResult(3, 'aaa'), # 'aaabcdefg'
]
# Parse the strings
results = []
for inp in self.inp_strings_num:
results.append(comb(inp))
print('%s results for each string :' % str(ks))
for n, r in enumerate(results):
print(n, r, repr(comb), self.inp_strings_num[n])
assert len(results) == len(exp_outputs)
for n, (exp, out) in enumerate(zip(exp_outputs, results)):
print("[%d / %d] : comparing %s -> %s" % \
(n, len(results), str(exp), str(out))
)
assert exp == out
def __call__(self,
inp: str,
parse_inp: ParseResult = None,
idx: int = 0) -> ParseResult:
# Handle the 'zeroth' time
empty_result = self.E(inp, parse_inp=parse_inp, idx=idx)
result = self.A(inp, parse_inp=empty_result)
if len(result) == 0:
return empty_result
if len(result) == 1:
return result
# run this zero or more times up to unlimited bound
start_idx = result.last_idx()
while True:
new_result = self.A(inp, idx=result.last_idx())
if new_result.last_idx() <= result.last_idx() or new_result.empty(
):
break
result = new_result
parser_out = ParseResult()
parser_out.extend(empty_result)
# if we dont accept partial results and haven't consumed the
# whole input then return an empty result
#if not self.accept_partial and result.last_idx() < len(inp):
# return parser_out
parser_out.add(result.last_idx(), inp[start_idx - 1:result.last_idx()])
return parser_out
def __call__(self,
inp: str,
parse_inp: ParseResult = None,
idx: int = 0) -> ParseResult:
start_idx = parse_inp.last_idx() if parse_inp is not None else idx
a_result = self.A(inp, parse_inp, idx=idx)
if a_result.empty() or a_result.last_idx() == start_idx:
return ParseResult()
b_result = self.B(inp, a_result)
if b_result.last_idx() <= a_result.last_idx():
return ParseResult()
parse_out = ParseResult()
parse_out.extend(a_result)
parse_out.update(b_result.last_idx(), b_result.last_str())
return parse_out
def test_alternation(self) -> None:
boo_parser = StringParser('boo')
boondoggle_parser = StringParser('boondoggle')
exp_outputs = [ParseResult()]
exp_outputs[0].add(3, 'boo')
alt_combo = combinator.OR(boo_parser, boondoggle_parser)
alt_result = alt_combo(self.inp_string_1)
print(alt_combo)
print(alt_result)
assert alt_result == exp_outputs[0]
def __call__(self,
inp: str,
parse_inp: ParseResult = None,
idx: int = 0) -> ParseResult:
result = self.A(inp, parse_inp=parse_inp)
if len(result) == 0:
return result
#if len(result) == 1:
# return result
# run this zero or more times up to unlimited bound
start_idx = result.last_idx()
while True:
new_result = self.A(inp, idx=result.last_idx())
if new_result.last_idx() <= result.last_idx() or new_result.empty(
):
break
result = new_result
parser_out = ParseResult()
parser_out.add(result.last_idx(), inp[start_idx - 1:result.last_idx()])
return parser_out
def test_quoted_string(self) -> None:
quoted_string = QuotedString()
expected_results = [
# We don't get the double-quotes in the string since the Left
# and Right combinators dont return them
ParseResult(31, "\"this is a valid quoted string\""),
ParseResult(
), # NOTE: with AND combinator this will be an empty result ParseResult(0, []),
ParseResult()
#ParseResult(2, "\"\""),
]
parse_results = []
for s in self.test_strings:
parse_results.append(quoted_string(s))
# Display
for n, res in enumerate(parse_results):
print(n, res)
assert len(parse_results) == len(expected_results)
for res, exp_res in zip(parse_results, expected_results):
assert res == exp_res
def test_kleene_star_num(self) -> None:
p = NumParser()
ks = combinator.KleeneStar(p)
exp_outputs = [
ParseResult(0, ''), # ''
ParseResult(0, ''), # '1'
ParseResult(0, ''), # '2'
ParseResult(0, ''), # '12'
ParseResult(0, ''), # '122'
ParseResult(0, ''), # '122a'
ParseResult(0, ''), # 'a112'
ParseResult(0, ''), # '1a2'
]
# '1'
exp_outputs[1].add(1, '1')
# '2'
exp_outputs[2].add(1, '2')
# '12'
exp_outputs[3].add(2, '12')
# '122'
exp_outputs[4].add(3, '122')
# '122a'
exp_outputs[5].add(3, '122')
# '1a2'
exp_outputs[7].add(1, '1') # once we see 'a' we stop
# Parse the strings
results = []
for inp in self.inp_strings_num:
results.append(ks(inp))
print('%s results for each string :' % str(ks))
for n, r in enumerate(results):
print(n, r, repr(ks), self.inp_strings_num[n])
assert len(results) == len(exp_outputs)
for n, (exp, out) in enumerate(zip(exp_outputs, results)):
print("[%d / %d] : comparing %s -> %s" % \
(n, len(results), str(exp), str(out))
)
assert exp == out
def test_inp_string_single(self) -> None:
alpha_parser = AlphaParser()
exp_outputs = [
ParseResult(), # ''
ParseResult(1, 'a'), # 'a'
ParseResult(1, 'b'), # 'b'
ParseResult(), # '1'
ParseResult(), # '2'
ParseResult(), # '{'
]
parser_outputs = []
for inp in self.inp_strings_single:
parser_outputs.append(alpha_parser(inp))
assert len(parser_outputs) == len(exp_outputs)
for exp_out, test_out in zip(exp_outputs, parser_outputs):
assert exp_out == test_out
def test_parse_char(self) -> None:
exp_outputs_1 = [
ParseResult(),
ParseResult(1, self.inp_strings_1[1][0]),
ParseResult(1, self.inp_strings_1[2][0]),
ParseResult(1, self.inp_strings_1[3][0]),
ParseResult(1, self.inp_strings_1[4][0]),
ParseResult()
]
cparser = CharParser('a')
parser_outputs = []
for i in self.inp_strings_1:
parser_outputs.append(cparser(i, idx=0))
# display
for n, o in enumerate(parser_outputs):
print('Input %d [%s] produced : %s' %
(n, self.inp_strings_1[n], str(o)))
assert len(parser_outputs) == len(exp_outputs_1)
for exp_out, test_out in zip(exp_outputs_1, parser_outputs):
assert exp_out == test_out
def test_word_one_or_more_chars(self) -> None:
exp_outputs = [
ParseResult(0, ''),
ParseResult(3, 'the'),
ParseResult(5, 'quick'),
ParseResult(0, ''),
ParseResult(0, ''),
ParseResult(6, 'jumped')
]
z_parser = AlphaParser()
word_parser = combinator.KleeneDot(z_parser)
parser_output = []
for inp in self.inp_strings_word:
parser_output.append(word_parser(inp))
for n, r in enumerate(parser_output):
print(n, r)
for n, (exp, out) in enumerate(zip(exp_outputs, parser_output)):
print("[%d / %d] : comparing %s -> %s" % \
(n, len(parser_output), str(exp), str(out))
)
assert exp == out
