def test_safety(self):
matcher3 = Delayed()
matcher4 = Delayed()
matcher1 = Any()[::'b', ...] & Eos()
with Separator(Drop(Any('a')[:])):
matcher2 = Any()[::'b', ...] & Eos()
# pylint: disable-msg=W0613
def target(matcher3=matcher3, matcher4=matcher4):
matcher3 += Any()[::'b', ...] & Eos()
with Separator(Drop(Any('b')[:])):
matcher4 += Any()[::'b', ...] & Eos()
t = Thread(target=target)
t.start()
t.join()
matcher5 = Any()[::'b', ...] & Eos()
matcher6 = Any()[::'b', ...] & Eos()
text = 'cababab'
assert text == matcher1.parse_string(text)[0], matcher1.parse_string(
text)
assert 'cbbb' == matcher2.parse_string(text)[0], matcher2.parse_string(
text)
assert text == matcher3.parse_string(text)[0], matcher3.parse_string(
text)
assert 'caaa' == matcher4.parse_string(text)[0], matcher4.parse_string(
text)
assert 'cbbb' == matcher5.parse_string(text)[0], matcher5.parse_string(
text)
assert text == matcher6.parse_string(text)[0], matcher6.parse_string(
text)
def test_ambiguity(self):
'''
A (signed) integer will consume a - sign.
'''
tokens = (Token(Integer()) | Token(r'\-'))[:] & Eos()
self.examples([(lambda: list(tokens.parse_all('1-2')),
"[['1', '-2']]")])
matchers = (Integer() | Literal('-'))[:] & Eos()
self.examples([(lambda: list(matchers.parse_all('1-2')),
"[['1', '-2'], ['1', '-', '2']]")])
def test_error(self):
#basicConfig(level=INFO)
class Term(Node):
pass
class Factor(Node):
pass
class Expression(Node):
pass
expression = Delayed()
number = Digit()[1:, ...] > 'number'
term = Or(
AnyBut(Space() | Digit() | '(')[1:, ...]
^ 'unexpected text: {results[0]}', number > Term,
number**make_error("no ( before '{stream_out}'") / ')' >>
node_throw, '(' / expression / ')' > Term,
('(' / expression / Eos())**make_error("no ) for '{stream_in}'") >>
node_throw)
muldiv = Any('*/') > 'operator'
factor = (term / (muldiv / term)[0:, r'\s*']) > Factor
addsub = Any('+-') > 'operator'
expression += (factor / (addsub / factor)[0:, r'\s*']) > Expression
line = expression / Eos()
parser = line.get_parse_string()
try:
parser('1 + 2 * 3 + 4 - 5)')[0]
assert False, 'expected error'
except SyntaxError as e:
assert e.msg == "no ( before ')'", e.msg
try:
parser('1 + 2 * (3 + 4 - 5')
assert False, 'expected error'
except SyntaxError as e:
assert e.msg == "no ) for '(3 + 4 - 5'", e.msg
try:
parser('1 + 2 * foo')
assert False, 'expected error'
except SyntaxError as e:
assert e.msg == "unexpected text: foo", e.msg
def test_location(self):
matcher = FullFirstMatch(Any('a')[:] & Eos())
matcher.config.clear()
try:
list(matcher.match_string('aab'))
assert False, 'expected error'
except FullFirstMatchException as e:
assert str(e) == """The match failed at 'b',
Line 1, character 2 of str: 'aab'.""", str(e)
def test_expr_with_functions(self):
'''
Expression with function calls and appropriate binding.
'''
#basicConfig(level=DEBUG)
# pylint: disable-msg=C0111, C0321
class Call(Node): pass
class Term(Node): pass
class Factor(Node): pass
class Expression(Node): pass
value = Token(Float()) > 'value'
name = Token('[a-z]+')
symbol = Token('[^a-zA-Z0-9\\. ]')
expr = Delayed()
open_ = ~symbol('(')
close = ~symbol(')')
funcn = name > 'name'
call = funcn & open_ & expr & close > Call
term = call | value | open_ & expr & close > Term
muldiv = symbol(Any('*/')) > 'operator'
factor = term & (muldiv & term)[:] > Factor
addsub = symbol(Any('+-')) > 'operator'
expr += factor & (addsub & factor)[:] > Expression
line = expr & Eos()
line.config.trace(True).lexer()
parser = line.get_parse_string()
results = str26(parser('1 + 2*sin(3+ 4) - 5')[0])
assert results == """Expression
+- Factor
| `- Term
| `- value '1'
+- operator '+'
+- Factor
| +- Term
| | `- value '2'
| +- operator '*'
| `- Term
| `- Call
| +- name 'sin'
| `- Expression
| +- Factor
| | `- Term
| | `- value '3'
| +- operator '+'
| `- Factor
| `- Term
| `- value '4'
+- operator '-'
`- Factor
`- Term
`- value '5'""", '[' + results + ']'
def _assert_string(self, separator, expecteds, streams=STREAMS_3):
with separator:
parser = And(Optional('a') & Optional('b') & 'c', Eos())
ok = True
parser.config.no_full_first_match()
for (stream, expected) in zip(streams, expecteds):
parsed = parser.parse_string(stream) is not None
if PRINT:
print('{0!r:9} : {1!r:5} {2!r:5}'
.format(stream, parsed, parsed == expected))
ok = ok and (parsed == expected)
assert ok
def test_smart_spaces(self):
with SmartSeparator1(Space()):
parser = 'a' & Optional('b') & 'c' & Eos()
parser.config.no_full_first_match()
assert parser.parse('a b c')
assert parser.parse('a c')
assert not parser.parse('a b c ')
assert not parser.parse('a c ')
assert not parser.parse('a bc')
assert not parser.parse('ab c')
assert not parser.parse('abc')
assert not parser.parse('ac')
assert not parser.parse('a c')
def test_complex(self):
#basicConfig(level=DEBUG)
class VerbPhrase(Node):
pass
class DetPhrase(Node):
pass
class SimpleTp(Node):
pass
class TermPhrase(Node):
pass
class Sentence(Node):
pass
verb = Literals('knows', 'respects', 'loves') > 'verb'
join = Literals('and', 'or') > 'join'
proper_noun = Literals('helen', 'john', 'pat') > 'proper_noun'
determiner = Literals('every', 'some') > 'determiner'
noun = Literals('boy', 'girl', 'man', 'woman') > 'noun'
verbphrase = Delayed()
verbphrase += verb | (verbphrase // join // verbphrase) > VerbPhrase
det_phrase = determiner // noun > DetPhrase
simple_tp = proper_noun | det_phrase > SimpleTp
termphrase = Delayed()
termphrase += simple_tp | (termphrase // join //
termphrase) > TermPhrase
sentence = termphrase // verbphrase // termphrase & Eos() > Sentence
sentence.config.clear().left_memoize()
p = sentence.get_match_string()
print(p.matcher.tree())
text = 'every boy or some girl and helen and john or pat knows ' \
'and respects or loves every boy or some girl and pat or ' \
'john and helen'
# text = 'every boy loves helen'
count = 0
for _meaning in p(text):
count += 1
if count < 3:
#print(_meaning[0][0])
pass
#print(count)
assert count == 392, count
class ExtensionParser(object):
"""
A class that parses extensions.
"""
class ExtensionCall(Node):
"""
An extension call.
"""
_name = None
_args = None
_kwargs = None
@property
def name(self):
return self._name[0] if self._name else None
@property
def args(self):
return tuple(self._args) if self._args else tuple()
@property
def kwargs(self):
return dict(self._kwargs) if self._kwargs else {}
COMMA = Drop(',')
NONE = Literal('None') >> (lambda x: None)
BOOL = (Literal('True') | Literal('False')) >> (lambda x: x == 'True')
IDENTIFIER = Word(Letter() | '_', Letter() | '_' | Digit())
FLOAT = Real() >> float
INTEGER = Integer() >> int
STRING = String() | String("'")
ITEM = STRING | INTEGER | FLOAT | NONE | BOOL | IDENTIFIER
with Separator(~Regexp(r'\s*')):
VALUE = Delayed()
LIST = Drop('[') & VALUE[:, COMMA] & Drop(']') > list
TUPLE = Drop('(') & VALUE[:, COMMA] & Drop(')') > tuple
VALUE += LIST | TUPLE | ITEM
ARGUMENT = VALUE >> '_args'
KWARGUMENT = (IDENTIFIER & Drop('=') & VALUE > tuple) >> '_kwargs'
ARGUMENTS = (KWARGUMENT | ARGUMENT)[:, COMMA]
NAME = IDENTIFIER > '_name'
EXTENSION = ((NAME & Drop('(') & ARGUMENTS & Drop(')'))
| NAME) & Eos() > ExtensionCall
@property
def parser(self):
return self.EXTENSION.get_parse_string()
def create_parser(delimiter):
space = Space()
comma = Drop(',') | Drop(',') + space
if delimiter == ',':
# by comma
seperator = Separator(~Regexp(r'\s*'))
delimiter = comma
else:
assert delimiter == ' ', 'delimiter "%s" not supported' % delimiter
seperator = DroppedSpace()
delimiter = space
none = Literal('None') >> (lambda x: None)
bool = (Literal('True') | Literal('False')) >> (lambda x: x == 'True')
ident = Word(Letter() | '_', Letter() | '_' | Digit())
float_ = Float() >> float
int_ = Integer() >> int
str_ = String() | String("'")
dict_key = str_ | int_ | float_ | Word()
dict_spaces = ~Whitespace()[:]
dict_value = dict_key
item = str_ | int_ | float_ | none | bool | ident | Word()
with seperator:
value = Delayed()
list_ = Drop('[') & value[:, comma] & Drop(']') > list
tuple_ = Drop('(') & value[:, comma] & Drop(')') > tuple
dict_el = dict_key & Drop(':') & value > tuple
dict_ = Drop('{') & dict_el[1:, Drop(',')] & Drop('}') > dict
value += list_ | tuple_ | dict_ | item | space
arg = value >> 'arg'
karg = (ident & Drop('=') & value > tuple) >> 'karg'
expr = (karg | arg)[:, delimiter] & Drop(Eos()) > Node
return expr.get_parse()
# # If you wish to allow use of your version of this file only under the # terms of the LGPL License and not to allow others to use your version # of this file under the MPL, indicate your decision by deleting the # provisions above and replace them with the notice and other provisions # required by the LGPL License. If you do not delete the provisions # above, a recipient may use your version of this file under either the # MPL or the LGPL License. ''' Performance related tests. ''' from lepl import Eos from lepl.apps.rfc3696 import _HttpUrl http = _HttpUrl() & Eos() # values are before/after adding IP addresses # value in pares is char count; other values are timing in secs # (89041/115471) 1.17/1.20 #http.config.clear() # (89441/115951) 1.02/1.08 #http.config.clear().direct_eval() # (3089/7835) 0.48/0.53 #http.config.clear().compile_to_nfa() # (3089/7835) 0.45/0.51 #http.config.clear().compile_to_dfa() # (3352/10148) 0.23/0.29
def test_expression2(self):
'''
As before, but with evaluation.
'''
#basicConfig(level=DEBUG)
# we could do evaluation directly in the parser actions. but by
# using the nodes instead we allow future expansion into a full
# interpreter
# pylint: disable-msg=C0111, C0321
class BinaryExpression(Node):
op = lambda x, y: None
def __float__(self):
return self.op(float(self[0]), float(self[1]))
class Sum(BinaryExpression): op = add
class Difference(BinaryExpression): op = sub
class Product(BinaryExpression): op = mul
class Ratio(BinaryExpression): op = truediv
class Call(Node):
funs = {'sin': sin,
'cos': cos}
def __float__(self):
return self.funs[self[0]](self[1])
# we use unsigned float then handle negative values explicitly;
# this lets us handle the ambiguity between subtraction and
# negation which requires context (not available to the the lexer)
# to resolve correctly.
number = Token(UnsignedFloat())
name = Token('[a-z]+')
symbol = Token('[^a-zA-Z0-9\\. ]')
expr = Delayed()
factor = Delayed()
float_ = Or(number >> float,
~symbol('-') & number >> (lambda x: -float(x)))
open_ = ~symbol('(')
close = ~symbol(')')
trig = name(Or('sin', 'cos'))
call = trig & open_ & expr & close > Call
parens = open_ & expr & close
value = parens | call | float_
ratio = value & ~symbol('/') & factor > Ratio
prod = value & ~symbol('*') & factor > Product
factor += prod | ratio | value
diff = factor & ~symbol('-') & expr > Difference
sum_ = factor & ~symbol('+') & expr > Sum
expr += sum_ | diff | factor | value
line = expr & Eos()
parser = line.get_parse()
def myeval(text):
return float(parser(text)[0])
self.assertAlmostEqual(myeval('1'), 1)
self.assertAlmostEqual(myeval('1 + 2*3'), 7)
self.assertAlmostEqual(myeval('1 - 4 / (3 - 1)'), -1)
self.assertAlmostEqual(myeval('1 -4 / (3 -1)'), -1)
self.assertAlmostEqual(myeval('1 + 2*sin(3+ 4) - 5'), -2.68602680256)
def test_calculation(self):
'''
We could do evaluation directly in the parser actions. but by
using the nodes instead we allow future expansion into a full
interpreter.
'''
# pylint: disable-msg=C0111, C0321
class BinaryExpression(Node):
op = lambda x, y: None
def __float__(self):
return self.op(float(self[0]), float(self[1]))
class Sum(BinaryExpression):
op = add
class Difference(BinaryExpression):
op = sub
class Product(BinaryExpression):
op = mul
class Ratio(BinaryExpression):
op = truediv
class Call(Node):
funs = {'sin': sin, 'cos': cos}
def __float__(self):
return self.funs[self[0]](self[1])
# we use unsigned float then handle negative values explicitly;
# this lets us handle the ambiguity between subtraction and
# negation which requires context (not available to the the lexer)
# to resolve correctly.
number = Token(UnsignedReal())
name = Token('[a-z]+')
symbol = Token('[^a-zA-Z0-9\\. ]')
expr = Delayed()
factor = Delayed()
real_ = Or(number >> float, ~symbol('-') & number >>
(lambda x: -float(x)))
open_ = ~symbol('(')
close = ~symbol(')')
trig = name(Or('sin', 'cos'))
call = trig & open_ & expr & close > Call
parens = open_ & expr & close
value = parens | call | real_
ratio = value & ~symbol('/') & factor > Ratio
prod = value & ~symbol('*') & factor > Product
factor += prod | ratio | value
diff = factor & ~symbol('-') & expr > Difference
sum_ = factor & ~symbol('+') & expr > Sum
expr += sum_ | diff | factor | value
line = expr & Eos()
parser = line.get_parse()
def calculate(text):
return float(parser(text)[0])
self.examples([(lambda: calculate('1'), '1.0'),
(lambda: calculate('1 + 2*3'), '7.0'),
(lambda: calculate('-1 - 4 / (3 - 1)'), '-3.0'),
(lambda: calculate('1 -4 / (3 -1)'), '-1.0'),
(lambda: str(calculate('1 + 2*sin(3+ 4) - 5'))[:5],
'-2.68')])
def target(matcher3=matcher3, matcher4=matcher4):
matcher3 += Any()[::'b', ...] & Eos()
with Separator(Drop(Any('b')[:])):
matcher4 += Any()[::'b', ...] & Eos()