def test_is_one_of_false_hierarchy(self):
t_liquid = RxType("liquid")
t_water = RxType("water", t_liquid)
t_lake = RxType("lake", t_water)
type_set = RxTypeSet()
type_set._types = {"liquid": t_liquid}
self.assertFalse(type_set.is_one_of("liquid", ["water", "lake"]))
def __init__(self,
printable_subset: Optional[Iterable[str]] = None) -> None:
self.omit_types: Set[str] = set()
self.omit_wrappers: Set[str] = set()
self._rxtypes = RxTypeSet()
self._char_sets = CharSets(self._rxtypes)
self._rxwrappers = RxWrapperSet(self._char_sets, printable_subset)
def test_add(self):
rxtype = RxType("liquid")
type_set = RxTypeSet(init_types=False)
type_set.add(rxtype)
self.assertEqual(len(type_set._types), 1)
self.assertEqual(type_set._types["liquid"], rxtype)
def test_is_of_false_unrelated(self):
t_liquid = RxType("liquid")
t_solid = RxType("solid")
type_set = RxTypeSet()
type_set._types = {"liquid": t_liquid, "solid": t_solid}
self.assertFalse(type_set.is_of("solid", "liquid"))
self.assertFalse(type_set.is_of("liquid", "solid"))
def test_is_of_excluded(self):
t_liquid = RxType("liquid")
t_water = RxType("water", t_liquid)
type_set = RxTypeSet()
type_set._types = {"liquid": t_liquid}
with self.assertRaises(KeyError):
type_set.is_of("water", "liquid")
def test_is_one_of_false_unrelated(self):
t_liquid = RxType("liquid")
t_solid = RxType("solid")
t_gas = RxType("gas")
t_plasma = RxType("plasma")
type_set = RxTypeSet()
type_set._types = {"plasma": t_plasma}
self.assertFalse(
type_set.is_one_of("plasma", ["liquid", "solid", "gas"]))
def test_is_one_of_true(self):
t_liquid = RxType("liquid")
t_solid = RxType("solid")
t_gas = RxType("gas")
t_water = RxType("water", t_liquid)
type_set = RxTypeSet()
type_set._types = {"water": t_water}
self.assertTrue(type_set.is_one_of("water",
["liquid", "solid", "gas"]))
def test_is_of_true(self):
t_liquid = RxType("liquid")
t_water = RxType("water", t_liquid)
t_lake = RxType("lake", t_water)
type_set = RxTypeSet()
type_set._types = {"water": t_water, "lake": t_lake}
self.assertTrue(type_set.is_of("water", "liquid"))
self.assertTrue(type_set.is_of("lake", "water"))
def test_init_types(self):
type_set = RxTypeSet(init_types=False)
type_set.init_types()
for settings in RXTYPE_SETTINGS:
self.assertIn(settings["name"], type_set._types)
rxtype = type_set._types[settings["name"]]
self.assertIsInstance(rxtype, RxType)
self.assertEqual(rxtype.name, settings["name"])
def test_init_char_set_empty_subset(self):
type_set = RxTypeSet()
type_set.init_types()
char_sets = CharSets(type_set)
type_name = "alphanum"
full_type_names = set(type_set.get_full_type_names(type_name))
printable_subset = set()
char_set = CHAR_SETS[type_name]
char_sets.init_char_set(type_name, printable_subset, char_set)
# no characters should be activated from type name character set
for char in char_set:
for name in full_type_names:
if name in char_sets._char_sets:
self.assertNotIn(char, char_sets._char_sets[name])
def test_init_char_set(self):
type_set = RxTypeSet()
type_set.init_types()
char_sets = CharSets(type_set)
printable_subset = set(CHAR_SETS.keys())
for settings in RXWRAPPER_SETTINGS:
if settings.get("is_char_set"):
type_name = settings.get("rxtype_name", settings["name"])
char_set = settings.get("char_set") or CHAR_SETS[type_name]
char_sets.init_char_set(type_name, printable_subset, char_set)
for type_name in CHAR_SETS.keys():
self.assertIn(type_name, char_sets._char_sets)
for char in CHAR_SETS[type_name]:
self.assertIn(char, char_sets._char_sets[type_name])
def setUp(self):
self.water_type = RxType("water")
self.fire_type = RxType("fire")
self.rxtypes = RxTypeSet()
self.char_sets = CharSets(self.rxtypes)
self.river_wrapper = RxWrapper("river", lambda node: "river",
self.water_type)
self.candle_wrapper = RxWrapper("candle", lambda node: "candle",
self.fire_type)
def test_init_char_set_full_subset(self):
type_set = RxTypeSet()
type_set.init_types()
char_sets = CharSets(type_set)
type_name = "alphanum"
full_type_names = set(type_set.get_full_type_names(type_name))
printable_subset = set(CHAR_SETS.keys())
char_set = CHAR_SETS[type_name]
char_sets.init_char_set(type_name, printable_subset, char_set)
# all related type names should contain all characters in alphanum char set
for char in char_set:
for name in full_type_names:
# only evaluate those type names that represent actual printable character sets
if name in char_sets._char_sets:
self.assertIn(char, char_sets._char_sets[name])
def test_init(self):
type_set = RxTypeSet()
char_sets = CharSets(type_set)
self.assertEqual(len(char_sets._char_sets), len(CHAR_SETS.keys()))
for key in CHAR_SETS.keys():
self.assertIn(key, char_sets._char_sets)
self.assertIsInstance(char_sets._char_sets[key], set)
self.assertEqual(len(char_sets._char_sets[key]), 0)
self.assertEqual(type_set, char_sets._rxtypes)
def test_init_char_set_partial_subset(self):
type_set = RxTypeSet()
type_set.init_types()
char_sets = CharSets(type_set)
type_names = ["digit", "alpha_lower", "alpha_upper"]
printable_subset = set(["digit", "alpha_upper"])
# Initialise character sets
for type_name in type_names:
char_set = CHAR_SETS[type_name]
char_sets.init_char_set(type_name, printable_subset, char_set)
# Search through all specified type names
for type_name in type_names:
# Confirm that type name is present in charsets
self.assertIn(type_name, char_sets._char_sets)
# Get the full list of character set type names that the current type name inherits from (inluding itself)
full_type_names = list(
filter(
lambda name: name in char_sets._char_sets,
type_set.get_full_type_names(type_name),
))
# For all characters in the character set corresponding to the current type name
for char in CHAR_SETS[type_name]:
# For all the character set types it inherits from (including itself)
for name in full_type_names:
# If the current type name is in the printable subset, it should be activated within the charset
if type_name in printable_subset:
self.assertIn(char, char_sets._char_sets[name])
# If the current type name is not in the printable subset, it should be excluded from the charset
else:
self.assertNotIn(char, char_sets._char_sets[name])
def test_empty_sets(self):
type_set = RxTypeSet()
char_sets = CharSets(type_set)
for key in char_sets._char_sets.keys():
char_sets._char_sets[key].add("a")
char_sets._char_sets[key].add("b")
char_sets._char_sets[key].add("c")
self.assertEqual(len(char_sets._char_sets[key]), 3)
char_sets.empty_sets()
self.assertEqual(len(char_sets._char_sets), len(CHAR_SETS.keys()))
for key in char_sets._char_sets.keys():
self.assertEqual(len(char_sets._char_sets[key]), 0)
def test_get_full_type_names(self):
t_liquid = RxType("liquid")
t_water = RxType("water", t_liquid)
t_lake = RxType("lake", t_water)
type_set = RxTypeSet()
type_set._types = {
"liquid": t_liquid,
"water": t_water,
"lake": t_lake
}
self.assertEqual(type_set.get_full_type_names("liquid"), ["liquid"])
self.assertEqual(type_set.get_full_type_names("water"),
["water", "liquid"])
self.assertEqual(type_set.get_full_type_names("lake"),
["lake", "water", "liquid"])
class RxNodeFactory:
def __init__(self,
printable_subset: Optional[Iterable[str]] = None) -> None:
self.omit_types: Set[str] = set()
self.omit_wrappers: Set[str] = set()
self._rxtypes = RxTypeSet()
self._char_sets = CharSets(self._rxtypes)
self._rxwrappers = RxWrapperSet(self._char_sets, printable_subset)
def set_omit(
self,
types: Optional[Union[str, Iterable[str]]] = None,
wrappers: Optional[Union[str, Iterable[str]]] = None,
) -> None:
if types and not isinstance(types, set):
types = set(types)
self.omit_types = types or set()
if wrappers and not isinstance(wrappers, set):
wrappers = set(wrappers)
self.omit_wrappers = wrappers or set()
def clear_omit(self, types: bool = False, wrappers: bool = False) -> None:
if not (types and wrappers):
self.clear_omit(types=True, wrappers=True)
if types:
self.omit_types.clear()
if wrappers:
self.omit_wrappers.clear()
def parse_rxspec(self, rxspec: RxSpec) -> NodeSpec:
if not isinstance(rxspec, list):
rxspec = [rxspec]
node_spec = {"rw_name": rxspec[0]}
for spec in rxspec[1:]:
if self._rxwrappers.wrapper_is_type(first_nested(spec), "mod"):
node_spec["modifier"] = self.parse_rxspec(spec)
else:
node_spec["children"] = [
self.parse_rxspec(child) for child in spec
]
return node_spec
def make_node(
self,
rw_name: Optional[str] = None,
children: Union[List[NodeSpec], int] = RAND,
modifier: Optional[Union[NodeSpec, int]] = None,
rxwrapper: Optional[RxWrapper] = None,
is_child: bool = False,
strict_type_match: bool = False,
) -> RxNode:
"""
children format: [{'rw_name': regex_wrapper_name, 'children': [<children>]})]
modifier format: {'rw_name': <modifier_name>, 'children': <children>, 'modifier': <modifier>}
"""
if not rxwrapper:
if not rw_name:
raise ValueError("must provide regex wrapper object or name")
rxwrapper = self._rxwrappers[rw_name]
child_nodes: List[RxNode] = []
if rxwrapper.child_count != 0:
if children == RAND:
child_types: List[str] = list(
filter(
lambda type_name: not self._rxtypes.is_one_of(
type_name, self.omit_types),
rxwrapper.child_types,
))
if rxwrapper.uniform_child_types:
child_types = sample(rxwrapper.child_types, 1)
child_nodes = [
self.make_random_node(choice(child_types), is_child=True)
for i in range(rxwrapper.get_child_count())
]
else:
for child in children:
child_nodes.append(self.make_node(**child, is_child=True))
node: RxNode = RxNode(self._char_sets, rxwrapper, child_nodes,
is_child)
if rxwrapper.is_modifiable:
if modifier == RAND:
# print("- ", node.name)
# print("- ", rxwrapper.name)
# print("- ", rxwrapper.rxtype)
# print("- ", rxwrapper.rxtype.is_type_name("mod"))
# if wrapper is not a modifier, build a modifier. Otherwise, build mod-modifier.
mod_type: str = ("mmod" if rxwrapper.rxtype.is_type_name("mod")
else "mod")
# print("- ", mod_type)
# if mod_type (to make) is mod, then don't build an mmod
# omit_types += ["mmod"] if mod_type == "mod" else []
# print(">> ", omit_types)
if mod_type not in self.omit_types:
modifier_node: RxNode = self.make_random_node(
mod_type, strict_typing=True)
node.set_modifier(modifier_node)
# print("-- ", modifier)
elif modifier:
modifier_node = self.make_node(**modifier)
node.set_modifier(modifier_node)
return node
def make_random_node(
self,
type_name: str = "re",
is_child: bool = False,
prob_modifier: float = P_MODIFIER,
strict_typing: bool = False,
) -> RxNode:
rxtype: RxType = self._rxtypes[type_name]
# filter RxWrapper.wrappers with items that match rxtype
filtered_wrappers: List[RxWrapper] = list(
filter(
lambda rxwrapper: rxwrapper.rxtype.is_type(
rxtype, strict=strict_typing),
self._rxwrappers.all(),
))
# filter out types specified for omission in node generation
for omit in self.omit_types:
omit_type: RxType = self._rxtypes[omit]
filtered_wrappers = list(
filter(
lambda rxwrapper: not rxwrapper.rxtype.is_type(omit_type),
filtered_wrappers,
))
# filter out characters if is root node and suppression parameter specified
if not is_child and SUPPRESS_ROOT_CHARS:
filtered_wrappers = list(
filter(
lambda rxwrapper: not rxwrapper.rxtype.is_type(
self._rxtypes["printable"]),
filtered_wrappers,
))
# filter out wrappers specified for omission in node generation
for omit in self.omit_wrappers:
filtered_wrappers = list(
filter(lambda rxwrapper: rxwrapper.name != omit,
filtered_wrappers))
rxwrapper: RxWrapper = choice(filtered_wrappers)
modifier: Optional[int] = None
if rxwrapper.is_modifiable and random() < prob_modifier:
modifier = RAND
return self.make_node(
rxwrapper=rxwrapper,
modifier=modifier,
is_child=is_child,
)
def setUp(self):
self.rxtypes = RxTypeSet()
self.char_sets = CharSets(self.rxtypes)
self.rxwrappers = RxWrapperSet(self.char_sets)
def test_getitem(self):
rxtype = RxType("liquid")
type_set = RxTypeSet()
type_set._types = {"liquid": rxtype}
self.assertEqual(type_set["liquid"], rxtype)
def test_getitem(self):
type_set = RxTypeSet()
char_sets = CharSets(type_set)
for key in CHAR_SETS.keys():
self.assertIsInstance(char_sets[key], set)
def test_rxtypes(self):
type_set = RxTypeSet()
char_sets = CharSets(type_set)
result = char_sets.rxtypes()
self.assertEqual(result, type_set)
def test_contains_false(self):
type_set = RxTypeSet()
char_sets = CharSets(type_set)
self.assertFalse("INVALID_KEY" in char_sets)
self.assertNotIn("INVALID_KEY", char_sets)
def test_contains_true(self):
type_set = RxTypeSet()
char_sets = CharSets(type_set)
for key in CHAR_SETS.keys():
self.assertTrue(key in char_sets)
self.assertIn(key, char_sets)