def obj_from_pddl(pddl):
if pddl in Object._obj_from_name:
return Object.from_name(pddl)
elif pddl in OptimisticObject._obj_from_name:
return OptimisticObject.from_name(pddl)
raise ValueError(pddl)
def reset_globals():
# TODO: maintain these dictionaries in an object
Object.reset()
OptimisticObject.reset()
RULES[:] = []
SOLUTIONS[:] = []
def obj_from_value_expression(parent):
return replace_expression(
parent, lambda o: o if is_parameter(o) else Object.from_value(o))
def convert_constants(fact):
# TODO: take the constant map as an input
# TODO: throw an error if undefined
return Fact(get_prefix(fact), [
p if is_parameter(p) else Object.from_name(p) for p in get_args(fact)
])
def to_obj(value):
# Allows both raw values as well as objects to be specified
if any(isinstance(value, Class) for Class in [Object, OptimisticObject]):
return value
return Object.from_value(value)
def mapping(self):
if self._mapping is None:
self._mapping = get_mapping(self.external.inputs, self.input_objects)
for constant in self.external.constants:
self._mapping[constant] = Object.from_name(constant)
return self._mapping
import sys
from pddlstream.algorithms.algorithm import get_predicates
from pddlstream.algorithms.downward import get_literals, get_conjunctive_parts, fd_from_fact, EQ, make_object, \
pddl_from_instance, DEFAULT_MAX_TIME
from pddlstream.language.object import Object
from pddlstream.language.conversion import obj_from_pddl, substitute_fact
from pddlstream.language.fluent import get_predicate_map, remap_certified
from pddlstream.language.stream import Stream
from pddlstream.utils import INF, invert_dict, get_mapping
# Intuition: static facts about whether this state satisfies a condition
# The state can be seen as a hidden parameter with a precondition that you are at it
PYPLANNERS_VAR = 'PYPLANNERS_PATH'
PLACEHOLDER = Object.from_value('~')
def get_pyplanners_path():
return os.environ.get(PYPLANNERS_VAR, None)
def has_attachments(domain):
return any(getattr(action, 'attachments', {}) for action in domain.actions)
def compile_fluents_as_attachments(domain, externals):
import pddl
state_streams = set(
filter(lambda e: isinstance(e, Stream) and e.is_fluent(),
externals)) # is_special
def obj_from_existential_expression(parent): # obj_from_value_expression
return replace_expression(
parent, lambda o: OptimisticObject.from_opt(o, o)
if is_parameter(o) else Object.from_value(o))
from pddlstream.algorithms.advanced import get_predicates
from pddlstream.algorithms.downward import get_literals, get_conjunctive_parts, fd_from_fact, EQ, make_object, \
pddl_from_instance, DEFAULT_MAX_TIME, get_cost_scale
from pddlstream.language.object import Object
from pddlstream.language.conversion import obj_from_pddl, substitute_fact
from pddlstream.language.fluent import get_predicate_map, remap_certified
from pddlstream.language.stream import Stream
from pddlstream.utils import INF, invert_dict, get_mapping, safe_zip
# Intuition: static facts about whether this state satisfies a condition
# The state can be seen as a hidden parameter with a precondition that you are at it
# TODO: refactor to algorithms
PYPLANNERS_VAR = 'PYPLANNERS_PATH'
PLACEHOLDER_OBJ = Object.from_value('~')
DEFAULT_PYPLANNER = {
'search': 'eager',
'evaluator': 'greedy',
'heuristic': 'ff',
'successors': 'all',
}
def get_pyplanners_path():
return os.environ.get(PYPLANNERS_VAR, None)
def has_attachments(domain):
return any(getattr(action, 'attachments', {}) for action in domain.actions)
def parse_value(value):
return OptimisticObject.from_opt(
value, value) if is_parameter(value) else Object.from_value(value)