Add ability to specify multiple buyers at once

Users can now specify multiple buyers, allowing the solver to figure out
the best skeleton that could be sold to any of them. This could be
useful for buyers that share a payout.
This commit is contained in:
Jeremy Saklad 2021-06-16 17:28:16 -05:00
parent 9b47e777b4
commit 5a43e0c9e6
Signed by: Jeremy Saklad
GPG Key ID: 9CA2149583EDBF84
1 changed files with 23 additions and 11 deletions

View File

@ -1158,7 +1158,7 @@ class OccasionalBuyer(enum.Enum):
]
def Solve(bone_market_fluctuations, zoological_mania, occasional_buyer = None, desired_buyer = None, maximum_cost = cp_model.INT32_MAX, maximum_exhaustion = cp_model.INT32_MAX, time_limit = float('inf'), stdscr = None):
def Solve(bone_market_fluctuations, zoological_mania, occasional_buyer = None, desired_buyers = [], maximum_cost = cp_model.INT32_MAX, maximum_exhaustion = cp_model.INT32_MAX, time_limit = float('inf'), stdscr = None):
model = cp_model.CpModel()
actions = {}
@ -1203,9 +1203,13 @@ def Solve(bone_market_fluctuations, zoological_mania, occasional_buyer = None, d
model.AddAssumptions([
actions[buyer].Not()
for unavailable_buyer in OccasionalBuyer if unavailable_buyer != occasional_buyer
for buyer in unavailable_buyer.value if buyer != desired_buyer
for buyer in unavailable_buyer.value if buyer not in desired_buyers
])
# Restrict to desired buyers
if desired_buyers:
model.Add(cp_model.LinearExpr.Sum([actions[desired_buyer] for desired_buyer in desired_buyers]) == 1)
# One torso
model.Add(cp_model.LinearExpr.Sum([value for (key, value) in actions.items() if isinstance(key, Torso)]) == 1)
@ -1216,9 +1220,6 @@ def Solve(bone_market_fluctuations, zoological_mania, occasional_buyer = None, d
# One buyer
model.Add(cp_model.LinearExpr.Sum([value for (key, value) in actions.items() if isinstance(key, Buyer)]) == 1)
# Set buyer
if desired_buyer is not None:
model.AddAssumption(actions[desired_buyer])
# Value calculation
original_value = model.NewIntVar(0, cp_model.INT32_MAX, 'original value')
@ -2137,19 +2138,28 @@ def Solve(bone_market_fluctuations, zoological_mania, occasional_buyer = None, d
class EnumAction(argparse.Action):
def __init__(self, **kwargs):
# Pop off the type value
enum = kwargs.pop("type", None)
enum = kwargs.pop('type', None)
nargs = kwargs.pop('nargs', None)
# Generate choices from the Enum
kwargs.setdefault("choices", tuple(e.name.lower() for e in enum))
kwargs.setdefault('choices', tuple(member.name.lower() for member in enum))
super(EnumAction, self).__init__(**kwargs)
self._enum = enum
self._nargs = nargs
def __call__(self, parser, namespace, values, option_string=None):
# Convert value back into an Enum
enum = self._enum[values.upper()]
if self._nargs is None or self._nargs == '?':
setattr(namespace, self.dest, enum)
else:
items = getattr(namespace, self.dest, list())
items.append(enum)
setattr(namespace, self.dest, items)
def main():
@ -2184,9 +2194,11 @@ def main():
buyer.add_argument(
'-b','--buyer', '--desired-buyer',
action=EnumAction,
nargs='+',
default=[],
type=Buyer,
help='specific buyer that skeleton should be designed for',
dest='desired_buyer'
help='specific buyer that skeleton should be designed for (if declared repeatedly, will choose from among those provided)',
dest='desired_buyers'
)
parser.add_argument(
@ -2222,7 +2234,7 @@ def main():
args = parser.parse_args()
arguments = (args.bone_market_fluctuations, args.zoological_mania, args.occasional_buyer, args.desired_buyer, args.maximum_cost, args.maximum_exhaustion, args.time_limit)
arguments = (args.bone_market_fluctuations, args.zoological_mania, args.occasional_buyer, args.desired_buyers, args.maximum_cost, args.maximum_exhaustion, args.time_limit)
if not args.verbose:
def WrappedSolve(stdscr, arguments):