{- Copyright (C) 2018 Dr. Alistair Ward This file is part of BishBosh. BishBosh is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version. BishBosh is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with BishBosh. If not, see . -} {- | [@AUTHOR@] Dr. Alistair Ward [@DESCRIPTION@] Searches for the optimal /move/ from those currently available. -} module BishBosh.Search.Search( -- * Types -- ** Data-types Result ( -- MkResult, getSearchState, getQuantifiedGames, getNMovesEvaluated ), -- * Constants showsSeparator, -- * Functions search, calculateBranchingFactor ) where import Control.Arrow((&&&)) import qualified BishBosh.Component.Move as Component.Move import qualified BishBosh.Data.Exception as Data.Exception import qualified BishBosh.Evaluation.PositionHashQuantifiedGameTree as Evaluation.PositionHashQuantifiedGameTree import qualified BishBosh.Evaluation.QuantifiedGame as Evaluation.QuantifiedGame import qualified BishBosh.Input.SearchOptions as Input.SearchOptions import qualified BishBosh.Model.Game as Model.Game import qualified BishBosh.Notation.MoveNotation as Notation.MoveNotation import qualified BishBosh.Search.AlphaBeta as Search.AlphaBeta import qualified BishBosh.Search.SearchState as Search.SearchState import qualified BishBosh.State.TurnsByLogicalColour as State.TurnsByLogicalColour import qualified BishBosh.Text.ShowList as Text.ShowList import qualified Control.DeepSeq import qualified Control.Exception import qualified Control.Monad.Reader import qualified Data.Maybe -- | The type returned by 'search'. data Result x y positionHash criterionValue weightedMean = MkResult { getSearchState :: Search.SearchState.SearchState x y positionHash criterionValue weightedMean, getQuantifiedGames :: [Evaluation.QuantifiedGame.QuantifiedGame x y criterionValue weightedMean], -- ^ The optimal path down the /positionHashQuantifiedGameTree/. getNMovesEvaluated :: Component.Move.NMoves -- ^ The total number of nodes in the /positionHashQuantifiedGameTree/ which were analysed. } instance Control.DeepSeq.NFData weightedMean => Control.DeepSeq.NFData (Result x y positionHash criterionValue weightedMean) where rnf MkResult { getQuantifiedGames = quantifiedGames } = Control.DeepSeq.rnf quantifiedGames -- CAVEAT: don't evaluate the search-state, since this contains the PositionHashQuantifiedGameTree ! -- | Used to format output. showsSeparator :: ShowS showsSeparator = showString " -> " instance (Enum x, Enum y, Real criterionValue, Real weightedMean) => Notation.MoveNotation.ShowNotationFloat (Result x y positionHash criterionValue weightedMean) where showsNotationFloat moveNotation showsDouble result@MkResult { getQuantifiedGames = quantifiedGames, getNMovesEvaluated = nMovesEvaluated } = Text.ShowList.showsFormattedList showsSeparator ( Notation.MoveNotation.showsNotationFloat moveNotation showsDouble ) quantifiedGames . showString "; selected after analysing " . shows nMovesEvaluated . showString " moves (branching-factor" . Text.ShowList.showsAssociation . showsDouble ( calculateBranchingFactor result ) . showChar ')' -- | Initiates the recursive function 'Search.AlphaBeta.negaMax', then unpacks the results. search :: ( Enum x, Enum y, Eq criterionValue, Num weightedMean, Ord weightedMean, Ord positionHash, Ord x, Ord y ) => Input.SearchOptions.SearchDepth -- ^ How deep down the tree to search. -> Search.SearchState.SearchState x y positionHash criterionValue weightedMean -> Input.SearchOptions.Reader (Result x y positionHash criterionValue weightedMean) search 0 _ = Control.Exception.throw . Data.Exception.mkOutOfBounds . showString "BishBosh.Search.Search.search:\t" . shows Input.SearchOptions.searchDepthTag . showString " must be at least " $ shows Input.SearchOptions.minimumSearchDepth "." search searchDepth searchState | Just terminationReason <- Model.Game.getMaybeTerminationReason game = Control.Exception.throw . Data.Exception.mkInvalidDatum . showString "BishBosh.Search.Search.search:\tthe game has already terminated; " $ shows terminationReason "." | otherwise = do (maybeRetireKillerMovesAfter, maybeRetireTranspositionsAfter) <- Control.Monad.Reader.asks $ Input.SearchOptions.getMaybeRetireKillerMovesAfter &&& Input.SearchOptions.maybeRetireTranspositionsAfter let nPlies = State.TurnsByLogicalColour.getNPlies $ Model.Game.getTurnsByLogicalColour game searchResult <- Search.AlphaBeta.negaMax searchDepth $ Search.SearchState.euthanise nPlies maybeRetireKillerMovesAfter maybeRetireTranspositionsAfter searchState case Search.AlphaBeta.extractSelectedTurns nPlies searchResult of (dynamicMoveData, turns@(turn : _), nMovesEvaluated) -> let isMatch turn' = (== turn') . Evaluation.QuantifiedGame.getLastTurn . Evaluation.PositionHashQuantifiedGameTree.getQuantifiedGame in return {-to Reader-monad-} MkResult { getSearchState = Search.SearchState.mkSearchState ( Data.Maybe.fromMaybe ( Control.Exception.throw $ Data.Exception.mkIncompatibleData "BishBosh.Search.Search.search:\tBishBosh.Data.RoseTree.reduce failed." ) $ Evaluation.PositionHashQuantifiedGameTree.reduce (isMatch turn) positionHashQuantifiedGameTree ) dynamicMoveData, getQuantifiedGames = map Evaluation.PositionHashQuantifiedGameTree.getQuantifiedGame . Data.Maybe.fromMaybe ( Control.Exception.throw $ Data.Exception.mkSearchFailure "BishBosh.Search.Search.search:\tBishBosh.Data.RoseTree.traceRoute failed." ) $ Evaluation.PositionHashQuantifiedGameTree.traceRoute isMatch positionHashQuantifiedGameTree turns, getNMovesEvaluated = nMovesEvaluated } _ -> Control.Exception.throw $ Data.Exception.mkNullDatum "BishBosh.Search.Search.search:\tzero turns selected." where positionHashQuantifiedGameTree = Search.SearchState.getPositionHashQuantifiedGameTree searchState game = Evaluation.QuantifiedGame.getGame $ Evaluation.PositionHashQuantifiedGameTree.getRootQuantifiedGame positionHashQuantifiedGameTree -- | Calculate the geometric-mean of the number of moves evaluated at each node. calculateBranchingFactor :: Floating branchingFactor => Result x y positionHash criterionValue weightedMean -> branchingFactor calculateBranchingFactor MkResult { getQuantifiedGames = quantifiedGames, getNMovesEvaluated = nMovesEvaluated } | null quantifiedGames = Control.Exception.throw $ Data.Exception.mkNullDatum "BishBosh.Search.Search.calculateBranchingFactor:\tnull quantifiedGames." | nMovesEvaluated == 0 = Control.Exception.throw $ Data.Exception.mkInsufficientData "BishBosh.Search.Search.calculateBranchingFactor:\tzero moves analysed." | otherwise = fromIntegral nMovesEvaluated ** recip ( fromIntegral $ length quantifiedGames -- The search-depth. )