{- Copyright (C) 2011-2015 Dr. Alistair Ward This program 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. This program 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 this program. If not, see . -} {- | [@AUTHOR@] Dr. Alistair Ward [@DESCRIPTION@] * Implements several different prime-factorisation algorithms. * . -} module Factory.Math.Implementations.PrimeFactorisation( -- * Types -- ** Data-types Algorithm( -- DixonsMethod, FermatsMethod, TrialDivision ) -- * Functions -- factoriseByDixonsMethod -- factoriseByFermatsMethod -- factoriseByTrialDivision ) where import Control.Arrow((&&&)) import qualified Control.Arrow import qualified Control.DeepSeq import qualified Control.Parallel.Strategies import qualified Data.Default import qualified Data.Maybe import qualified Data.Numbers.Primes import qualified Factory.Data.Exponential as Data.Exponential import Factory.Data.Exponential((<^)) import qualified Factory.Data.PrimeFactors as Data.PrimeFactors import qualified Factory.Math.PerfectPower as Math.PerfectPower import qualified Factory.Math.Power as Math.Power import qualified Factory.Math.PrimeFactorisation as Math.PrimeFactorisation import qualified ToolShed.Data.Pair -- | The algorithms by which prime-factorisation has been implemented. data Algorithm = DixonsMethod -- ^ . | FermatsMethod -- ^ . | TrialDivision -- ^ . deriving (Eq, Read, Show) instance Data.Default.Default Algorithm where def = TrialDivision instance Math.PrimeFactorisation.Algorithmic Algorithm where primeFactors algorithm = case algorithm of DixonsMethod -> factoriseByDixonsMethod FermatsMethod -> Data.PrimeFactors.reduce . factoriseByFermatsMethod TrialDivision -> factoriseByTrialDivision -- | . factoriseByDixonsMethod :: Integral base => base -> Data.PrimeFactors.Factors base exponent factoriseByDixonsMethod = undefined {- | * . * . * . * @i = f1 * f2@ Assume a non-trivial factorisation, ie. one in which both factors exceed one. => @i = (larger + smaller) * (larger - smaller)@ Represent the co-factors as a sum and difference. => @i = larger^2 - smaller^2@ Which has an integral solution if @i@ is neither /even/ nor a /perfect square/. => @sqrt (larger^2 - i) = smaller@ Search for /larger/, which results in an integral value for /smaller/. * Given that the smaller factor /f2/, can't be less than 3 (/i/ isn't /even/), then the larger /f1/, can't be greater than @(i `div` 3)@. So: @(f2 >= 3) && (f1 <= i `div` 3)@ Two equations which can be used to solve for /larger/. => @(larger - smaller >= 3) && (larger + smaller <= i `div` 3)@ Add these to eliminate /smaller/. => @larger <= (i + 9) `div` 6@ The upper bound of the search-space. * This algorithm works best when there's a factor close to the /square-root/. -} factoriseByFermatsMethod :: ( Control.DeepSeq.NFData base, Control.DeepSeq.NFData exponent, Integral base, Num exponent ) => base -> Data.PrimeFactors.Factors base exponent factoriseByFermatsMethod i | i <= 3 = [Data.Exponential.rightIdentity i] | even i = Data.Exponential.rightIdentity 2 : factoriseByFermatsMethod (i `div` 2) {-recurse-} | Data.Maybe.isJust maybeSquareNumber = (<^ 2) `map` factoriseByFermatsMethod (Data.Maybe.fromJust maybeSquareNumber) {-recurse-} | null factors = [Data.Exponential.rightIdentity i] -- Prime. | otherwise = uncurry (++) . Control.Parallel.Strategies.withStrategy ( Control.Parallel.Strategies.parTuple2 Control.Parallel.Strategies.rdeepseq Control.Parallel.Strategies.rdeepseq -- CAVEAT: unproductive on the size of integers tested so far. ) . ToolShed.Data.Pair.mirror factoriseByFermatsMethod $ head factors where -- maybeSquareNumber :: Integral i => Maybe i maybeSquareNumber = Math.PerfectPower.maybeSquareNumber i -- factors :: Integral i => [i] factors = map ( ( uncurry (+) &&& uncurry (-) -- Construct the co-factors as the sum and difference of /larger/ and /smaller/. ) . Control.Arrow.second Data.Maybe.fromJust ) . filter ( Data.Maybe.isJust . snd -- Search for a perfect square. ) . map ( Control.Arrow.second $ Math.PerfectPower.maybeSquareNumber {-hotspot-} . (+ negate i) -- Associate the corresponding value of /smaller/. ) . takeWhile ( (<= (i + 9) `div` 6) . fst -- Terminate the search at the maximum value of /larger/. ) . Math.Power.squaresFrom {-hotspot-} . ceiling $ sqrt (fromIntegral i :: Double) -- Start the search at the minimum value of /larger/. {- | * Decomposes the specified integer, into a product of /prime/-factors, using , AKA . * This works best when the factors are small. -} factoriseByTrialDivision :: (Integral base, Num exponent) => base -> Data.PrimeFactors.Factors base exponent factoriseByTrialDivision = slave Data.Numbers.Primes.primes where slave primes i | null primeCandidates = [Data.Exponential.rightIdentity i] | otherwise = Data.Exponential.rightIdentity lowestPrimeFactor `Data.PrimeFactors.insert'` slave primeCandidates (i `quot` lowestPrimeFactor) where primeCandidates = dropWhile ( (/= 0) . (i `rem`) ) $ takeWhile ( <= Math.PrimeFactorisation.maxBoundPrimeFactor i ) primes lowestPrimeFactor = head primeCandidates