# Prime numbers

### From HaskellWiki

m |
m |

## Revision as of 07:07, 28 December 2009

## Contents |

## 1 Prime Number Resources

In mathematics, a *prime number* (or a *prime*) is a natural number which has exactly two distinct natural number divisors: 1 and itself. The smallest prime is thus 2.

Prime Numbers at Wikipedia.

Sieve of Eratosthenes at Wikipedia.

HackageDB packages:

Numbers: An assortment of number theoretic functions.

NumberSieves: Number Theoretic Sieves: primes, factorization, and Euler's Totient.

primes: Efficient, purely functional generation of prime numbers.

## 2 Finding Primes

Any natural number is representable as a product of powers of its prime factors, and so a prime has no prime divisors other then itself. That means that starting with 2, *for each* newly found *prime* we can *eliminate* from the rest of the numbers *all such numbers* that have this prime as their divisor, giving us the *next available* number as next prime. This is known as *sieving* the natural numbers, so that in the end what we are left with are just primes.

### 2.1 The Classic Simple Primes Sieve

Attributed to David Turner *(SASL Language Manual, 1975)*, the following is a direct translation of that idea, generating a list of all prime numbers:

primes :: [Integer] primes = sieve [2..] where sieve (p:xs) = p : sieve [x | x<-xs, x `mod` p /= 0] -- or: filter ((/=0).(`mod`p)) xs

This should only be considered a *specification*, not a *code*. When run as is, it is *extremely inefficient* because it starts up the filters prematurely, immediately after each prime, instead of only after the prime's square has been reached. To be admitted as prime, *each number* will be *tested for divisibility* here by all its preceding primes, while just those not greater than its square root would suffice. This means that e.g. to find the **1001**st prime (`7927`

), **1000** filters are used, when in fact just the first **24** are needed (upto `89`

's filter only).

So this in effect creates a cascade of nested filters in front of the infinite numbers supply, and in *extremely premature* fashion at that. One way of fixing that would be to *postpone* the creation of filters until the right moment, by decoupling the primes supply from the numbers supply, as in

### 2.2 Postponed Filters Sieve

primes :: [Integer] primes = 2: 3: sieve (tail primes) [5,7..] where sieve (p:ps) xs = h ++ sieve ps [x | x<-t, x `rem` p /= 0] -- or: filter ((/=0).(`rem`p)) t where (h,~(_:t)) = span (< p*p) xs

This can be seen as essential framework for all the code to come. It only tests odd numbers, and only by the primes that are needed, for each *numbers span* between successive squares of primes. To find the **1001**st prime, the divisibility test is performed by only **24** nested filters corresponding to the first **24** odd primes.

Whereas the first version exhibits near O(*n*^{2}) behavior, this one exhibits near O(*n*^{1.5}) behavior, with an orders-of-magnitude speedup.

There is another way for the composites to be found - by generating all the multiples of successive primes, in advance. Any number thus generated will of course be divisible by the corresponding prime.

#### 2.2.1 Postponed Multiples Removal i.e. Euler's Sieve

Instead of testing *each number* for divisibility by a prime we can just *remove* the prime's *multiples* in advance. We gain in speed because we now get the primes *for free*, after all the multiples are removed on a particular span, *without* performing any divisibility tests *at all*:

primes :: [Integer] primes = 2: 3: sieve (tail primes) [5,7..] where sieve (p:ps) xs = h ++ sieve ps (t `minus` tail [q,q+2*p..]) where (h,~(_:t)) = span (< q) xs q = p*p minus :: (Ord a) => [a] -> [a] -> [a] minus a@(x:xs) b@(y:ys) = case compare x y of LT -> x: xs `minus` b EQ -> xs `minus` ys GT -> a `minus` ys minus a b = a

This is, in fact, Euler's sieve.

#### 2.2.2 Merged Multiples Removal Sieve

`(...((s-a)-b)-...)`

is the same as `(s-(a+b+...))`

, and so we can just remove the *merged* infinite primes multiples, each starting at its prime's square, from the *initial* numbers supply. This way we needn't explicitly jump to a prime's square because it's where its multiples start anyway:

primes :: [Integer] primes = 2:primes' where primes' = [3] ++ [5,7..] `minus` foldr merge' [] mults mults = map (\p->let q=p*p in (q,tail [q,q+2*p..])) $ primes' merge' (q,qs) xs = q : merge qs xs merge :: (Ord a) => [a] -> [a] -> [a] merge a@(x:xs) b@(y:ys) = case compare x y of LT -> x: merge xs b EQ -> x: merge xs ys GT -> y: merge a ys merge a b = if null a then b else a

This code is yet faster. Its main deficiency still is that it creates a linear nested merging structure, instead of a tree-like structure. Each multiple produced by a prime has to percolate to the top eventually, so it's better to make its path shorter. It'll have to go through fewer merge nodes this way.

The linearity is imposed by type asymmetry of our `(merge' :: a -> b -> b)`

function, forcing us into the `1+(1+(1+(1+...)))`

pattern, `+`

standing for `merge'`

(which was defined that way to prevent the run-ahead when folding over the infinite list of lists of multiples).

We need to turn it into an associative operation of uniform type `(:: a -> a -> a)`

to be able to freely rearrange the combinations into arbitrary tree-like patterns, as in e.g. ((1+1)+(2+2))+(...) etc. The type uniformity is what makes compositionality possible.

The code in the "Implicit Heap" section below improves on that, and is essentially equivalent to using a treefold instead of a standard linear `foldr`

, as in:

#### 2.2.3 Treefold Merged Multiples Removal

primes :: [Integer] primes = 2:primes' where primes' = [3,5] ++ drop 2 [3,5..] `minus` comps mults = map (\p-> let q=p*p in ([q],tail [q,q+2*p..])) $ primes' comps = fst $ tfold mergeSP (pairwise mergeSP mults) pairwise f (x:y:ys) = f x y : pairwise f ys tfold f (a: ~(b: ~(c:xs))) = (a `f` (b `f` c)) `f` tfold f (pairwise f xs) mergeSP (a,b) ~(c,d) = let (bc,b') = spMerge b c in (a ++ bc, merge b' d) where spMerge :: (Ord a) => [a] -> [a] -> ([a],[a]) spMerge a@(x:xs) b@(y:ys) = case compare x y of LT -> (x:c,d) where (c,d) = spMerge xs b EQ -> (x:c,d) where (c,d) = spMerge xs ys GT -> (y:c,d) where (c,d) = spMerge a ys spMerge a [] = ([] ,a) spMerge [] b = ([] ,b)

The fold used here creates a `(2+(2+2))+( (4+(4+4)) + ( (8+(8+8)) + ... ))`

structure, better adjusted for primes multiples production than `1+(2+(4+(8+...)))`

, used by the "Implicit Heap" code, giving it additional 10% speedup.

` mergeSP `

is an associative operation, preserving of the invariant such that for a list of multiples `[(a,b),(c,d), ...]`

, it's always the case that ` last a < head b && last a < head c`

. These "split pairs" represent ordered list as a pair of its known and (currently) finite prefix, and the rest of it. Such pairs under `mergeSP`

operation form a *monoid*, and if we were to declare a

`mergeSP`

its `tfold`

its `comps`

would just become This code exhibits approximately O(*n*^{1.20})..O(*n*^{1.15}) local asymptotic behavior (tested interpreted, in GHCi, for 10,000 to 300,000 primes produced). When compared with Melissa O'Neill's PQ code from the ZIP package which was modified to work on odds only as well, it is 3.2x times faster, with used memory reported about 2.5x times smaller.

It can be further improved with the wheel optimization (as described below), giving it a further performance boost of about 30%-35%:

#### 2.2.4 Treefold Merged Multiples, with Wheel

primes :: [Integer] primes = 2:3:5:7:primes' where primes' = [11,13] ++ drop 2 (wheelSupply ()) `minus` comps mults = map (\p-> let q=p*p in ([q],tail [q,q+2*p..])) $ primes' comps = fst $ tfold mergeSP (pairwise mergeSP mults) wheel = 2:4:2:4:6:2:6:4:2:4:6:6:2:6:4:2:6:4:6:8:4:2:4:2: 4:8:6:4:6:2:4:6:2:6:6:4:2:4:6:2:6:4:2:4:2:10:2:10:wheel roll n (x:wh) = n : roll (n+x) wh wheelSupply () = roll 11 wheel

This runs at about 20% faster than the Priority Queue-based code as present in Melissa O'Neill's ZIP package, with *exactly* the same local asymptotic behavior of about O(*n*^{1.22})..O(*n*^{1.18}) (tested interpreted, in GHCi, for 10,000 to 300,000 primes produced), with used memory reported about the same.

### 2.3 More Filtering Sieves

The primes list creation with divisibility testing can be reformulated in a few more ways, using the list of primes *as it is being built* (a la "circular programming").

#### 2.3.1 Odd numbers, by Trial Division

This is also good for generating a few 100,000s primes (when GHC-compiled as well):

primes :: [Integer] primes = 2: 3: filter isPrime [5,7..] where isPrime n = all (notDivs n) $ takeWhile (\p-> p*p <= n) (tail primes) notDivs n p = n `mod` p /= 0

Instead of relying on nested filters, it tests each odd number by an explicit list of all the needed prime factors. But for each number tested it re-fetches this list *anew* which will be *the same* for the increasing spans of numbers between the successive squares of primes.

#### 2.3.2 Generated Spans, by Nested Filters

The other way to go instead of concentrating on the numbers supply, is to directly work on the successive spans between the primes squares.

This version is a bit faster still, creating *158,000 primes* (again, GHC-compiled) in the same time as the postponed filters does 100,000 primes:

primes :: [Integer] primes = 2: 3: sieve [] (tail primes) 5 where notDivsBy d n = n `mod` d /= 0 sieve ds (p:ps) x = foldr (filter . notDivsBy) [x, x+2..p*p-2] ds ++ sieve (p:ds) ps (p*p+2)

This one explicitly maintains the list of primes needed for testing each odds span between successive primes squares, which it also explicitly generates. But it tests with nested `filter`

s, which it repeatedly recreates.

#### 2.3.3 Generated Spans, by List of Primes

The list of primes needed to test each range of odds is actually just the prefix of the primes list itself, of known length, and need not be specifically generated at all. Combined with one-call testing by the explicit list of primes, and direct generation of odds between the successive primes squares, this leads to:

primes :: [Integer] primes = 2: 3: sieve 0 (tail primes) 5 sieve k (p:ps) x = [x | x<-[x,x+2..p*p-2], and [x`rem`p/=0 | p<-fs]] -- or: all ((>0).(x`rem`)) fs ++ sieve (k+1) ps (p*p+2) where fs = take k (tail primes)

It produces about *222,000 primes* in the same amount of time, and is good for creating about a million first primes, compiled.

The reason to have `sieve`

function available separately too is that it can also be used to produce primes above a given number, as in

primesFrom m = sieve (length h) ps $ m`div`2*2+1 where (h,(_:ps)) = span (<= (floor.sqrt.fromIntegral) m) primes

It can thus produce a few primes e.g. above `239812076741689`

, which is a square of the millionth odd prime, without having to compute all the preceding primes (which would number in trillions).

#### 2.3.4 Multiples Removal on Generated Spans, or Sieve of Eratosthenes

The divisibility testing too should be considered a specification (as in "no multiples of p"), and not a code per se, because although testing composites is cheap (as most of them will have small factors, so the test is soon aborted), testing prime numbers is costly, and is to be avoided.

All the filtering versions thus far try to *keep the primes* among all numbers by testing *each number* in isolation. Instead, all the relevant primes' multiples can be removed from the corresponding segments of odds, and what's left after that will be just primes:

primes :: [Integer] primes = 2: 3: sieve [] (tail primes) 5 where sieve fs (p:ps) x = [x,x+2..q-2] `minus` foldl merge [] mults ++ sieve ((2*p,q):fs') ps (q+2) where q = p*p mults = [ [y+s,y+2*s..q] | (s,y)<- fs] fs' = [ (s,last ms) | ((s,_),ms)<- zip fs mults]

This modifies the preceding sieve to "mark" the odd composites in a given range (instead of testing their divisibility) by generating them - just as a person performing the original sieve of Eratosthenes would do, marking one by one the multiples of the relevant primes.

Compared with the previous version, interpreted under both GHCi and WinHugs, it runs *faster*, takes *less* memory, and has better asymptotic behavior, its performance approximately the same as in the Merged Multiples Removal sieve. The advantage in working with spans explicitly is that this code is easily amendable to using arrays for the composites marking and removal on each *finite* span.

### 2.4 Using Arrays

#### 2.4.1 Generating Segments of Primes

The removal of multiples on each segment of odds in the sieve of Eratosthenes can be done by actually marking them in an array, instead of manipulating lists with "minus" and "merge":

primes :: [Integer] primes = 2: 3: sieve [] (tail primes) 5 where sieve fs (p:ps) x = [i | i<- [x,x+2..q-2], a!i] ++ sieve ((2*p,q):fs') ps (q+2) where q = p*p mults = [ [y+s,y+2*s..q] | (s,y)<- fs] fs' = [ (s,last ms) | ((s,_),ms)<- zip fs mults] a = accumArray (\a b->False) True (x,q-2) [(i,()) | ms<- mults, i<- ms]

Apparently, arrays are * very* fast. This code runs faster than the Treefold Merged with Wheel version (itself 20% faster than Melissa O'Neill's PQ version), by 2.0x for 10,000, 2.25x for 100,000 and 2.4x times faster for 300,000 primes produced (testing it interpreted in GHCi), with about twice less memory usage reported.

#### 2.4.2 Calculating Primes Upto a Given Value

primesToN n = 2: [i | i<-[3,5..n], ar!i] where ar = f 5 $ accumArray (\a b->False) True (3,n) [(i,()) | i<- [9,15..n]] f p a | q>=n = a | True = if null x then a' else f (head x) a' where q = p*p a'= a//[(i,False)|i<-[q,q+2*p..n]] x = [i | i<-[p+2,p+4..n], a' !i]

#### 2.4.3 Calculating Primes in a Given Range

primesFromTo a b = (if a<3 then [2] else []) ++ [i | i<-[o,o+2..b], ar!i] where o = max (if even a then a+1 else a) 3 r = (1+).floor.sqrt.fromInteger $ b ar = accumArray (\a b->False) True (o,b) [(i,()) | p <- [3,5..r] , let q = p*p s = 2*p (n,x) = quotRem (o-q) s q' = if o <= q then q else if x==0 then q+n*s else q+(n+1)*s , i<- [q',q'+s..b] ]

Although testing by odds instead of by primes, the array generation is so fast that it is very much feasible and even preferable for quick generation of some short spans of relatively big primes.

### 2.5 Implicit Heap

The following is a more efficient prime generator, implementing the sieve of Eratosthenes.

See also the message threads Re: "no-coding" functional data structures via lazyness for more about how merging ordered lists amounts to creating an implicit heap and Re: Code and Perf. Data for Prime Finders for an explanation of thedata People a = VIP a (People a) | Crowd [a] mergeP :: Ord a => People a -> People a -> People a mergeP (VIP x xt) ys = VIP x $ mergeP xt ys mergeP (Crowd xs) (Crowd ys) = Crowd $ merge xs ys mergeP xs@(Crowd (x:xt)) ys@(VIP y yt) = case compare x y of LT -> VIP x $ mergeP (Crowd xt) ys EQ -> VIP x $ mergeP (Crowd xt) yt GT -> VIP y $ mergeP xs yt merge :: Ord a => [a] -> [a] -> [a] merge xs@(x:xt) ys@(y:yt) = case compare x y of LT -> x : merge xt ys EQ -> x : merge xt yt GT -> y : merge xs yt diff xs@(x:xt) ys@(y:yt) = case compare x y of LT -> x : diff xt ys EQ -> diff xt yt GT -> diff xs yt foldTree :: (a -> a -> a) -> [a] -> a foldTree f ~(x:xs) = x `f` foldTree f (pairs xs) where pairs ~(x: ~(y:ys)) = f x y : pairs ys primes, nonprimes :: [Integer] primes = 2:3:diff [5,7..] nonprimes nonprimes = serve . foldTree mergeP . map multiples $ tail primes where multiples p = vip [p*p,p*p+2*p..] vip (x:xs) = VIP x $ Crowd xs serve (VIP x xs) = x:serve xs serve (Crowd xs) = xs

### 2.6 Prime Wheels

The idea of only testing odd numbers can be extended further. For instance, it is a useful fact that every prime number other than 2 and 3 must be of the form 6*k* + 1 or 6*k* + 5. Thus, we only need to test these numbers:

primes :: [Integer] primes = 2:3:primes' where 1:p:candidates = [6*k+r | k <- [0..], r <- [1,5]] primes' = p : filter isPrime candidates isPrime n = all (not . divides n) $ takeWhile (\p -> p*p <= n) primes' divides n p = n `mod` p == 0

Such a scheme to generate candidate numbers first that avoid a given set of primes as divisors is called a **prime wheel**. Imagine that you had a wheel of circumference 6 to be rolled along the number line. With spikes positioned 1 and 5 units around the circumference, rolling the wheel will prick holes exactly in those positions on the line whose numbers are not divisible by 2 and 3.

A wheel can be represented by its circumference and the spiked positions.

data Wheel = Wheel Integer [Integer]

We prick out numbers by rolling the wheel.

roll (Wheel n rs) = [n*k+r | k <- [0..], r <- rs]

The smallest wheel is the unit wheel with one spike, it will prick out every number.

w0 = Wheel 1 [1]

nextSize (Wheel n rs) p = Wheel (p*n) [r' | k <- [0..(p-1)], r <- rs, let r' = n*k+r, r' `mod` p /= 0]

mkWheel ds = foldl nextSize w0 ds

Now, we can generate prime numbers with a wheel that for instance avoids all multiples of 2, 3, 5 and 7.

primes :: [Integer] primes = small ++ large where 1:p:candidates = roll $ mkWheel small small = [2,3,5,7] large = p : filter isPrime candidates isPrime n = all (not . divides n) $ takeWhile (\p -> p*p <= n) large divides n p = n `mod` p == 0

It's a pretty big wheel with a circumference of 210 and allows us to calculate the first 10000 primes in convenient time.

A fixed size wheel is fine, but how about adapting the wheel size while generating prime numbers? See the functional pearl titled Lazy wheel sieves and spirals of primes for more.

### 2.7 Bitwise prime sieve

Count the number of prime below a given 'n'. Shows fast bitwise arrays, and an example of Template Haskell to defeat your enemies.

{-# OPTIONS -O2 -optc-O -XBangPatterns #-} module Primes (nthPrime) where import Control.Monad.ST import Data.Array.ST import Data.Array.Base import System import Control.Monad import Data.Bits nthPrime :: Int -> Int nthPrime n = runST (sieve n) sieve n = do a <- newArray (3,n) True :: ST s (STUArray s Int Bool) let cutoff = truncate (sqrt $ fromIntegral n) + 1 go a n cutoff 3 1 go !a !m cutoff !n !c | n >= m = return c | otherwise = do e <- unsafeRead a n if e then if n < cutoff then let loop !j | j < m = do x <- unsafeRead a j when x $ unsafeWrite a j False loop (j+n) | otherwise = go a m cutoff (n+2) (c+1) in loop ( if n < 46340 then n * n else n `shiftL` 1) else go a m cutoff (n+2) (c+1) else go a m cutoff (n+2) c

And places in a module:

{-# OPTIONS -fth #-} import Primes main = print $( let x = nthPrime 10000000 in [| x |] )

Run as:

$ ghc --make -o primes Main.hs $ time ./primes 664579 ./primes 0.00s user 0.01s system 228% cpu 0.003 total

### 2.8 Using IntSet for a traditional sieve

module Sieve where import qualified Data.IntSet as I -- findNext - finds the next member of an IntSet. findNext c is | I.member c is = c | c > I.findMax is = error "Ooops. No next number in set." | otherwise = findNext (c+1) is -- mark - delete all multiples of n from n*n to the end of the set mark n is = is I.\\ (I.fromAscList (takeWhile (<=end) (map (n*) [n..]))) where end = I.findMax is -- primes - gives all primes up to n primes n = worker 2 (I.fromAscList [2..n]) where worker x is | (x*x) > n = is | otherwise = worker (findNext (x+1) is) (mark x is)

## 3 Testing Primality

### 3.1 Primality Test and Integer Factorization

Given an infinite list of prime numbers, we can implement primality tests and integer factorization:

isPrime n = n > 1 && n == head (primeFactors n) primeFactors 1 = [] primeFactors n = go n primes where go n ps@(p:pt) | p*p > n = [n] | n `rem` p == 0 = p : go (n `quot` p) ps | otherwise = go n pt

### 3.2 Miller-Rabin Primality Test

find2km :: Integral a => a -> (a,a) find2km n = f 0 n where f k m | r == 1 = (k,m) | otherwise = f (k+1) q where (q,r) = quotRem m 2 millerRabinPrimality :: Integer -> Integer -> Bool millerRabinPrimality n a | a <= 1 || a >= n-1 = error $ "millerRabinPrimality: a out of range (" ++ show a ++ " for "++ show n ++ ")" | n < 2 = False | even n = False | b0 == 1 || b0 == n' = True | otherwise = iter (tail b) where n' = n-1 (k,m) = find2km n' b0 = powMod n a m b = take (fromIntegral k) $ iterate (squareMod n) b0 iter [] = False iter (x:xs) | x == 1 = False | x == n' = True | otherwise = iter xs pow' :: (Num a, Integral b) => (a -> a -> a) -> (a -> a) -> a -> b -> a pow' _ _ _ 0 = 1 pow' mul sq x' n' = f x' n' 1 where f x n y | n == 1 = x `mul` y | r == 0 = f x2 q y | otherwise = f x2 q (x `mul` y) where (q,r) = quotRem n 2 x2 = sq x mulMod :: Integral a => a -> a -> a -> a mulMod a b c = (b * c) `mod` a squareMod :: Integral a => a -> a -> a squareMod a b = (b * b) `rem` a powMod :: Integral a => a -> a -> a -> a powMod m = pow' (mulMod m) (squareMod m)

## 4 External links

- A collection of prime generators; the file "ONeillPrimes.hs" contains one of the fastest pure-Haskell prime generators. WARNING: Don't use the priority queue from that file for your projects: it's broken and works for primes only by a lucky chance.