Text.AhoCorasick

Aho-Corasick string matching algorithm

Simplest example:

example1 = mapM_ print $ findAll simpleSM "ushers" where
    simpleSM = makeSimpleStateMachine ["he","she","his","hers"]

Position {pIndex = 1, pLength = 3, pVal = "she"}
Position {pIndex = 2, pLength = 2, pVal = "he"}
Position {pIndex = 2, pLength = 4, pVal = "hers"}

With data:

example2 = mapM_ print $ findAll sm "ushers" where
    sm = makeStateMachine [("he",0),("she",1),("his",2),("hers",3)]

Position {pIndex = 1, pLength = 3, pVal = 1}
Position {pIndex = 2, pLength = 2, pVal = 0}
Position {pIndex = 2, pLength = 4, pVal = 3}

Step-by-step state machine evaluation:

example3 = mapM_ print $ next sm "ushers" where
    sm = makeSimpleStateMachine ["he","she","his","hers"]
    next _ [] = []
    next sm (s:n) = let (SMStepRes match nextSM) = stateMachineStep sm s in
        (s, match) : next nextSM n

(u,[])
(s,[])
(h,[])
(e,[(3,"she"),(2,"he")])
(r,[])
(s,[(4,"hers")])

Documentation

makeStateMachine :: (Eq keySymb, Hashable keySymb) => [([keySymb], val)] -> StateMachine keySymb val

makeSimpleStateMachine :: (Eq keySymb, Hashable keySymb) => [[keySymb]] -> StateMachine keySymb [keySymb]

findAll :: (Eq keySymb, Hashable keySymb) => StateMachine keySymb val -> [keySymb] -> [Position val]

data Position val

Constructors

Position

Fields

pIndex :: Int

pLength :: Int

pVal :: val

Instances

Show val => Show (Position val)

stateMachineStep :: (Eq keySymb, Hashable keySymb) => StateMachine keySymb val -> keySymb -> SMStepRes keySymb val

type KeyLength = Int

data (Eq keySymb, Hashable keySymb) => SMStepRes keySymb val

Constructors

SMStepRes

Fields

smsrMatch :: [(KeyLength, val)]

smsrNextSM :: StateMachine keySymb val

Instances

(Eq keySymb, Hashable keySymb, Show keySymb, Show val) => Show (SMStepRes keySymb val)

resetStateMachine :: (Eq keySymb, Hashable keySymb) => StateMachine keySymb val -> StateMachine keySymb val