bloomfilter 1.0 → 1.0.1

raw patch · 5 files changed

+27/−8 lines, 5 filesPVP ok

version bump matches the API change (PVP)

API changes (from Hackage documentation)

Files

• Data/BloomFilter.hs +8/−0
• Data/BloomFilter/Easy.hs +2/−1
• bloomfilter.cabal +1/−1
• examples/SpellChecker.hs +2/−4
• examples/Words.hs +14/−2

@@ -49,6 +49,7 @@     -- ** Accessors     , lengthB     , elemB+    , notElemB      -- * Mutable Bloom filters     -- ** Creation@@ -219,6 +220,13 @@ elemB elt ub = all test (hashesU ub elt)   where test (off :* bit) = (bitArrayB ub `unsafeAt` off) .&. (1 `shiftL` bit) /= 0           +-- | Query an immutable Bloom filter for non-membership.  If the value+-- /is/ present, return @False@.  If the value is not present, there+-- is /still/ some possibility that @False@ will be returned.+notElemB :: a -> Bloom a -> Bool+notElemB elt ub = any test (hashesU ub elt)+  where test (off :* bit) = (bitArrayB ub `unsafeAt` off) .&. (1 `shiftL` bit) == 0+ -- | Create an immutable Bloom filter from a mutable one.  The mutable -- filter /must not/ be modified afterwards, or a runtime crash may -- occur.  For a safer creation interface, use 'createB'.

@@ -17,6 +17,7 @@       Bloom     , easyList     , elemB+    , notElemB     , lengthB      -- ** Example: a spell checker@@ -26,7 +27,7 @@     , suggestSizing     ) where -import Data.BloomFilter (Bloom, elemB, fromListB, lengthB)+import Data.BloomFilter (Bloom, elemB, fromListB, lengthB, notElemB) import Data.BloomFilter.Hash (Hashable, cheapHashes) import Data.BloomFilter.Util (nextPowerOfTwo) import qualified Data.ByteString as SB

@@ -1,5 +1,5 @@ name:            bloomfilter-version:         1.0+version:         1.0.1 license:         BSD3 license-file:    LICENSE author:          Bryan O'Sullivan <bos@serpentine.com>

@@ -1,7 +1,5 @@-import Data.BloomFilter.Easy (easyList, elemB)+import Data.BloomFilter.Easy (easyList, notElemB)  main = do   filt <- (easyList 0.01 . words) `fmap` readFile "/usr/share/dict/words"-  let check word | word `elemB` filt = ""-                 | otherwise         = word ++ "\n"-  interact (concat . map check . lines)+  interact (unlines . filter (`notElemB` filt) . lines)

@@ -3,11 +3,23 @@ -- queries it exhaustively.  import Control.Monad (forM_, mapM_)-import Data.BloomFilter.Easy (easyList, elemB, lengthB)+import Data.BloomFilter (Bloom, fromListB, elemB, lengthB)+import Data.BloomFilter.Hash (cheapHashes)+import Data.BloomFilter.Easy (easyList, suggestSizing) import qualified Data.ByteString.Char8 as B import Data.Time.Clock (diffUTCTime, getCurrentTime) import System.Environment (getArgs) +conservative, aggressive :: Double -> [B.ByteString] -> Bloom B.ByteString+conservative = easyList++aggressive fpr xs+    = let (size, numHashes) = suggestSizing (length xs) fpr+          k = 3+      in fromListB (cheapHashes (numHashes - k)) (size * k) xs++testFunction = aggressive+ main = do   args <- getArgs   let files | null args = ["/usr/share/dict/words"]@@ -18,7 +30,7 @@     putStrLn $ {-# SCC "words/length" #-} show (length words) ++ " words"     b <- getCurrentTime     putStrLn $ show (diffUTCTime b a) ++ "s to count words"-    let filt = {-# SCC "construct" #-} easyList 0.01 words+    let filt = {-# SCC "construct" #-} testFunction 0.01 words     print filt     c <- getCurrentTime     putStrLn $ show (diffUTCTime c b) ++ "s to construct filter"