[Haskell-cafe] Billion-triple RDF store

[Bulat Ziganshin]

Hello Joel,

Monday, April 24, 2006, 5:50:57 AM, you wrote:

> For example, you could use tries to store all strings but you could
> easily get to the point where you can't load them all into, say, 1Gb  
> of memory which is all I have on my machine. I'm thinking that the  
> required data could be pulled from disk lazily (as needed) and  
> laziness is a Haskell forte.

this should be rather easy to implement:

type Key = String
type Value = String

-- | Opens a file with given name and returns function that maps keys to values
open :: String -> IO (Key -> Value)
open filename = do h <- openBinaryFile filename ReadMode
                   let lookup key = unsafePerformIO $ do
                         -- your genuine algorithm in IO monad that
                         -- finds value for `key` and returns it
                         return value
                   return lookup

main = do lookup <- open "database"
          print (lookup "key1", lookup "key2")

-- 
Best regards,
 Bulat                            mailto:Bulat.Ziganshin at gmail.com