[Haskell-cafe] Re: Stream processors

[Peter Simons]

Ben Rudiak-Gould writes:

 > Must contexts be used in a single-threaded manner? If so,
 > I would expect this interface:

 >     start  :: IO ctx
 >     feed   :: ctx -> Buffer -> IO ()
 >     commit :: ctx -> IO a

'feed' cannot have this signature because it needs to update
the context. 


 > If not, I would expect this interface:

 >     start  :: ctx
 >     feed   :: ctx -> Buffer -> IO ctx
 >     commit :: ctx -> a

Both 'start' and 'commit' need to be in the IO monad,
because creating and finalizing the context may involve IO
calls. (Just think of a computation that does internal
buffering in memory which is accessed through another Ptr.)


 > Additionally, I don't think (Ptr Word8, Int) is general
 > enough for all reasonable uses of this interface.

That's true. I used this because I have the data in a memory
buffer already, so this is the API with the best performance
for me because it handles the input without marshaling. I
agree it would be nice to have something more general and
generally smarter than the my type definition -- that's why
I asked on the list. :-)


 >     feedBuffer   :: ctx -> Buffer               -> IO ctx
 >     feedSTUArray :: ctx -> STUArray s Int Word8 -> ST s ctx
 >     feedUArray   :: ctx -> UArray Int Word8     -> ctx

I would implement feedSTUArray and friends as wrappers
around the Ptr interface, not as primitive computations of
the stream processor. But nonetheless, having those would be
nice.  

Peter