On 20 April 2005 22:35, Scott Turner wrote:
On 2005 April 20 Wednesday 15:56, Iavor Diatchki wrote:
I've been playing around with decoding binary files, and I couldn't find a function in the libraries to read a byte from a file. Does anyone know if there is such a function? ....... it seems that we should have: hGetByte :: Handle -> IO Word8
hGetByte could be implemented on top of hGetBuf and StorableArray, but the overhead would be ugly. Recently I needed binary input for similar reasons, and would have used hGetByte had it been available. I ended up performing input using a reasonable-sized buffer.
In GHC, you can assume that hGetByte h = do c <- hGetChar h; return (fromIntegral (ord c)) you must open the file in binary mode (openBinaryFile). I don't think this works with Hugs, though. Personally, I think this ought to be part of the redesign of the I/O library. hGetByte and hPutByte would be a stopgap. But, not much progress is being made on an I/O redesign right now. Cheers, SImon
On Thu, Apr 21, 2005 at 10:32:59AM +0100, Simon Marlow wrote:
On 20 April 2005 22:35, Scott Turner wrote:
On 2005 April 20 Wednesday 15:56, Iavor Diatchki wrote:
I've been playing around with decoding binary files, and I couldn't find a function in the libraries to read a byte from a file. Does anyone know if there is such a function? ....... it seems that we should have: hGetByte :: Handle -> IO Word8
hGetByte could be implemented on top of hGetBuf and StorableArray, but the overhead would be ugly. Recently I needed binary input for similar reasons, and would have used hGetByte had it been available. I ended up performing input using a reasonable-sized buffer.
In GHC, you can assume that
hGetByte h = do c <- hGetChar h; return (fromIntegral (ord c))
you must open the file in binary mode (openBinaryFile). I don't think this works with Hugs, though.
Personally, I think this ought to be part of the redesign of the I/O library. hGetByte and hPutByte would be a stopgap. But, not much progress is being made on an I/O redesign right now.
but a useful stopgap, because their use would guarentee your code wouldn't suddenly stop working when ghc changes its IO system. I'd like to see all of: readBinaryFile :: String -> IO [Word8] writeBinaryFile :: String -> [Word8] -> IO () hGetByte :: Handle -> IO Word8 hPutByte :: Handle -> Word8 -> IO () added to all the compilers. simply so there is a straightforward (if not ideal) portable way to do binary IO. We can worry about the ideal pecfect low level interface later. John -- John Meacham - ⑆repetae.net⑆john⑈
John Meacham wrote:
Personally, I think this ought to be part of the redesign of the I/O library. hGetByte and hPutByte would be a stopgap. But, not much progress is being made on an I/O redesign right now.
but a useful stopgap, because their use would guarentee your code wouldn't suddenly stop working when ghc changes its IO system.
I'd like to see all of: readBinaryFile :: String -> IO [Word8] writeBinaryFile :: String -> [Word8] -> IO () hGetByte :: Handle -> IO Word8 hPutByte :: Handle -> Word8 -> IO ()
added to all the compilers. simply so there is a straightforward (if not ideal) portable way to do binary IO. We can worry about the ideal pecfect low level interface later.
Personally, I think that h{Get,Put}Byte should be an integral part of
the I/O interface. I don't see them as just a stopgap.
--
Glynn Clements
On Thu, Apr 21, 2005 at 10:32:59AM +0100, Simon Marlow wrote:
In GHC, you can assume that
hGetByte h = do c <- hGetChar h; return (fromIntegral (ord c))
you must open the file in binary mode (openBinaryFile). I don't think this works with Hugs, though.
It does, provided the file was opened in binary mode.
Just a thought, what about the other Simon's BlockIO library, its certainly very efficient, and probably what I am going to use for binaryIO from now on. Keean. Simon Marlow wrote:
On 20 April 2005 22:35, Scott Turner wrote:
On 2005 April 20 Wednesday 15:56, Iavor Diatchki wrote:
I've been playing around with decoding binary files, and I couldn't find a function in the libraries to read a byte from a file. Does anyone know if there is such a function? ....... it seems that we should have: hGetByte :: Handle -> IO Word8
hGetByte could be implemented on top of hGetBuf and StorableArray, but the overhead would be ugly. Recently I needed binary input for similar reasons, and would have used hGetByte had it been available. I ended up performing input using a reasonable-sized buffer.
In GHC, you can assume that
hGetByte h = do c <- hGetChar h; return (fromIntegral (ord c))
you must open the file in binary mode (openBinaryFile). I don't think this works with Hugs, though.
Personally, I think this ought to be part of the redesign of the I/O library. hGetByte and hPutByte would be a stopgap. But, not much progress is being made on an I/O redesign right now.
Cheers, SImon _______________________________________________ Libraries mailing list Libraries@haskell.org http://www.haskell.org/mailman/listinfo/libraries
Actually thats Peter Simons... (too many Simons...) http://cryp.to/blockio/ Keean. Keean Schupke wrote:
Just a thought, what about the other Simon's BlockIO library, its certainly very efficient, and probably what I am going to use for binaryIO from now on.
Keean.
Simon Marlow wrote:
On 20 April 2005 22:35, Scott Turner wrote:
On 2005 April 20 Wednesday 15:56, Iavor Diatchki wrote:
I've been playing around with decoding binary files, and I couldn't find a function in the libraries to read a byte from a file. Does anyone know if there is such a function? ....... it seems that we should have: hGetByte :: Handle -> IO Word8
hGetByte could be implemented on top of hGetBuf and StorableArray, but the overhead would be ugly. Recently I needed binary input for similar reasons, and would have used hGetByte had it been available. I ended up performing input using a reasonable-sized buffer.
In GHC, you can assume that hGetByte h = do c <- hGetChar h; return (fromIntegral (ord c))
you must open the file in binary mode (openBinaryFile). I don't think this works with Hugs, though.
Personally, I think this ought to be part of the redesign of the I/O library. hGetByte and hPutByte would be a stopgap. But, not much progress is being made on an I/O redesign right now.
Cheers, SImon _______________________________________________ Libraries mailing list Libraries@haskell.org http://www.haskell.org/mailman/listinfo/libraries
_______________________________________________ Libraries mailing list Libraries@haskell.org http://www.haskell.org/mailman/listinfo/libraries
participants (5)
-
Glynn Clements -
John Meacham -
Keean Schupke -
Ross Paterson -
Simon Marlow