>> I didn't try ghc-5.02 since I haven't installed it yet.
> Could you do that? The IO library in 5.00.2 had a bug in fdToHandle
> which meant that it didn't put the file descriptor into non-blocking
> mode, and I suspect this is the cause of your problem.
It works. Thank you.
Max.
__
> I didn't try ghc-5.02 since I haven't installed it yet.
Could you do that? The IO library in 5.00.2 had a bug in fdToHandle
which meant that it didn't put the file descriptor into non-blocking
mode, and I suspect this is the cause of your problem.
Cheers,
Simon
_
On Mon, Oct 15, 2001 at 04:47:25PM +0100, Simon Marlow wrote:
> I tried this here on Linux/x86 with 5.02 and it seems to work fine.
> Instead of the tcl/tk script you mentioned I used a FIFO in /tmp/fifo
> and made the runProcess just call "cat /tmp/fifo".
> Which GHC version and platform is this
> Playing with GHC, I met some oddity.
> Consider the example:
>
> > import IO
> > import Concurrent
> > import Posix
> > import PosixIO
> >
> > main = do
> > (fdIn, fdIn_send) <- createPipe
> > hIn_send <- fdToHandle fdIn_send
> > hIn <- fdToHandle fdIn
> > -- The "waiter"
On Thu, Oct 11, 2001 at 01:28:39PM +0200, Volker Stolz wrote:
> In local.glasgow-haskell-users, you wrote:
>>> interact hIn = do
>>> line <- hGetLine hIn
>>> numIn <- handleToInt {-made from handleToFd & fdToInt-}
>>> -- threadWaitRead numIn {-3-}
>
> Why should uncomme
I forgot to say: that's ghc-5.00.1, linux.
___
Glasgow-haskell-users mailing list
[EMAIL PROTECTED]
http://www.haskell.org/mailman/listinfo/glasgow-haskell-users
