On Tue, Sep 14, 2010 at 11:29 AM, Bryan O'Sullivan
<bos@serpentine.com> wrote:
On Tue, Sep 14, 2010 at 11:21 AM, Edward Z. Yang
<ezyang@mit.edu> wrote:
Pure code can always be safely asynchronously interrupted (even code
using state like the ST monad), and IO code can be made to interact
correctly with thread termination simply by using appropriate bracketing
functions that would handle normal IO exceptions.
Ertugrul's advice is still correct. I'd wager there are very few concurrent applications that could survive a killThread without disaster. People simply don't write or test code with that in mind, and even when they do, it's more likely than not to be wrong.
I don't use killThread, and I write what I'd call somewhat complex concurrent Haskell software for a living right now :-).
Instead I have a TChan of commands that I can send to a thread, either from the outside or inside, and that thread will eventually come back to it's event loop that looks at such messages, and shut down gracefully from there.
Of course the only time this would happen is if something goes wrong and I'm going to restart and forget all the data I have accumulated thus far anyway.
_______________________________________________
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe