<span class="Apple-style-span" style="border-collapse: collapse; "><div>forwarded:</div><div><br></div>Yes! if no state is passed, the optimization makes sense and the term is not executed, like any lazy evaluation. For example, I used the debugger (that is, without optimizations) to verify it with the Maybe monad:<div>
<br>op x= x+x</div><div><br>print $ Just (op 1) >>= \y-> return (Just 2)<br><br>does not evaluate op 1</div><div><br>but</div><div><br>print $ Just (op 1) >>= \y-> return y<br><br><span style="font-style: italic; ">does </span>execute it.<br>
<br><br><br>The trace of the first:<br><br>[1 of 1] Compiling Main ( test.hs, interpreted )<br>Ok, modules loaded: Main.<br>*Main> :set stop :list<br>*Main> :step main<br>Stopped at test.hs:4:6-43<br>_result :: IO () = _<br>
3<br>4 main= print $ Just (op 1) >>= \y-> return 2<br> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^<br>5<br>[test.hs:4:6-43] *Main> :step<br>Stopped at test.hs:4:14-43<br>_result :: Maybe Integer = _<br>
3<br>4 main= print $ Just (op 1) >>= \y-> return 2<br> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^<br>5<br>[test.hs:4:14-43] *Main> :step<br>Stopped at test.hs:4:14-24<br>_result :: Maybe Integer = _<br>
3<br>4 main= print $ Just (op 1) >>= \y-> return 2<br> ^^^^^^^^^^^<br>5<br>[test.hs:4:14-24] *Main> :step<br>Stopped at test.hs:4:35-43<br>_result :: Maybe Integer = _<br>3<br>4 main= print $ Just (op 1) >>= \y-> return 2<br>
^^^^^^^^^<br>5<br>[test.hs:4:35-43] *Main> :step<br>Just 2<br><br><br><span style="font-style: italic; ">But in the second case op is executed:<br></span><br><br>*Main> :step main<br>
Stopped at test.hs:4:6-43<br>_result :: IO () = _<br>3<br>4 main= print $ Just (op 1) >>= \y-> return y<br> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^<br>5<br>[test.hs:4:6-43] *Main> :step<br>Stopped at test.hs:4:14-43<br>
_result :: Maybe Integer = _<br>3<br>4 main= print $ Just (op 1) >>= \y-> return y<br> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^<br>5<br>[test.hs:4:14-43] *Main> :step<br>Stopped at test.hs:4:14-24<br>
_result :: Maybe Integer = _<br>3<br>4 main= print $ Just (op 1) >>= \y-> return y<br> ^^^^^^^^^^^<br>5<br>[test.hs:4:14-24] *Main> :step<br>Stopped at test.hs:4:35-43<br>_result :: Maybe Integer = _<br>
y :: Integer = _<br>3<br>4 main= print $ Just (op 1) >>= \y-> return y<br> ^^^^^^^^^<br>5<br>[test.hs:4:35-43] *Main> :step<br>Just Stopped at test.hs:4:20-23<br>
_result :: Integer = _<br>3<br>4 main= print $ Just (op 1) >>= \y-> return y<br> ^^^^<br>5<br>[test.hs:4:20-23] *Main> :step<br>Stopped at test.hs:6:0-8<br>
_result :: Integer = _<br>5<br>6 <span style="font-weight: bold; "> op x= x+x</span><br> ^^^^^^^^^<br>7<br>[test.hs:6:0-8] *Main> :step<br>Stopped at test.hs:6:6-8<br>_result :: Integer = _<br>x :: Integer = _<br>5<br>
6 <span style="font-weight: bold; ">o</span><span style="font-weight: bold; ">p x= x+x</span><br> ^^^<br>7<br>[test.hs:6:6-8] *Main> :step<br>Just 2<br><br><br><br><br>2009/2/5 Gregg Reynolds <<a href="mailto:dev@mobileink.com" target="_blank" style="color: rgb(42, 93, 176); ">dev@mobileink.com</a>><div>
<div><span id="q_11f60636b3e3ee54_1" class="WQ9l9c" style="cursor: pointer; color: rgb(80, 0, 80); font-size: 9px; ">- Ocultar texto citado -</span></div><div class="Ih2E3d" style="color: rgb(80, 0, 80); "><br>><br>> On Thu, Feb 5, 2009 at 9:27 AM, Bulat Ziganshin <<a href="mailto:bulat.ziganshin@gmail.com" target="_blank" style="color: rgb(42, 93, 176); ">bulat.ziganshin@gmail.com</a>> wrote:<br>
>><br>>> Hello Gregg,<br>>><br>>> Thursday, February 5, 2009, 6:20:06 PM, you wrote:<br>>><br>>> > An optimizer can see that the result of the first getChar is<br>>> > discarded and replace the entire expression with one getChar without<br>
>> > changing the formal semantics. <br>>><br>>> this is prohibited by using pseudo-value of type RealWorld which is<br>>> passed through entire action stream. actually, order of execution is<br>
>> controlled by dependencies on this values<br>>><br>>> <a href="http://haskell.org/haskellwiki/IO_inside" target="_blank" style="color: rgb(42, 93, 176); ">http://haskell.org/haskellwiki/IO_inside</a><br>
>><br>> Thanks. I actually read that a few weeks ago and forgot all about it. So the gist is that type IO has special magic semantics. Formal, but hidden. Which means monad semantics are built in to the language, at least for that type. The Haskell Report doesn't seem to say anything about this, which seems odd. <br>
><br>> But then for non-IO monads, the optimization would be allowed, no? Of course; only the IO monad has external world behavior.<br>><br>> Thanks,<br>><br>> gregg<br>><br></div></div><div class="Ih2E3d" style="color: rgb(80, 0, 80); ">
> _______________________________________________<br>> Haskell-Cafe mailing list<br>> <a href="mailto:Haskell-Cafe@haskell.org" target="_blank" style="color: rgb(42, 93, 176); ">Haskell-Cafe@haskell.org</a><br>> <a href="http://www.haskell.org/mailman/listinfo/haskell-cafe" target="_blank" style="color: rgb(42, 93, 176); ">http://www.haskell.org/mailman/listinfo/haskell-cafe</a><br>
><br></div></div></span>