Question

data Thing = Thing {a :: Int, b :: Int, c :: Int, (...) , z :: Int} deriving Show

foo = Thing 1 2 3 4 5 (...) 26
mkBar x = x { c = 30 }

main = do print $ mkBar foo

What is copied over when I mutate foo in this way? As opposed to mutating part of a structure directly.

Data Thing = Thing {a :: IORef Int, b :: IORef Int, (...) , z :: IORef Int}
instance Show Thing where
(...something something unsafePerformIO...)

mkFoo = do a <- newIORef 1
           (...)
           z <- newIORef 26
           return Thing a b (...) z
mkBar x = writeIORef (c x) 30

main = do foo <- mkFoo
          mkBar foo
          print foo

Does compiling with optimizations change this behavior?

Answer 1

In the first example, the pointers to the unchanged Int components are copied (and the constructor tag, if you wish to say so). It doesn’t make much difference whether an Int or a pointer to one is copied, but if the components were large structures, it would.

Since the fields are not strict, the behaviour is, afaik, independent of optimisation. If the fields were strict, with optimisations, they might be unpacked into the constructor and then the raw Int# values would be copied.

In the second example, nothing is copied, the contents of the IORef are overwritten.