On 2015-04-15 05:07 AM, Jon Schneider wrote:
With lazy evaluation where is it written that if you write things with no dependencies with a "do" things will be done in order ? Or isn't it ?
Is it a feature of the language we're supposed to accept ?
It is an axiomatic feature. It is unfortunately that the Haskell Report is only half-explicit on this. But here it goes. In Chapter 7, opening: "The order of evaluation of expressions in Haskell is constrained only by data dependencies; an implementation has a great deal of freedom in choosing this order. Actions, however, must be ordered in a well-defined manner for program execution – and I/O in particular – to be meaningful. Haskell’s I/O monad provides the user with a way to specify the sequential chaining of actions, and an implementation is obliged to preserve this order." It does not say clearly how you specify an order, but it is going to be the >>= operator. For example, main = getLine >>= \_ -> putStrLn "bye" specifies to stall for your input, and then, to tell you "bye". In that order. (Perform an experiment to confirm or refute it!) * It stalls for your input, even if your input is not needed. * It tells you "bye", even if you don't need to hear it. * And it stalls for your input before outputting, not the other way round. There is no laziness or optimizer re-ordering for this. "An implementation is obliged to preserve the order." In the rest of Chapter 7, several I/O actions from the library are described. A few are decribed as "read lazily" --- these are in fact the odd men out who postpone inputting, not the common case. The common case, where it does not say "read lazily", is to grab input here-and-now and produce output here-and-now. Lastly, throughout the Haskell Report, apart from the few I/O actions that "read lazily", there is no other laziness specified. That is, lazy evaluation is *not* specified. "An implementation has a great deal of freedom in choosing this order."