在 GHC 中编译

Question

I'm learning Haskell and a simple doubt struck me, what is the difference between these two commands to compile using GHC (fac is a program to calculate factorial) :

ghc -o fac fac.hs

and

ghc fac.hs

Now to run the program I just do fac.exe (I'm using Windows).

Given that both works which one is right? both? And which one is the most advisable to use?

Answer 1

Both are OK. The only thing the -o command does is determine the name of the executable file that is generated. As far as the Haskell goes it doesn't make any difference.

Personally, I like always using "-o" to make things explicit (since I like avoiding implicit defaults in general) and to make it look more similar to other compilers (for example, the gcc C compiler kind of forces you to use its "-o" option, since the default name it uses otherwise is very ugly)

Answer 2

It used to be that writing

ghc fac.hs

didn't really work. Instead you had to say

ghc --make fac.hs

The difference is that the former compiles fac.hs and does nothing else, while the latter checks to see if fac.hs uses any other modules that need recompiling, automatically includes any packages required, and so on and so forth.

However, the default behavior has been changed, and the above two commands are now identical. (I forget exactly when this happened, but it was a while ago now.) You may still see documentation and examples with the explicit --make switch though, and personally I like to include it. But that's just me.

The only thing the -o switch gives you is the ability to change the output name. This can be useful if you have, say, Main.hs and you want the output to be called MyCoolProgram or whatever. One thing I will say: under Linux, fac.hs (if it's the Main module) will compile to an executable named fac , but under Windows it will be called fac.exe . If you use -o to hard-code the binary name, this won't happen. And I'm not sure you can run a Windows binary that doesn't end .exe ...