[Haskell-cafe] Design of a DSL in Haskell

[Joerg Fritsch]

Hi Tillmann,

is a shallow embedded DSL == an internal DSL and a deeply embedded DSL == an external DSL or the other way around?

--Joerg

On Dec 3, 2012, at 11:40 PM, Tillmann Rendel wrote:

> Hi,
> 
> Joerg Fritsch wrote:
>> I am working on a DSL that eventuyally would allow me to say:
>> 
>> import language.cwmwl
>> 
>> main = runCWMWL $ do
>> 
>>     eval ("isFib::", 1000, ?BOOL)
>> 
>> 
>> I have just started to work on the interpreter-function runCWMWL and I
>> wonder whether it is possible to escape to real Haskell somehow (and
>> how?) either inside ot outside the do-block.
> 
> You can already use Haskell in your DSL. A simple example:
> 
>  main = runCWMWL $ do
>    eval ("isFib::", 500 + 500, ?BOOL)
> 
> The (+) operator is taken from Haskell, and it is available in your DSL program. This use of Haskell is completely for free: You don't have to do anything special with your DSL implementation to support it. I consider this the main benefit of internal vs. external DSLs.
> 
> 
> A more complex example:
> 
>  main = runCWMWL $ do
>    foo <- eval ("isFib::", 1000, ?BOOL)
>    if foo
>      then return 27
>      else return 42
> 
> Here, you are using the Haskell if-then-else expression to decide which DSL program to run. Note that this example also uses (>>=) and return, so it only works because your DSL is monadic. Beyond writing the Monad instance, you don't have to do anything special to support this. In particular, you might not need an additional embed function if you've already implemented return from the Monad type class. I consider this the main benefit of the Monad type class.
> 
>  Tillmann