[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Chicken-hackers] -block: a solution leading to more questions

From: Jörg F . Wittenberger
Subject: Re: [Chicken-hackers] -block: a solution leading to more questions
Date: 12 Apr 2013 20:05:43 +0200

On Apr 11 2013, Christian Kellermann wrote:

* Jörg F. Wittenberger <address@hidden> [130411 18:36]:
To sum up: I feel this would be a task to be performed inside
the compiler and not by some precompile script.

Feasible? Rational?  How do you guys thing ot it.

Find attached the modified code of yesterdays foobar example,
which does work.  (Including a Makefile ;-)

Are you aware of the "filename" form of the module construct?

You can have your portable code in the file as it is, then build a
module file which loads your implementation and you can specify
which identifiers should be visible to the outside world. That
sounds a lot easier to me than what you are currently doing.

For example:

File foo.scm:
(module foo (procedure1 procedure2) "foo-impl.scm")

No, that is yes: I've not been aware of this for.

I'm using (module foo (<exports>) (include "foo.imple.scm")
almost everywhere, however.

foo-impl.scm contains your original code. Or does this miss an
important point here?

Maybe: what I'm doing there is (or that is seems to be - to me)
only needed when one tries to use (block) optimization.
So what I'm doing is only to manage "export via backdoor".

I noticed however, that it have the nice side effect of
better insulation.  When I change the implementation, I
might need/want to change the import list, which would
trigger a recompile of depending modules.  That's not the
case with the binding construct.

Kind regards,



reply via email to

[Prev in Thread] Current Thread [Next in Thread]