Boris Kolpackov wrote:
> > The meat of make is rule-based so I would think that replacing this with
> > a functional programming language would not be a good thing.
>
> I noticed that as a make-based build systems gets more elaborate
> the more and more code in makefiles is procedural rather than
> declarative (rules). Though I believe rules are the most important
> property of make.
I agree (although I'd say "imperitive" rather than "procedural").
> When I contemplated make-like inference system in lisp I didn't say
> I suggest to throw away rules. We should still have them:
>
> hello : hello.c
> gcc -o $@ $<
>
> (defrule ("hello") ("hello.c")
> "gcc -o $@ $<")
>
> Not as succinct but way more powerful.
I actually like the rule syntax the way they are. It's very Prolog-like which may not
be by coincidence.
I suppose all this may not matter depending upon how one language is embedded within
the other. One thing I really hate about the existing macro language is its behaviour
with regards to line endings and white space. I'm hoping the planned embedded Guile
will do away with this limitation.
Noel
--
NOTICE: If received in error, please destroy and notify sender. Sender does not waive
confidentiality or privilege, and use is prohibited.
_______________________________________________
Help-make mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/help-make
