tech-toolchain archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
Re: Make and makefile bugs (PRs 49085, 49086, 49087)
On Sun, Aug 17, 2014 at 04:08:42PM +0400, Valery Ushakov wrote:
> On Sun, Aug 17, 2014 at 13:16:11 +0300, Jarmo Jaakkola wrote:
> > Do you mean that .NULL should support multiple suffixes? Sounds a bit
> > hairy.
>
> I don't understand. .NULL is a syntactic crutch that is ncessary
> since transformation rule syntax requires two suffixes. But there's
> just one empty suffix, so why would you need multiple .NULL suffixes -
> except for obvious considerations of modularity, but that's somewhat
> orthogonal, as .NULL suffix already conflicts with a real suffix.
I'd suppose that the reason would be that the fake suffix could be
different in some cases. The rule "foo" -> "foo.a" could be the same as
the rule to make "foo.b" -> "foo.a", but for "foo" -> "foo.x" it could
be "foo.y" -> "foo.x" instead of "foo.b" -> "foo.x".
Of course, in this case it would actually make sense to use ".dummy"
(or ".crap" as it so eloquently put before ;). So:
.SUFFIXES: .a .b .x .y .dummy
.b.a .dummy.a:
...
.y.x .dummy.x:
.NULL: .dummy
But hey, thanks! This gave me some ideas I'll present when I reply to
Christos' mail.
--
Jarmo Jaakkola
- References:
- Make and makefile bugs (PRs 49085, 49086, 49087)
- Re: Make and makefile bugs (PRs 49085, 49086, 49087)
- Re: Make and makefile bugs (PRs 49085, 49086, 49087)
- Re: Make and makefile bugs (PRs 49085, 49086, 49087)
- Re: Make and makefile bugs (PRs 49085, 49086, 49087)
Home |
Main Index |
Thread Index |
Old Index