darwin.jam vs gcc.jam

Brad Howes <howesll.mit.edu> writes:

> On Dec 11, 2006, at 8:06 AM, David Abrahams wrote:
>
>     I know you understand this Rene, but to be
completely clear for
> everyone else: I mean that we should have one top-level
"gcc" toolset
> that everyone uses even if they're targeting Darwin.
>
> Hey Dave -- I worked at Dragon Systems while you and
Tim Peters were
> there -- 

Hi Brad.

> sorry if this has already been answered, but why the
insistence that
> gcc.jam and darwin.jam merge? I thought the whole point
of the tool
> inheritance framework was to allow for this kind of
derivation?


Users of GCC shouldn't have to wonder whether they should
use the gcc
toolset (obvious; it's the name of the compiler like all the
other
toolsets) or the darwin toolset (non-obvious: it's the name
of the OS,
sorta, and it's not unlikely that we'll have toolsets for
OSes someday
-- think cross-compilation).  It's just confusing.  I don't
care how
the code is organized; this is a user-interface issue.

> I'm happily using BBv2 on a PowerBook and I would hate
to
> encounter a ton of churn to back get to where I am
today. Seems kind
> of unusual to me to include a ton of Darwin/MacOS
X-specific rules and
> actions to support bundles and frameworks in gcc.jam.

Take a good look at gcc.jam.  It's already chock full of
platform-specific rules.  Anyway, it wouldn't be hard to
allow darwin
to be used as a toolset name transparently as a
backward-compatibility
measure.

-- 
Dave Abrahams
Boost Consulting
www.boost-consulting.com

_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost-build