Re: [jamboost] Boost.Build, legacy changes...

From: Rene Rivera (grafik666_at_[hidden])
Date: 2002-02-10 23:29:53

On 2002-02-10 at 10:19 PM, david.abrahams_at_[hidden] (David Abrahams) wrote:

>
>----- Original Message -----
>From: "Rene Rivera" <grafik666_at_[hidden]>
>
>> Ahh, I'm there with you, what I did is even more flexible :-) I use a
>global
>> variable to map from the target type and the dependency type:
>>
>> gTARGET_TYPE_ID(pyd) = PYD ; # similar for others
>
>A bit more work for new target types, but no worries, this branch of the
>code is dead anyway ;^). In future we'll specify these things with rule
>invocations (and some sort of variables only under-the-covers).

I would hope that we don't use variables at all in the new system :-) It would
make more sense to make the target types objects. But that supposses that we
use the object system on V2.

>> >P.S. How does the idea of implementing the initialization and PWD
>features
>> >for V2 sit with you?
>>
>> I was expecting to :-] But I'm still working on the 2.4 Perforce merge.
>
>How's it going? Need help? If you want to check in an intermediate state on
>a branch, that might be a good idea.

Slowly... I'll create a branch for it soon, end of this week, maybe. So yes
help will be welcome at that point ;-)

>> And I
>> have to think more about the difficulties that Vladimir brought with the
>crawl
>> up behaviour.
>
>I wasn't really clear on what the problem was. Wasn't he just contending
>that crawl-up should use GLOB to look for the files? I was expecting that
>anyway.

Nah, that's the easy part :-) The hard question is what happens when we
include a Jamfile from an "external" project. Do we search for the
project-root.jam again? And if so we have to do the search relative to the
Jamfile we are including. And I'm sure there are other problems I can't think
of right now that "external" inclusion brings up.

-- grafik - Don't Assume Anything
-- rrivera_at_[hidden] - grafik_at_[hidden]
-- 102708583_at_icq - Grafik666_at_AIM - Grafik_at_[hidden]