On 18-Jun-08, at 5:12 PM, Dan Fabulich wrote:

Oleg Gusakov wrote:

SAT based resolver in the sandbox branch works differently from the old one, and as such - it may break a few builds that rely on bugs in the old resolver.

I think we know it will break builds that rely on bugs in the old resolver, right?


Yes, but it is also possible to house both entirely separate systems and turn on an old capability mode. It would be a huge drag to have to go futz with an existing project and we might just have to live with both mechanisms in there for a while. That's no so bad.

And we need to test is more thoroughly.

What sort of testing do you have in mind, exactly? Just general testing out in the field? Field testing won't necessarily get us that much if we know that we're going to break builds that rely on bugs.

Are there additional tests we could write today?

The next release will still use old resolver, but the intermediate, pre-SAT graph-based solution will be there as an independent component of artifact, it is used by visualization and APIs, not by the build.

Do you mean the "next release" of maven-artifact? (2.0.10? 3.0?) Or the next release of Maven? (2.0.10? 2.1?)?


2.0.x is not going to use the SAT solver. We're talking 2.1.x but even in that capacity as people transition the old behavior must be maintained or mimicked. I don't know yet how hard it would be to mimic. We were going for correct with the SAT solver.

-Dan

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


Thanks,

Jason

----------------------------------------------------------
Jason van Zyl
Founder,  Apache Maven
jason at sonatype dot com
----------------------------------------------------------

People develop abstractions by generalizing from concrete examples.
Every attempt to determine the correct abstraction on paper without
actually developing a running system is doomed to failure. No one
is that smart. A framework is a resuable design, so you develop it by
looking at the things it is supposed to be a design of. The more examples
you look at, the more general your framework will be.

  -- Ralph Johnson & Don Roberts, Patterns for Evolving Frameworks


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to