Hi Toan,
if you have some free time feel free to start and draft ideas already. I'm open
to any patch proposal anyone sends in ;-)!
René
On 09.11.2012, at 20:56, Toan Pham wrote:
> Yes, we had removing this static number sort priority on our T2 wishlist for
> some time already. But as trunk appears to work quite well for most of us
> (embedded, server, VM, etc. pp use) and most of us appear to be a little busy
> with work, side projects and so on we did not untertake too experimental
> changes in the latest time.
>
> If we touch the dependency thing we should get rid of this numeric priority
> entirely.
>
> Whenever you have time to re-do dependency build order, please let me know.
> I have three different ideas and they can definitely improve t2SDE.
>
>
> I would suggest to go the route to consider all dependencies hard, and we
> just mark optional. As optional deps are usual at fancy UI apps etc. That
> should be easier and faster to get working fast.
>
>
> There are two types of dependency: (1) build-time and (2) run-time.
> 1. build-time dependency is pretty much a minimal tool chain to build a
> package. ( autoconfig, sed, linux-header etc...)
> 2. run-time dependency is, on the other hand, part of the rootfs. It
> changes depending on a package's configuration. ie. cairo with x11 backend
> vs. cairo with directfb backend. However, some dependencies are hard linked.
> For instance, cairo always uses libpixman/libc.
>
> I agree with you with making runtime dependency optional. Some dependencies
> are hard, regardless. We need to have three dependency tags to address the
> three types of dependency.
>
>
> Which scenario of build setup and packages are you having trouble with right
> now?
>
> From the top of my head, I think it was libxt that failed. it depended on
> util-linux. I resolved this issue temporarily by running pkgloop several
> times if I detected an initial build condition.
--
René Rebe, ExactCODE GmbH, Jaegerstr. 67, DE-10117 Berlin
http://exactcode.com | http://t2-project.org | http://rene.rebe.de
-----------------------------------------------------------
If you wish to unsubscribe from this mailing, send mail to
[email protected] with a subject of: unsubscribe t2
