On Tue, Mar 23, 2010 at 2:32 AM, Frans Meulenbroeks < [email protected]> wrote:
> Chris, while this is a nice idea it conflicts with having checksums in > the recipe, as that makes the recipes unique. > Unless of course there is (or we create) a way to have multiple > checksums in a recipe and pick the one that is for the version we are > building. > See the gist for nano that I linked in the email, I set it up so that works. SRC_URI_<pv> flags are automatically transferred to SRC_URI via an anonymous python function. I have a checksums.inc that sets them all. Then again I feel it is in most cases better to move forward. E.g. for > your nano example: why would people want to build say nano 1.0.2 if > there is also a working 1.0.6 recipe (or even a 2.2.3 one or whatever > version it is at). I feel it is better to spent time to > fix/improve/add/repair the latest version than spend time fixing old > code. > (generally speaking that is) > Well, in part I agree, and in part I disagree. This is why I started this thread about the worth of the feature. For companies and the like, keeping old versions around indefinitely is a good thing, and while I know many will argue that they can keep their recipes for it around, that just encourages deviation from OE, and believe me, I know how much of a pain it is for a company to stay in sync with OE :) Thanks for the comments. -- Christopher Larson clarson at kergoth dot com Founder - BitBake, OpenEmbedded, OpenZaurus Maintainer - Tslib Senior Software Engineer, Mentor Graphics _______________________________________________ Openembedded-devel mailing list [email protected] http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
