On 10/24/10, Chris Larson <[email protected]> wrote: > On Sun, Oct 24, 2010 at 12:56 PM, Holger Freyther > <[email protected]<holger%[email protected]> >> wrote: > >> On 10/24/2010 08:35 PM, Khem Raj wrote: >> >> >> >> >> this is used in all efl recipes and I've noticed it in elementary only >> >> because python-elementary depends on such new version. >> > >> > Do you have a list of recipes which use this ? I am sure with git it >> > will be hard to use such a thing >> > as the git revs are random and simlarily many other SCMs >> >> >> Speaking as upstream co-author of OpenBSC. We are using something like >> this >> to >> get the revision from the last tag of the repository. If someone is doing >> a >> make dist(check) we will copy a special version file into the tar.gz and >> use >> that instead. >> >> From an upstream point of view I think it is acceptable to say that the >> ones >> that build stuff from a SCM (in contrast to a tarball) should have the >> files >> of the SCM around... >> >> PS: I liked the --exclude patch a lot too... > > > I don't think its necessarily the fact that we're building from an SCM > that's the differentiating factor here, it's the build from a branch as > opposed to a tag. The tagged sources in an scm may be (though aren't > always) exactly the same as the tarball contents, in which case the scm > files shouldn't be necessary. Ideally, the buildsystems would work with or > without them, but I agree, I guess we may have to revert the patch :(
another option is a to add a qualifier to bitbake fetcher to tell it that we want to keep SCM metadata for a given fetch and keep the default to exclude it but enable it for such cases. > -- > Christopher Larson > clarson at kergoth dot com > Founder - BitBake, OpenEmbedded, OpenZaurus > Maintainer - Tslib > Senior Software Engineer, Mentor Graphics > -- -Khem _______________________________________________ Bitbake-dev mailing list [email protected] https://lists.berlios.de/mailman/listinfo/bitbake-dev
