Excerpts from Michael K. Johnson's message of 2014-03-19 01:42:10 +0100: > > > * Using Conary's extensive package build automation to make it easier > > > to build better packages for Fedora. > > > > can you elaborate how we would do that? > > a recipe that depends on and invokes rpm build tools and spits out an > > encapsulated rpm instead of a native conary package? > > Essentially, yes. > > Basically, taking advantage of Conary's superclasses, rich policy, > automatic componentization, etc, and making it available to Fedora > packagers so that they have less work to do to make high quality > packages, and can concentrate only on the useful packaging. Most > of what you're used to with writing recipes for Foresight... > > > well, i suppose it would be a factory that can handle all rpm packages > > otherwise a packager would have two, the conary recipe and the rpm spec to > > manage for each package. > > What I had in mind here would not require a spec file to be > maintained. Conary would probably write a minimal one to > call rpmbuild but control the build process. There is more > than one way to do that, and if there is interest on the Fedora > side we can discuss which way works best then...
well, this topic opens a lot of questions that we won't get answers to in the next two months i think, since we'll still be busy with the base. we may get those answers once we come to creating additional packages as it could be interesting just for us to not only build foresight packages, but rpms that we can submit to fedora. (and here i believe the specfile must be complete so that the rpm can be rebuilt without conary tools. not sure if a generated spec can do that, unless conary can generate the spec from a recipe. now that would be interesting!) > This is something that Andy Grimm did a lightning talk on at a > previous fudcon, to some interest. can you help me track that down? > > the summary is just the kind of outline i was looking for. > > i found nothing that i could add or remove from it or from the text. > > i'd present the talk as made by the foresight development team with me just > > acting as a speaker. > > In that case, though, I suggest you develop it further on the list > here, so that everyone can contribute. good idea. the talk is definitely far from finished. what you provided is suitable for a submission and to frame the topic. next the picture needs to be filled with detail. greetings, martin. -- eKita - the online platform for your entire academic life hackerspace beijing - http://qike.info -- chief engineer eKita.co pike programmer pike.lysator.liu.se caudium.net societyserver.org BLUG secretary beijinglug.org foresight developer foresightlinux.org realss.com unix sysadmin Martin Bähr working in china http://societyserver.org/mbaehr/ _______________________________________________ Foresight-devel mailing list [email protected] https://lists.foresightlinux.org/mailman/listinfo/foresight-devel
