On Sun, 8 Jul 2012 23:40:29 +0200 "Andreas K. Huettel" <dilfri...@gentoo.org> wrote:
> Am Sonntag 08 Juli 2012, 22:10:02 schrieb Michał Górny: > > On Sun, 08 Jul 2012 19:49:25 +0200 > > > > René Neumann <li...@necoro.eu> wrote: > > > Hi all, > > > > > > I'd like just to receive a short clarification about the 'status' > > > of base.eclass: Is this eclass expected to be available > > > everywhere, i.e. should each eclass make sure it imports and > > > incorporates it. Or is it just an eclass like the others and > > > ebuilds should make sure they inherit it if needed? > > > > No. It is unmaintained, has serious design flaws and it simply > > should not be used anywhere. At least in EAPI != [01]. > > Please clarify this. > > A lot of (inheriting eclasses and) packages depend on features > provided by base.eclass (e.g., PATCHES), which are pretty neat and > which I would sorely miss. So I would certainly object to deprecating > base.eclass, unless its relevant functionality is only moving to a > better place. base.eclass is randomly exporting non-requested, non-wanted phase functions colliding with other inherited eclasses. It's just the lexical order of inherits what stops mayhem from happening. In other words, base.eclass is only suitable if you are expecting to export *all* phase functions which simply doesn't happen in eclasses. For example, if distutils used base eclass, every VCS eclass inherited before it would be ignored (due to src_unpack() redefined to default one for no good reason). -- Best regards, Michał Górny
signature.asc
Description: PGP signature
