| 1 |
> |
| 2 |
> Sort of responding to/being inspired by haubi's comment on -dev about |
| 3 |
> "inherit eapi 4", I wondered whether we should make a prefix.eclass. |
| 4 |
> |
| 5 |
> Currently we have eprefixify as function provided by Portage, and hence |
| 6 |
> ebuilds that use eprefixify cannot be straight ported to gentoo-x86. |
| 7 |
> Most notable are the eselect-* ebuilds that some gentoo-x86 devs like |
| 8 |
> to |
| 9 |
> give full Prefix support, but simply can't because eprefixify (which we |
| 10 |
> would use) cannot be used in gentoo-x86. |
| 11 |
> |
| 12 |
> So, should we start an eclass with for now eprefixify in there, and for |
| 13 |
> every ebuild where we use it, start inheriting prefix and nuke the |
| 14 |
> function from Portage? Somehow I think this is a good idea. |
| 15 |
|
| 16 |
Hmm, sounds like a good idea. This would indeed solve some of the posed problems. |
| 17 |
|
| 18 |
> |
| 19 |
> Since I like haubi's idea about inherit doing the magic for eapis, |
| 20 |
> perhaps this can even solve the problem we have with our EAPI mungling |
| 21 |
> (all our ebuilds just inherit eapi prefix to "tag" them) and all |
| 22 |
> eclasses (it remains a hell of a job to keep it working), as well as |
| 23 |
> having prefix ebuilds live happily next to non-prefix ones. But for |
| 24 |
> that we really need Portage to be able to mask based on what's in the |
| 25 |
> inherit line, e.g. the resolver needs to take it into account. |
| 26 |
> |
| 27 |
|
| 28 |
I believe this can be done without too much effort, although I'm not sure if this can be considered "clean". OTOH, why not? This would enable users to express things like: "I want live ebuilds, but none that come from svn" (and thus inherit subversion.eclass). |
| 29 |
|
| 30 |
Cheers, Markus |
| 31 |
|
| 32 |
> |
| 33 |
> -- |
| 34 |
> Fabian Groffen |
| 35 |
> Gentoo on a different level |
Archives