| 1 |
On 23:23 Wed 25 Mar , Ciaran McCreesh wrote: |
| 2 |
> On Wed, 25 Mar 2009 23:06:37 +0100 |
| 3 |
> Donnie Berkholz <dberkholz@g.o> wrote: |
| 4 |
> > > 10) dohard and dosed banned in EAPI 3 |
| 5 |
> > |
| 6 |
> > I think I missed the reasoning for removing these, particularly |
| 7 |
> > dosed. pybugz didn't see any open bugs. |
| 8 |
> |
| 9 |
> Portage doesn't merge hardlinks correctly, so dohard is bad. And |
| 10 |
> dosed's been considered deprecated for years. |
| 11 |
|
| 12 |
I like dosed because I've used it many times for the $D-removing |
| 13 |
feature. If there was (is?) an automatic filter at install-time that |
| 14 |
scanned files to remove $D references, that would work for me instead. |
| 15 |
|
| 16 |
> > > 11) doinclude, newinclude for EAPI 3 |
| 17 |
> > |
| 18 |
> > Is installing to /usr/include by default useful for most packages |
| 19 |
> > that want to use this? Or would they /usr/include/${PN}? If you have |
| 20 |
> > to change it often, aren't you just as well off using insinto/doins? |
| 21 |
> > Should there be an "includeinto"? |
| 22 |
> |
| 23 |
> I'd be inclined to agree on that one, but people seem to be after more |
| 24 |
> of these do* things. |
| 25 |
|
| 26 |
I'm still not convinced this one adds anything significant beyond doins |
| 27 |
by default that it's worth adding. In half the cases |
| 28 |
(/usr/include/$PN/foo.h or anywhere else vs /usr/include/foo.h), you'd |
| 29 |
need to customize its install directory and so effectively are just |
| 30 |
running insinto/doins. Plus it's only useful for ebuilds that have a |
| 31 |
crappy install anyway, so wouldn't we rather fix the build system? |
| 32 |
|
| 33 |
> > > 21) REPLACING_VERSIONS and REPLACED_BY_VERSION in EAPI 3 |
| 34 |
> > |
| 35 |
> > I'm curious why it isn't global. Seems like it would make sense to |
| 36 |
> > put it near dependencies. Also I could be wrong, but wouldn't you |
| 37 |
> > want to be able to cache this and show smart pretend output, etc? |
| 38 |
> |
| 39 |
> I think you're misunderstanding what this is for. It's to allow |
| 40 |
> packages to work out whether they're upgrading / downgrading / |
| 41 |
> reinstalling / whatever, since Zac broke the devmanual-documented and |
| 42 |
> PMS-required way of doing it using has_version and refuses to revert it. |
| 43 |
|
| 44 |
Ah, I see. This is the first I'd heard about this anywhere (the problem |
| 45 |
or the proposal). So running has_version in setup/preinst and again in |
| 46 |
postinst doesn't do what you'd want anymore? |
| 47 |
|
| 48 |
-- |
| 49 |
Thanks, |
| 50 |
Donnie |
| 51 |
|
| 52 |
Donnie Berkholz |
| 53 |
Developer, Gentoo Linux |
| 54 |
Blog: http://dberkholz.wordpress.com |
Archives