| 1 |
On Fri, 30 Oct 2015 23:40:28 +0100 |
| 2 |
hasufell <hasufell@g.o> wrote: |
| 3 |
|
| 4 |
> On 10/30/2015 10:16 PM, Anthony G. Basile wrote: |
| 5 |
> > On 10/30/15 3:35 PM, hasufell wrote: |
| 6 |
> >> On 10/30/2015 06:55 PM, Michał Górny wrote: |
| 7 |
> >>> We have no way of saying 'I prefer polarssl, then gnutls, then |
| 8 |
> >>> libressl, and never openssl'. |
| 9 |
> >> I don't think this is something that can be reasonably supported and it |
| 10 |
> >> sounds awfully automagic. And I don't see how this is possible right |
| 11 |
> >> now, so I'm not really sure what you expect to get worse. |
| 12 |
> >> |
| 13 |
> >> E.g. -gnutls pulling in dev-libs/openssl is not really something you'd |
| 14 |
> >> expect. If we go for provider USE flags, then things become consistent, |
| 15 |
> >> explicit and unambiguous. The only problem is our crappy implementation |
| 16 |
> >> of providers USE flags via REQUIRED_USE. |
| 17 |
> >> |
| 18 |
> > I'm not sure what mgorny has in mind, but the problem I see with saying |
| 19 |
> > I want just X to be my provider system wide is that some pkgs build with |
| 20 |
> > X others don't, other pkgs might need a different provider. So it might |
| 21 |
> > make sense to order them in terms of preference: X1 > X2 > X3 ... and |
| 22 |
> > then when emerging a package, the first provider in the preference list |
| 23 |
> > that works is pulled in for that package. |
| 24 |
> > |
| 25 |
> |
| 26 |
> Isn't that basically what the proposal B already was, except that we |
| 27 |
> don't use REQUIRED_USE for it but some sort of pkg_setup/pkg_pretend |
| 28 |
> function? I don't see how those ideas even conflict. |
| 29 |
|
| 30 |
And some sort of magical USE flag meanings? Please stop this right |
| 31 |
here. We don't need 16 USE flag package variants which mean 4 things in |
| 32 |
different, random and unexpected ways. |
| 33 |
|
| 34 |
-- |
| 35 |
Best regards, |
| 36 |
Michał Górny |
| 37 |
<http://dev.gentoo.org/~mgorny/> |
Archives