1 |
On Tue, 27 Oct 2015 22:46:35 -0400 |
2 |
Rich Freeman <rich0@g.o> wrote: |
3 |
|
4 |
> On Tue, Oct 27, 2015 at 10:06 PM, hasufell <hasufell@g.o> wrote: |
5 |
> > |
6 |
> > B) 1 feature flag, 3 strict provider flags |
7 |
> > * ssl: enable any sort of SSL/TLS support |
8 |
> > * gnutls: only to enable gnutls provided ssl support in case there |
9 |
> > is a choice |
10 |
> > * openssl: only to enable openssl provided ssl support in case |
11 |
> > there is a choice (should not be implemented as !gnutls?) |
12 |
> > * libressl: only to enable libressl provided ssl support in case there |
13 |
> > is a choice, must conflict with 'openssl' USE flag |
14 |
> > |
15 |
> > consequences: |
16 |
> > * REQUIRED_USE="^^ ( openssl libressl )" is not only allowed, it is |
17 |
> > _mandatory_ |
18 |
> > * packages like media-video/ffmpeg _must_ switch the USE flag |
19 |
> > openssl->ssl to avoid breaking global USE flags |
20 |
> > * !gnutls? ( dev-libs/openssl:0 ) will be bad form or even disallowed |
21 |
> > |
22 |
> > B will definitely be more work, but ofc is also a lot cleaner and |
23 |
> > totally unambigous. |
24 |
> > |
25 |
> |
26 |
> ++ |
27 |
> |
28 |
> The pain is for a short time. Then we have to live with this for a |
29 |
> long time. USE flags should have one meaning. The fact that this |
30 |
> isn't the case right now is already a bug. We don't need to |
31 |
> perpetuate it. |
32 |
|
33 |
No, the pain is neverending. You define a number of flags which are |
34 |
scattered all over the place and there's practically no good value but |
35 |
the 'default'. |
36 |
|
37 |
We have no way of saying 'I prefer polarssl, then gnutls, then |
38 |
libressl, and never openssl'. Whatever I put in USE, I'm going to hit |
39 |
one kind of REQUIRED_USE issues, or other. And in the end, I end up |
40 |
having huge package.use just to make things work. |
41 |
|
42 |
How is that a 'short time' pain? |
43 |
|
44 |
-- |
45 |
Best regards, |
46 |
Michał Górny |
47 |
<http://dev.gentoo.org/~mgorny/> |