1 |
On Fri, 2020-06-26 at 09:51 +0100, Sergei Trofimovich wrote: |
2 |
> On Fri, 26 Jun 2020 07:29:45 +0000 |
3 |
> Michał Górny <mgorny@g.o> wrote: |
4 |
> |
5 |
> > Dnia June 26, 2020 6:42:57 AM UTC, Sergei Trofimovich <slyfox@g.o> napisał(a): |
6 |
> > > On Sat, 20 Jun 2020 16:29:53 +0100 |
7 |
> > > Sergei Trofimovich <slyfox@g.o> wrote: |
8 |
> > > |
9 |
> > > > On Sat, 20 Jun 2020 16:05:38 +0200 |
10 |
> > > > Michał Górny <mgorny@g.o> wrote: |
11 |
> > > > |
12 |
> > > > > On Sat, 2020-06-20 at 14:57 +0100, Sergei Trofimovich wrote: |
13 |
> > > > > > Give maintainers the chance to act and flag packages that pull in |
14 |
> > > python:2.7. |
15 |
> > > > > > Signed-off-by: Sergei Trofimovich <slyfox@g.o> |
16 |
> > > > > > --- |
17 |
> > > > > > profiles/package.deprecated | 4 ++++ |
18 |
> > > > > > 1 file changed, 4 insertions(+) |
19 |
> > > > > > |
20 |
> > > > > > diff --git a/profiles/package.deprecated |
21 |
> > > b/profiles/package.deprecated |
22 |
> > > > > > index a756e845f47..bb661571962 100644 |
23 |
> > > > > > --- a/profiles/package.deprecated |
24 |
> > > > > > +++ b/profiles/package.deprecated |
25 |
> > > > > > @@ -17,6 +17,10 @@ |
26 |
> > > > > > |
27 |
> > > > > > #--- END OF EXAMPLES --- |
28 |
> > > > > > |
29 |
> > > > > > +# Sergei Trofimovich <slyfox@g.o> (2020-06-20) |
30 |
> > > > > > +# Deprecated. Consider poring to python 3 and drop support for |
31 |
> > > python2. |
32 |
> > > > > > +dev-lang/python:2.7 |
33 |
> > > > > > + |
34 |
> > > > > > # Sergei Trofimovich <slyfox@g.o> (2020-02-22) |
35 |
> > > > > > # virtual/libstdc++ has only one sys-libs/libstdc++-v3 provider. |
36 |
> > > > > > # Use that instead. Or even better use none of them. It's a |
37 |
> > > > > |
38 |
> > > > |
39 |
> > > > > It will trigger the same for packages that support *only* |
40 |
> > > > > Python 2.7, as well as these that support 2.7 in addition to 3 |
41 |
> > > because |
42 |
> > > > > they have 2.7 deps. |
43 |
> > > > |
44 |
> > > > If we expect actions by developers on both cases I don't see a |
45 |
> > > problem with that. |
46 |
> > > |
47 |
> > > Pushed as: |
48 |
> > > https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=79d65d6641cfc0ef7b44df491c390e8c880e3049 |
49 |
> > > with full text being: |
50 |
> > > |
51 |
> > > +# Sergei Trofimovich <slyfox@g.o> (2020-06-26) |
52 |
> > > +# Deprecated. |
53 |
> > > +# - optional python:2.7 dependency should be dropped if no reverse |
54 |
> > > +# dependencies are using it. |
55 |
> > > +# - mandatory python:2.7 depepndency will require package porting |
56 |
> > > +# or package removal if no reverse dependencies are using it. |
57 |
> > > +dev-lang/python:2.7 |
58 |
> > |
59 |
> > You've just introduced 829 CI warnings |
60 |
> |
61 |
> That's the intention. |
62 |
> |
63 |
> > effectively disabling the ability to distinguish *new* problems in these packages. |
64 |
> |
65 |
> Correct. Citing above: |
66 |
> |
67 |
> "If we expect actions by developers on both cases I don't see a problem with that." |
68 |
> |
69 |
> I assume we still do. |
70 |
|
71 |
Not exactly. You've pinpointed the wrong target. |
72 |
|
73 |
First of all, we want people to support Python 3. Removing support for |
74 |
Python 2 is a secondary goal. |
75 |
|
76 |
Flagging packages that support Python 2 in addition to Python 3 |
77 |
and cause no trouble in py2 cleanup is doubtful. |
78 |
|
79 |
Flagging packages that support 2+3 because of their revdeps is not |
80 |
helpful at all. It's just noise to the maintainer who can't remove py2 |
81 |
because of revdeps. |
82 |
|
83 |
Flagging dev-python/pypy* which needs py2 but is entirely outside |
84 |
the eclass system is not helpful at all. |
85 |
|
86 |
-- |
87 |
Best regards, |
88 |
Michał Górny |