1 |
Dear all, |
2 |
|
3 |
Bug 684962 (dev-python/ipython-7.5.0: package conflicts) has |
4 |
demonstrated a painful consequence when upstream start to release |
5 |
python3 only versions. |
6 |
|
7 |
Upstream has dropped python-2.7 support in dev-python/ipython-7.5.0, |
8 |
thus there is no python_targets_python2_7 USE flag for the ebuild. |
9 |
dev-python/qtconsole, a dependant of dev-python/ipython, still supports |
10 |
python-2.7. When qtconsole get emerged with |
11 |
USE="python_targets_python2_7 python_targets_python3_6" for example, old |
12 |
dev-python/ipython-5.8.0-r1 is drawn, resulting in conflict against |
13 |
dev-python/ipython-7.5.0. USE=python_targets_python2_7 had to be |
14 |
removed from dev-python/qtconsole to avoid it. |
15 |
|
16 |
If one package drops python-2.7, all its dependants have to drop |
17 |
python-2.7 even if they can work with python-2.7. |
18 |
|
19 |
|
20 |
Given the python-2 countdown deadline being 2020-01-01, a month away, |
21 |
shall we get rid of python-2? |
22 |
|
23 |
|
24 |
If the answer is yes, we will need to decide on the following |
25 |
python-2-only packages. |
26 |
|
27 |
,---- |
28 |
| $ comm -23 <(equery -qC h python_targets_python2_7 | sort ) <(equery -qC h python_targets_python3_6 | sort) |
29 |
| dev-lang/yasm-1.3.0 |
30 |
| dev-libs/libxslt-1.1.33-r1 |
31 |
| dev-python/backports-functools-lru-cache-1.5 |
32 |
| dev-python/enum34-1.1.6-r1 |
33 |
| dev-python/functools32-3.2.3 |
34 |
| dev-python/futures-3.2.0 |
35 |
| dev-python/pygobject-2.28.6-r55 |
36 |
| dev-python/pygtk-2.24.0-r4 |
37 |
| dev-python/subprocess32-3.2.7 |
38 |
| dev-util/boost-build-1.70.0 |
39 |
| dev-vcs/subversion-1.12.2 |
40 |
| gnome-base/libglade-2.6.4-r2 |
41 |
| net-analyzer/nmap-7.70 |
42 |
| sys-devel/clang-8.0.1 |
43 |
| x11-wm/xpra-2.4.3 |
44 |
`---- |
45 |
|
46 |
If the answer is no, to avoid holding back new versions having only |
47 |
python3, such as bug 671796 for dev-python/matplotlib bump, old versions |
48 |
with python_targets_python2_7 and new versions without should be |
49 |
co-installable into different SLOTs. |
50 |
|
51 |
What do you think? |
52 |
|
53 |
Yours, |
54 |
Benda |