| 1 |
Ühel kenal päeval, E, 10.04.2017 kell 14:44, kirjutas William L. |
| 2 |
Thomson Jr.: |
| 3 |
> Again go modify a few hundred python packages to remove say 3.4. I |
| 4 |
> think about 10-20 ebuilds in. You will be scripting and looking for |
| 5 |
> another way.... |
| 6 |
|
| 7 |
No, for that you simple remove python3_4 from _PYTHON_ALL_IMPLS in |
| 8 |
python-utils-r1.eclass and call it a day. Some other day you might make |
| 9 |
a mass commit to remove 3_4 from PYTHON_COMPAT of all in-tree ebuilds, |
| 10 |
but that's just janitorial and no other effect. |
| 11 |
|
| 12 |
The current implementation makes perfect sense to me, and follows one |
| 13 |
of the zens of python: |
| 14 |
Explicit is better than implicit. |
| 15 |
|
| 16 |
Declare explicitly what version is supported, don't implicitly do so |
| 17 |
and merely hope there are no issues. If some lower level module doesn't |
| 18 |
work with new python and your higher level module wants to use the new |
| 19 |
python, repoman will catch it for you due to it having been explicit |
| 20 |
via PYTHON_USE_DEP. |
| 21 |
|
| 22 |
There is no difference with reverse approach if one would mass commit |
| 23 |
the new COMPAT into all ebuilds upon the introduction of a new python |
| 24 |
slot, but this is not done, because things would break. |
| 25 |
|
| 26 |
|
| 27 |
Mart |
Archives