| 1 |
>>>>> On Sun, 20 Jun 2021, Michał Górny wrote: |
| 2 |
|
| 3 |
> @@ -712,8 +714,13 @@ python_newexe() { |
| 4 |
> ) |
| 5 |
> |
| 6 |
> # install the wrapper |
| 7 |
> - _python_ln_rel "${ED%/}"/usr/lib/python-exec/python-exec2 \ |
| 8 |
> - "${ED%/}/${wrapd}/${newfn}" || die |
| 9 |
> + if [[ ${EAPI} != [67] ]]; then |
| 10 |
> + dosym -r /usr/lib/python-exec/python-exec2 \ |
| 11 |
> + "${wrapd}/${newfn}" |
| 12 |
> + else |
| 13 |
> + _python_ln_rel "${ED%/}"/usr/lib/python-exec/python-exec2 \ |
| 14 |
> + "${ED%/}/${wrapd}/${newfn}" || die |
| 15 |
> + fi |
| 16 |
|
| 17 |
Wouldn't it be cleaner to inherit eapi8-dosym.eclass in EAPIs 6 and 7 |
| 18 |
instead? Then dosym -r could be used unconditionally, and function |
| 19 |
_python_ln_rel could be dropped altogether? |
| 20 |
|
| 21 |
Same for python-r1.eclass (patch 20/28) and distutils-r1.eclass |
| 22 |
(patch 27/28). |
| 23 |
|
| 24 |
Ulrich |
Archives