| 1 |
Mike Gilbert schrieb: |
| 2 |
> On Tue, Jul 31, 2012 at 11:31 AM, Mike Gilbert <floppym@g.o> wrote: |
| 3 |
>> On Mon, Jul 30, 2012 at 1:05 PM, Mike Gilbert <floppym@g.o> wrote: |
| 4 |
>>>> So do you know how many ebuilds we'd have to update to get it right? |
| 5 |
>>>> |
| 6 |
>>> |
| 7 |
>>> At the moment, no. If someone could help me write a script to identify |
| 8 |
>>> affected packages, that would be great. |
| 9 |
>>> |
| 10 |
>>> If said script uses the portage api to determine the dependency tree, |
| 11 |
>>> we will want to modify python.eclass first to reduce the number of |
| 12 |
>>> hits. This could be done in a local copy of the tree. |
| 13 |
>> |
| 14 |
>> I've created a quick little script and posted it in my devspace. Now I |
| 15 |
>> just need to mess with python.eclass. |
| 16 |
>> |
| 17 |
>> http://dev.gentoo.org/~floppym/python-deps.py |
| 18 |
> |
| 19 |
> After modifying python.eclass, distutils.eclass, java-ant-2.eclass, |
| 20 |
> and waf-utils.eclass, there are 532 dependencies that would need to be |
| 21 |
> modified to make the || dep plan work. I have attached the list for |
| 22 |
> reference. |
| 23 |
> |
| 24 |
> Does this seem like a viable option? Is anyone going to volunteer to |
| 25 |
> take it from here? |
| 26 |
> |
| 27 |
|
| 28 |
If everyone agrees with this route, i suggest a post to the gentoo-dev |
| 29 |
mailing list about this planned change and a request for maintainers to |
| 30 |
change their dependencies accordingly. |
| 31 |
Additionally i suggest to add it also to some written doc, e.g. the |
| 32 |
python developer guide. |
| 33 |
After giving maintainers some time for that change, we could then open |
| 34 |
bugs or do the remaining needed conversions (when the maintainer is |
| 35 |
missing or gives his ok). |
| 36 |
|
| 37 |
|
| 38 |
-- |
| 39 |
|
| 40 |
Thomas Sachau |
| 41 |
Gentoo Linux Developer |
Archives