| 1 |
Mark Loeser wrote: |
| 2 |
> Mike Frysinger <vapier@g.o> said: |
| 3 |
>> On Sunday 30 March 2008, Mark Loeser wrote: |
| 4 |
>>> Actually, I'd say this should just be removed. If a user wants to apply |
| 5 |
>>> a patch, they can put their own ebuild into an overlay and do it |
| 6 |
>>> themselves (presumably if they want to patch something, they'll know how |
| 7 |
>>> to make the simple modifications to an ebuild). By allowing the user to |
| 8 |
>>> arbitrarily patch something means we have no idea what the user has |
| 9 |
>>> built and is filing a bug about. If they installed an ebuild from an |
| 10 |
>>> overlay it is a lot easier to identify what they built. Sure, they |
| 11 |
>>> could patch the ebuild in their tree, but by supporting user supplied |
| 12 |
>>> patches easily in this way, we are encouraging them to patch things |
| 13 |
>>> without our knowledge. If we start supporting this across the board, I |
| 14 |
>>> can see bugs being filed when their patches break and they don't |
| 15 |
>>> understand what is happening. |
| 16 |
>> that's actually exactly what i'm encouraging. i'm not worried about such |
| 17 |
>> issues as they're easily resolved by people posting the full build log. |
| 18 |
> |
| 19 |
> Which is great, but I think this is something we should discuss and |
| 20 |
> figure out if this is something we want to introduce into the tree (too |
| 21 |
> late now, but better late than never). If it is something we want to |
| 22 |
> move forward with, it should be introduced at the package manager level |
| 23 |
> instead of being an in-tree package manager specific feature. |
| 24 |
|
| 25 |
I think that maybe we should first introduce new patching phase and then |
| 26 |
make this user patch really usable feature. For example if you want to |
| 27 |
patch something that's input to running autotools, doing it in |
| 28 |
post_src_unpack is too late... |
| 29 |
|
| 30 |
Caster |
| 31 |
-- |
| 32 |
gentoo-dev@l.g.o mailing list |
Archives