| 1 |
Jason Stubbs wrote: |
| 2 |
> On Monday 31 October 2005 05:29, Zac Medico wrote: |
| 3 |
> |
| 4 |
>>I choose the bandaid method in order to keep the patch as small as possible |
| 5 |
>>and still keep PORTAGE_TMPDIR clean. AFAICT it would take a *much* larger |
| 6 |
>>patch to clean up all the same cases without the bandaid approach. |
| 7 |
>> |
| 8 |
>>On a side note, my patch should be changed so that it doesn't try to |
| 9 |
>>clean_builddir when things like "--pretend" and "--fetchonly" are in |
| 10 |
>>emerge's myopts. |
| 11 |
> |
| 12 |
> |
| 13 |
> The exceptions are likely not the only ones. The exceptions to the exceptions |
| 14 |
> to the ... is what is bad about bandaids. I don't know what the "proper" fix |
| 15 |
> is in this case, but if you try it you'll probably find it's smaller and/or |
| 16 |
> less work that you guessed it to be. |
| 17 |
> |
| 18 |
|
| 19 |
Yeah, I agree with you and Brian. I don't expect a proper fix to be difficult or time consuming (though I expect the patch size to be a bit larger due to minor refactoring that seems necessary). Anyways, I wanted to see what a small bandaid patch would look like so that it could be compared in size and complexity to a more proper fix. |
| 20 |
|
| 21 |
Zac |
| 22 |
-- |
| 23 |
gentoo-portage-dev@g.o mailing list |
Archives