| 1 |
On Mon, 4 Aug 2008 21:14:44 +0000 (UTC) |
| 2 |
Duncan <1i5t5.duncan@×××.net> wrote: |
| 3 |
> Has any tracking been done on the tree as it currently exists such |
| 4 |
> that an estimate of the number of such bugs expected can be made? |
| 5 |
> Surely, while thinking about making a commitment to the statement |
| 6 |
> above, the number of bugs it's going to trigger is apropos. |
| 7 |
|
| 8 |
The only tracking that's done currently, and the only tracking we can |
| 9 |
really work out how to do, is "stuff that breaks with Paludis". And |
| 10 |
that doesn't get everything, because quite a few of the flukes in how |
| 11 |
Paludis behaves are similar to some of the flukes in how Portage |
| 12 |
behaves. |
| 13 |
|
| 14 |
Some things we can validate easily -- for example, we can check that |
| 15 |
depend strings parse using a strictly PMS-enforcing parser. But for |
| 16 |
weird quirks that make EAPI 0 so difficult to pin down, there's not |
| 17 |
much that can be done. |
| 18 |
|
| 19 |
-- |
| 20 |
Ciaran McCreesh |
Archives