| 1 |
On Sun, May 06, 2007 at 04:10:56PM +0100, Stephen Bennett wrote: |
| 2 |
> On Sat, 05 May 2007 18:40:13 -0700 |
| 3 |
> Zac Medico <zmedico@g.o> wrote: |
| 4 |
> |
| 5 |
> > Let's sure we talking about the same thing when we say "implicit |
| 6 |
> > _p0". The patch attached to bug 171259 will make ntp-4.2.4_p0 |
| 7 |
> > greater than ntp-4.2.4, but ntp-4.2.4_p will still be considered |
| 8 |
> > equal to ntp-4.2.4_p0. |
| 9 |
> |
| 10 |
> OK, that change makes sense, and is in fact what PMS in its current |
| 11 |
> wording requires. One or the other should be changed to match, and I |
| 12 |
> think the PMS version at the moment makes more sense. |
| 13 |
|
| 14 |
As indicated above, that's actually a change to the long standing |
| 15 |
behaviour; personally, I'm inclined to just block _p0 from being used |
| 16 |
in ebuild version (meaning repoman). |
| 17 |
|
| 18 |
Reasoning is simple enough- we disallow -r0 from being used for |
| 19 |
similar reasons (if it's implicit, adding it makes uniqueness annoying |
| 20 |
further it's unneeded). |
| 21 |
|
| 22 |
My 2 cents at least. |
| 23 |
~harring |
Archives