| 1 |
-----BEGIN PGP SIGNED MESSAGE----- |
| 2 |
Hash: SHA256 |
| 3 |
|
| 4 |
On 06/08/13 04:22 PM, Alexis Ballier wrote: |
| 5 |
> On Tue, 6 Aug 2013 20:44:57 +0100 Ciaran McCreesh |
| 6 |
> <ciaran.mccreesh@××××××××××.com> wrote: |
| 7 |
> |
| 8 |
>> On Tue, 6 Aug 2013 15:31:14 -0400 Alexis Ballier |
| 9 |
>> <aballier@g.o> wrote: |
| 10 |
>>> Well, ok, but this doesn't relate to what I was writing. |
| 11 |
>>> Subslot, or slot emulators or whatever, in their current usage |
| 12 |
>>> with := dependencies, are not fine grained enough for some use |
| 13 |
>>> cases. Those cause regressions if used improperly. |
| 14 |
>> |
| 15 |
>> There is no regression. Previously, packages sometimes broke |
| 16 |
>> when doing an upgrade. Now, packages do not break when doing an |
| 17 |
>> upgrade. |
| 18 |
> |
| 19 |
> The regression is the useless rebuild. Without preserve-libs, |
| 20 |
> packages break even more: cf the libc example. |
| 21 |
> |
| 22 |
|
| 23 |
Terminology issue. Useless rebuilds are not regressions. they might |
| 24 |
be undesirable, they might be bugs, but they are not something that |
| 25 |
was happening before, and then was fixed, and now is happening again. |
| 26 |
|
| 27 |
The only thing "regression" can be applied to in this discussion is if |
| 28 |
a consumer is NOT using a := slot-operator on a dep like poppler, and |
| 29 |
the poppler libs that consumer uses are updated and breakage occurrs |
| 30 |
- -- ie, the exact same case that was happening all the time prior to |
| 31 |
EAPI5 adoption. IE, exactly what you are proposing to do in order to |
| 32 |
reduce the extra rebuilds. |
| 33 |
|
| 34 |
|
| 35 |
|
| 36 |
-----BEGIN PGP SIGNATURE----- |
| 37 |
Version: GnuPG v2.0.20 (GNU/Linux) |
| 38 |
|
| 39 |
iF4EAREIAAYFAlIBXFYACgkQ2ugaI38ACPBwlQEAp9nVnOws3WxWyfmK7xuuCzfb |
| 40 |
sUwYmIB6lJv9G3FljpkA/RtMWa55WxihhfML9VqlQJ+HqSfP69RMTrbIjmGKufbv |
| 41 |
=PyPl |
| 42 |
-----END PGP SIGNATURE----- |
Archives