| 1 |
-----BEGIN PGP SIGNED MESSAGE----- |
| 2 |
Hash: SHA256 |
| 3 |
|
| 4 |
On 26/09/13 06:51 AM, Michael Palimaka wrote: |
| 5 |
> On 26/09/2013 17:53, Michał Górny wrote: |
| 6 |
>> How do we handle packages which install multiple libraries? I'm |
| 7 |
>> afraid forcing such a policy and/or hurrying developers to adapt |
| 8 |
>> will only cause more of poppler-like issues to occur. |
| 9 |
>> |
| 10 |
> There isn't a 100% perfect solution currently, and I agree that |
| 11 |
> hurrying people will simply move us from "not enough rebuilds" to |
| 12 |
> "too many rebuilds". |
| 13 |
|
| 14 |
Enforcing consistency is much more important imo than "emerge -uDN |
| 15 |
@world" efficiency. For those users that need more efficiency they |
| 16 |
can always get it by upgrading individual packages with |
| 17 |
'--rebuild-ignore' or '--ignore-built-slot-operator-deps y' after |
| 18 |
seeing what all is going to be rebuilt via 'emerge -uDNav' |
| 19 |
|
| 20 |
To be honest, the main issue that I see keeping slot-operators from |
| 21 |
being properly used has to do with virtuals, because subslots (or full |
| 22 |
slots for that matter) do not propagate through them. But that |
| 23 |
doesn't hold back converting libs and rdeps for everything not falling |
| 24 |
to a virtual. |
| 25 |
|
| 26 |
> Poppler was a great example of what can go wrong. Apart from |
| 27 |
> people being forced to rebuild packages that link only against one |
| 28 |
> of the stable interfaces, I even saw rebuilds forced for packages |
| 29 |
> that didn't even link against the libraries. |
| 30 |
|
| 31 |
The latter in that case was a mis-use of the ':=' on the poppler atom |
| 32 |
in *DEPEND. |
| 33 |
|
| 34 |
-----BEGIN PGP SIGNATURE----- |
| 35 |
Version: GnuPG v2.0.20 (GNU/Linux) |
| 36 |
|
| 37 |
iF4EAREIAAYFAlJEQOkACgkQ2ugaI38ACPCp1QEAsSX+efIdTGRZ94EIgyYzQcSF |
| 38 |
4TeEmFvzanp5A/6DL94BAIVnw3ayjTmOmUYevRl/Hr0cEyVv4X9T+bFnhngW6Ops |
| 39 |
=CsnH |
| 40 |
-----END PGP SIGNATURE----- |
Archives