| 1 |
Am 24.04.2010 13:32, schrieb Gentoo: |
| 2 |
> On Fri, 2010-04-23 at 22:31 -0700, Zac Medico wrote: |
| 3 |
>> On 04/23/2010 05:43 AM, Sebastian Luther wrote: |
| 4 |
>>> Someone might come up with some logic to detect new use flags in |
| 5 |
>>> *DEPEND, but this looks like a hack to me. |
| 6 |
>> |
| 7 |
>> It doesn't seem too bad to me. |
| 8 |
|
| 9 |
It doesn't work, because it's not guaranteed, that only use flag from |
| 10 |
IUSE are used in use conditionals. That means you can't do it reliably |
| 11 |
without the unevaluated value. |
| 12 |
|
| 13 |
>> |
| 14 |
>>> The clean solution is to |
| 15 |
>>> store the unevaluated string. |
| 16 |
>> |
| 17 |
>> Do you want to do this for $PKGDIR/Packages as well? We've always |
| 18 |
>> evaluated USE conditionals in there since we were copying the |
| 19 |
>> behavior of the older genpkgindex tool and that's how it behaved. |
| 20 |
>> |
| 21 |
|
| 22 |
We should do it there too for the same reason as for storing it in the |
| 23 |
vdb. Never heard of that tool, but anyone handling portage's binpkgs |
| 24 |
should use the portage api which provides an easy way to evaluate the |
| 25 |
use conditionals. |
| 26 |
|
| 27 |
>> Also note that if we want to rely on having unevaluated strings then |
| 28 |
>> we'll probably want to try to get alternative package managers to |
| 29 |
>> behave the same way (maybe specify it in PMS). |
| 30 |
|
| 31 |
The vdb isn't covered by PMS. Paludis stores the unevaluated value, |
| 32 |
pkgcore stores the evaluated value. |
| 33 |
|
| 34 |
>> |
| 35 |
>>> Question is: Does anyone have a good argument to not use the old |
| 36 |
>>> behavior again? |
| 37 |
> |
| 38 |
> space and ease of parsing for minimal pkg mergers. |
| 39 |
> |
| 40 |
> |
| 41 |
|
| 42 |
Minimal mergers have to handle it anyway, since this has been the old |
| 43 |
behavior until some weeks ago. For portage API users, the difference is |
| 44 |
a (rather long) one liner. What do you mean with space? |
| 45 |
|
| 46 |
>>> |
| 47 |
>>> Sebastian |
| 48 |
>>> |
| 49 |
>>> [1] commit e6be6590e99522f9be69e2af8eff87919d9bf31f on 2010-02-14 |
| 50 |
>> |
| 51 |
>> I think we'll have to handle the evaluated strings anyway since this |
| 52 |
>> code has already been released and stabilized in portage-2.1.8.x, |
| 53 |
>> and USE conditionals have been evaluate in $PKGDIR/Packages for even |
| 54 |
>> longer. Because of this, I see little or no benefit in changing it |
| 55 |
>> back to unevaluated strings at this point. |
| 56 |
> |
| 57 |
> Good. Thanks for not reverting back to those old behaviors. |
| 58 |
|
| 59 |
And the new use case isn't of any relevance? |
| 60 |
|
| 61 |
As a compromise: What about storing both values? |
Archives