1 |
Kevin F. Quinn wrote: |
2 |
> On Tue, 9 Jan 2007 23:23:55 +0000 |
3 |
> Ciaran McCreesh <ciaranm@×××××××.org> wrote: |
4 |
> |
5 |
>> On Tue, 09 Jan 2007 14:41:50 -0800 Zac Medico <zmedico@g.o> |
6 |
>> wrote: |
7 |
>> | Bug #161045 [1] requests that portage support RESTRICT=sandbox. |
8 |
>> | This is certainly a valid request but a user may wish to reject a |
9 |
>> | package based on certain questionable values of RESTRICT. |
10 |
>> |
11 |
>> If a RESTRICT value is questionable, it shouldn't be supported or |
12 |
>> used. |
13 |
>> |
14 |
>> Honestly, this strikes me as rather silly and rather dangerous. |
15 |
>> RESTRICT is not something about which the end user should have to |
16 |
>> know or care; it should be something entirely between ebuilds and the |
17 |
>> package manager. And sandbox is not something that should be turned |
18 |
>> off lightly; making it so easy will only encourage developers to take |
19 |
>> the nasty way out rather than fixing simple bugs. |
20 |
|
21 |
RESTRICT="fetch" is between the package, the manager, and the user (as |
22 |
someone has to fetch the files). |
23 |
|
24 |
RESTRICT="unattended" is also between the package, the manager, and the |
25 |
user, as the user has to insert CD's for certain packages (most in |
26 |
games). However unattanded is not currently implemented, only proposed. |
27 |
|
28 |
Maybe if we move these to a new metadata key or something, Looking over |
29 |
the other RESTRICT values they all seem to be (ebuild,PM) tuples, not |
30 |
anything the user would care about. |
31 |
|
32 |
|
33 |
-- |
34 |
gentoo-dev@g.o mailing list |