1 |
On 15/06/2014 20:27, Mike Gilbert wrote: |
2 |
> On Sat, Jun 14, 2014 at 7:31 PM, Alan McKinnon <alan.mckinnon@×××××.com> wrote: |
3 |
>> I'm at a loss to explain why this is a good idea or desirable: |
4 |
>> |
5 |
>> !!! The ebuild selected to satisfy |
6 |
>> ">=net-libs/gupnp-0.18[abi_x86_32(-)?,abi_x86_64(-)?,abi_x86_x32(-)?,abi_mips_n32(-)?,abi_mips_n64(-)?,abi_mips_o32(-)?]" |
7 |
>> has unmet requirements. |
8 |
>> - net-libs/gupnp-0.20.12-r1::gentoo USE="introspection -connman |
9 |
>> -networkmanager" ABI_X86="64 -32 -x32" PYTHON_TARGETS="python2_7 -python2_6" |
10 |
>> |
11 |
>> The following REQUIRED_USE flag constraints are unsatisfied: |
12 |
>> exactly-one-of ( connman networkmanager ) |
13 |
>> |
14 |
>> |
15 |
>> |
16 |
>> USE="upnp" is pulling this in, the major user of that flag here is |
17 |
>> farstream. |
18 |
>> |
19 |
>> Maybe I want farstream. Maybe I want upnp for it. |
20 |
>> Maybe I don't care for connman or networkmanager (I use wicd, some users |
21 |
>> want no nw manager at all) |
22 |
>> |
23 |
>> So, what is the sense behind a feature of an instant messenger framework |
24 |
>> causing me to decide between two undesirable connection managers? |
25 |
>> |
26 |
>> Anyone see some valid logic that I miss? |
27 |
>> |
28 |
> |
29 |
> I'm pretty sure this was a simple developer error. |
30 |
> |
31 |
> REQUIRED_USE="^^ ( connman networkmanager )" should have been |
32 |
> REQUIRED_USE="?? ( connman networkmanager )". |
33 |
> |
34 |
> The former requires that exactly one flag be enabled. The latter |
35 |
> requires that at most one flag may be enabled. |
36 |
> |
37 |
> |
38 |
> |
39 |
|
40 |
|
41 |
You are probably correct, I filed a bug: |
42 |
|
43 |
https://bugs.gentoo.org/show_bug.cgi?id=513310 |
44 |
|
45 |
-- |
46 |
Alan McKinnon |
47 |
alan.mckinnon@×××××.com |