| 1 |
Dnia 2013-12-18, o godz. 22:58:07 |
| 2 |
heroxbd@g.o napisał(a): |
| 3 |
|
| 4 |
> Hey, Michał, |
| 5 |
> |
| 6 |
> Michał Górny <mgorny@g.o> writes: |
| 7 |
> |
| 8 |
> > a) adding USE=c++11 and USE-deps to all the packages in question, |
| 9 |
> |
| 10 |
> I think it is better achieved by a (simple and stupid) global |
| 11 |
> CXXFLAGS. Adding an extra USE flag feels a little over-engineering. |
| 12 |
|
| 13 |
This is nowhere near a good solution IMO. |
| 14 |
|
| 15 |
First of all, it doesn't give us a way of ensuring ABI compatibility. |
| 16 |
Users switch the flags and have to rebuild all C++ packages to regain |
| 17 |
the ABI compatibility. The system ends up borked quite easily. |
| 18 |
|
| 19 |
Then, we don't have a good way of finding packages to rebuild. Users |
| 20 |
could try to find out which libraries used C++ but well... it's nowhere |
| 21 |
near good. Or they just rebuild everything... |
| 22 |
|
| 23 |
Then, many developers just won't bother. Users will be the ones to hit |
| 24 |
the incompatible package build failures first. |
| 25 |
|
| 26 |
Lastly, this gives us no way of switching to C++11 completely without |
| 27 |
modifying the compiler defaults. Even if we put '-std=c++11' into |
| 28 |
profiles, most of the people override CXXFLAGS and won't have it. |
| 29 |
|
| 30 |
> Any anyway, if it is only for lldb, a piece of elog conveying a |
| 31 |
> preferred solution would suffice. |
| 32 |
|
| 33 |
elog? I think you mean dying with CXXFLAGS that don't specify |
| 34 |
the necessary standard. Which is kinda backwards to REQUIRED_USE... |
| 35 |
|
| 36 |
And then, simple CXXFLAGS solution would end up breaking users' |
| 37 |
systems... |
| 38 |
|
| 39 |
-- |
| 40 |
Best regards, |
| 41 |
Michał Górny |
Archives