| 1 |
On Tue, 28 May 2013 16:43:10 +0200 |
| 2 |
David Carlos Manuelda <stormbyte@×××××.com> wrote: |
| 3 |
|
| 4 |
> I posted a bug about that along with a suggestion, despite sometimes I do not |
| 5 |
> explain myself correctly (I am very sorry): bug #471590 |
| 6 |
> |
| 7 |
> Many packages are bundling its own libraries rather than link against system |
| 8 |
> ones, and there is a bug tracker for that (bug #251464) |
| 9 |
> [...] |
| 10 |
> What I propose for example, is a very good and simple approach: to have an |
| 11 |
> option in portage's make.conf, something like that (the name may change): |
| 12 |
> |
| 13 |
> 1.- USE_SYSTEM_LIBRARIES="cairo sqlite XXX" |
| 14 |
> 2.- USE_SYSTEM_LIBRARIES="* -cairo" |
| 15 |
> 3.- USE_SYSTEM_LIBRARIES="*" |
| 16 |
|
| 17 |
I don't think we should do it like this. |
| 18 |
|
| 19 |
Bundling libraries is a pathological case. In general, we should work |
| 20 |
on fixing this and getting rid of bundled libraries. In that general |
| 21 |
case, the flags are not required. |
| 22 |
|
| 23 |
The remaining cases usually involve libraries that are somehow forked |
| 24 |
or otherwise deviated from upstream. For example, the sqlite within |
| 25 |
Mozilla products has a few braindead features enabled that are not |
| 26 |
really suitable for system-wide use. |
| 27 |
|
| 28 |
In those cases, every library needs to be considered separately |
| 29 |
and the concept of local flag matches that. Making it anyhow a global |
| 30 |
matter would suggest that each bundled sqlite is equal and each reason |
| 31 |
for bundling is exact the same. |
| 32 |
|
| 33 |
That said, I don't see why a user would want to choose to have some |
| 34 |
libraries bundled, and others not in a global manner. |
| 35 |
|
| 36 |
-- |
| 37 |
Best regards, |
| 38 |
Michał Górny |
Archives