1 |
On 06/06/2012 01:28 AM, Pacho Ramos wrote: |
2 |
> El mar, 05-06-2012 a las 16:07 -0700, Zac Medico escribió: |
3 |
>> The "SLOT operator" dependencies that Ciaran has been advocating are |
4 |
>> very close to a good solution. However, if we want it to work with |
5 |
>> unslotted packages, then we need to introduce a separate ABI_SLOT |
6 |
>> variable as discussed here: |
7 |
>> |
8 |
>> https://bugs.gentoo.org/show_bug.cgi?id=192319#c18 |
9 |
>> |
10 |
>> It's really no more difficult to do than "SLOT operator" dependencies, |
11 |
>> it's more flexible, and we can do it in EAPI 5. |
12 |
> |
13 |
> In that case, I obviously wouldn't have any problem with that approach |
14 |
> (it sound even better :)). Is there any place where I could get a bit |
15 |
> more documentation about how this "SLOT operator" way would work? For |
16 |
> example, how would work for rebuilding x11 drivers after updating xorg |
17 |
> or rebuilding gobject-introspection after major glib update... |
18 |
|
19 |
Whenever you have an ABI change, the developer doing the version bump |
20 |
needs to increment the SLOT (or ABI_SLOT if we use a separate variable) |
21 |
in the package. Packages that depend on the package with the ABI change |
22 |
(reverse dependencies) append a := operator to their dependency atoms, |
23 |
indicating that they are locked to the ABI of the SLOT that they are |
24 |
built against. The package manager translates the := operators into a |
25 |
dependencies on specific SLOTs at build time, so that when you update |
26 |
your system next time, it can use this information to trigger rebuilds |
27 |
automatically when necessary. |
28 |
-- |
29 |
Thanks, |
30 |
Zac |