1 |
W dniu czw, 29.03.2018 o godzinie 11∶34 -0500, użytkownik William Hubbs |
2 |
napisał: |
3 |
> On Thu, Mar 29, 2018 at 12:24:38PM -0400, Michael Orlitzky wrote: |
4 |
> > On 03/29/2018 11:57 AM, William Hubbs wrote: |
5 |
> > > > |
6 |
> > > > The PMS says that empty directories are undefined, so the portage |
7 |
> > > > behavior of installing them and leaving them alone leads to |
8 |
> > > > incompatibilities. Ebuilds rely on the portage behavior, and if another |
9 |
> > > > PM (within its rights) deletes them, then the package breaks with the |
10 |
> > > > non-portage PM. |
11 |
> > > |
12 |
> > > Maybe so, but you just made the argument for doing nothing different in |
13 |
> > > current eapis and proposing stripping empty directories in eapi 7. |
14 |
> > > However, this should be stripping empty directories combined with |
15 |
> > > failing the emerge. |
16 |
> > |
17 |
> > If we strip them only in EAPI=7, then that still leaves all of these |
18 |
> > packages broken with respect to the PMS in the other EAPIs. |
19 |
> > |
20 |
> > Stripping the empty directories isn't my favorite approach, but leaving |
21 |
> > things broken looks bad on paper too. |
22 |
> |
23 |
> If we are going to strip the empty directories, we should hard fail the |
24 |
> emerge at the same time. Otherwise there is no way to know whether the |
25 |
> package we successfully install will now run. |
26 |
> |
27 |
|
28 |
The developer is supposed to *look* at what the package installs. |
29 |
If people just commit ebuilds based on 'Portage did not explode', |
30 |
then maybe they shouldn't have commit access in the first place. |
31 |
|
32 |
-- |
33 |
Best regards, |
34 |
Michał Górny |