1 |
On Thu, Mar 29, 2018 at 12:24:38PM -0400, Michael Orlitzky wrote: |
2 |
> On 03/29/2018 11:57 AM, William Hubbs wrote: |
3 |
> >> |
4 |
> >> The PMS says that empty directories are undefined, so the portage |
5 |
> >> behavior of installing them and leaving them alone leads to |
6 |
> >> incompatibilities. Ebuilds rely on the portage behavior, and if another |
7 |
> >> PM (within its rights) deletes them, then the package breaks with the |
8 |
> >> non-portage PM. |
9 |
> > |
10 |
> > Maybe so, but you just made the argument for doing nothing different in |
11 |
> > current eapis and proposing stripping empty directories in eapi 7. |
12 |
> > However, this should be stripping empty directories combined with |
13 |
> > failing the emerge. |
14 |
> |
15 |
> If we strip them only in EAPI=7, then that still leaves all of these |
16 |
> packages broken with respect to the PMS in the other EAPIs. |
17 |
> |
18 |
> Stripping the empty directories isn't my favorite approach, but leaving |
19 |
> things broken looks bad on paper too. |
20 |
|
21 |
If we are going to strip the empty directories, we should hard fail the |
22 |
emerge at the same time. Otherwise there is no way to know whether the |
23 |
package we successfully install will now run. |
24 |
|
25 |
William |