| 1 |
On Mon, 2021-09-27 at 21:03 -0400, Mike Gilbert wrote: |
| 2 |
> On Mon, Sep 27, 2021 at 1:20 PM Michał Górny <mgorny@g.o> wrote: |
| 3 |
> > |
| 4 |
> > Warn the developers if ebuilds install files with xattrs to ${ED}. |
| 5 |
> > The xattrs may or may not be preserved when installing the package, |
| 6 |
> > making them unreliable on one hand, and somewhat suprising in other |
| 7 |
> > cases (e.g. when they unintentionally leak from developer's system). |
| 8 |
> > |
| 9 |
> > This is the first step towards restoring PMS compliance and *not* |
| 10 |
> > preserving extended metadata. |
| 11 |
> |
| 12 |
> How does preserving xattrs conflict with PMS? |
| 13 |
|
| 14 |
The PMS doesn't specify that xattrs, ACLs, caps etc. are preserved. |
| 15 |
By doing that, Portage allows developers to commit ebuilds that are not |
| 16 |
going to work reliably without even realizing it. In fact, this can't |
| 17 |
even work reliably inside Portage itself, depending on the filesystem |
| 18 |
used for $D. |
| 19 |
|
| 20 |
Furthermore, doexe preserving stuff goes contrary to common sense. Why |
| 21 |
would helpers preserve xattrs when they are supposed to reset things |
| 22 |
like mode and ownership by design? |
| 23 |
|
| 24 |
> Is there a bug report you could reference? |
| 25 |
|
| 26 |
It starts with https://bugs.gentoo.org/814857. |
| 27 |
|
| 28 |
|
| 29 |
|
| 30 |
-- |
| 31 |
Best regards, |
| 32 |
Michał Górny |
Archives