| 1 |
On 12/9/19 3:17 AM, Michał Górny wrote: |
| 2 |
> |
| 3 |
> 2. Mailing list reviews don't serve their original purpose. |
| 4 |
> |
| 5 |
> The original purpose of mailing list reviews was to verify that |
| 6 |
> the developers use new packages correctly. For example, Michael |
| 7 |
> Orlitzky has found a lot of unnecessary home directories specified. |
| 8 |
> Of course, that works only if people submit *ebuilds* for review. |
| 9 |
> |
| 10 |
> However, at some points developers arbitrarily decided to send only |
| 11 |
> numbers for review. This defeats the purpose of the review in the first |
| 12 |
> place. |
| 13 |
> |
| 14 |
|
| 15 |
For users like "git" and "ntp" that are shared by multiple packages, I |
| 16 |
think the mailing list review is still helpful. We fixed big problems |
| 17 |
with those two that only came to light during review. If somebody else |
| 18 |
requests a UID for "your" user, that gives you a chance to review their |
| 19 |
acct-user ebuild and make sure that it's compatible with your package. |
| 20 |
|
| 21 |
For esoteric packages with a dedicated user, though, you're probably |
| 22 |
right. The main benefit of the mailing list posts so far is that they |
| 23 |
let me track down pull requests and suggest that people ignore the |
| 24 |
example in the devmanual. |
| 25 |
|
| 26 |
If we can remove |
| 27 |
|
| 28 |
ACCT_USER_SHELL=/usr/bin/foo |
| 29 |
ACCT_USER_HOME=/var/lib/foo |
| 30 |
ACCT_USER_HOME_OWNER=foo:bar |
| 31 |
ACCT_USER_HOME_PERMS=0775 |
| 32 |
|
| 33 |
from [1], that would really help people guess the right thing to do the |
| 34 |
first time around. |
| 35 |
|
| 36 |
|
| 37 |
[1] https://devmanual.gentoo.org/ebuild-writing/users-and-groups/index.html |
Archives