Is there a reason the patched version (thanks) is going through [testing]? We now have a package that causes problems in [core] and a version that doesn't cause these problems and for the rest is not worse in [testing].

Just curious, - Jouke

On Sat, Nov 16, 2013 at 11:19 PM, Jouke Witteveen j.witteveen@gmail.com wrote:

Although it slipped through review on the list and made it past [testing], there is a bug in netctl 1.4 that can prevent automatically generated wireless profiles from working. The bad commit was f5065. I propose adding commit 3dc6d, which fixes the issue, directly as a patch to the package in [core].

WPA profiles generated with `wifi-menu -o` will stop working in 1.4 because the detection of hex keys looks at $key instead of at $Key (capital K). The commit referenced above fixes this issue. If we apply it quickly, we might have it in before the majority of users has upgraded to netctl 1.4.

Sorry,

• Jouke

No exceptions? In this case, it is not very sensible, is it?

To me, these questions are no longer about netctl 1.4-2. I even think it is a very good rule to have all [core] packages go through [testing] first. However, there might be cases where it is not what you'd want (of which this incarnation of netctl might be one).

On Mon, Nov 18, 2013 at 10:49 AM, Allan McRae allan@archlinux.org wrote:

On 18/11/13 19:47, Jouke Witteveen wrote:

...

Is there a reason the patched version (thanks) is going through [testing]? We now have a package that causes problems in [core] and a version that doesn't cause these problems and for the rest is not worse in [testing].

Just curious,

• Jouke

All [core] packages MUST go through [testing].

Allan