Exactly. Stef was referring to the normalize function here: https://github.com/cockpit-project/cockpit/blob/0a0250e8118c47ced529772758d49d6c5a60924f/pkg/users/test-list-public-keys.sh#L8

Parsing text output is always problematic and has been an issue in the past, hence the normalization function.

Maybe you can adjust it to work with the new output.

-Dominik

On 02/01/2016 09:32 PM, Jared Sutton wrote:
It looks like "ssh-keygen -l -f <authorized_keys_file>" produces different output on wily (expected output) vs vivid (unexpected output)...

Inline image 1

On Mon, Feb 1, 2016 at 9:49 AM, Stef Walter <stefw@fedoraproject.org> wrote:
Looks like the tests produce different output than are expected.

Perhaps some changes are necessary to the normalize() function in test-list-public-keys.sh file?
_______________________________________________
cockpit-devel mailing list
cockpit-devel@lists.fedorahosted.org
https://lists.fedorahosted.org/admin/lists/cockpit-devel@lists.fedorahosted.org



_______________________________________________
cockpit-devel mailing list
cockpit-devel@lists.fedorahosted.org
https://lists.fedorahosted.org/admin/lists/cockpit-devel@lists.fedorahosted.org