[ovs-dev] [Single DP 09/15] tests: Rewrite unit tests to not expect bridge with odp zero.

[Justin Pettit]

On Oct 22, 2012, at 3:21 PM, Ben Pfaff <blp at nicira.com> wrote:

> "git am" says:
> 
>    Applying: tests: Rewrite unit tests to not expect bridge with odp zero.
>    /home/blp/ovs/.git/rebase-apply/patch:2440: trailing whitespace.
>      [add-br test-br0], 
>    /home/blp/ovs/.git/rebase-apply/patch:2472: trailing whitespace.
>    AT_CHECK([RUN_OVS_VSCTL([list b], [get b test-br0 _uuid])], 
>    /home/blp/ovs/.git/rebase-apply/patch:2526: trailing whitespace.
>         add bridge test-br0 external_ids '"roles"="local; remote; cloud"'])], 
>    /home/blp/ovs/.git/rebase-apply/patch:2529: trailing whitespace.
>    AT_CHECK([RUN_OVS_VSCTL_ONELINE([get bridge test-br0 other_config external-ids])], 
>    /home/blp/ovs/.git/rebase-apply/patch:2533: trailing whitespace.
>    AT_CHECK([RUN_OVS_VSCTL([get bridge test-br0 other_config:hwaddr -- --if-exists get bridge test-br0 other-config:nonexistent])], 
>    warning: squelched 21 whitespace errors
>    warning: 26 lines add whitespace errors.

Okay.  Should be better now.

> This isn't a particularly elegant solution, but it's only
> incrementally less elegant than what came before it, and it's only
> test code so I don't feel too bad about it.

Agreed.

> If there's one change that I'd make, it's to just get rid of the
> "test-" prefix everywhere.  netdev-dummy is only good for testing
> anyway and I wouldn't feel bad about special-casing all br* devices
> there.  This would also make the diff slightly smaller.  But I don't
> feel bad about keeping it, either, if you don't want to deal with it.

Yes, that makes it a much smaller patch.  I didn't do it originally, because I wasn't sure we'd want to be special-case such a common name.  However, since you're fine with it, I just went ahead and changed it.

--Justin