[Qgis-developer] Failing tests

[Matthias Kuhn]

Hi Jürgen

On 06/02/2015 09:03 AM, Jürgen E. Fischer wrote:
> Hi Nyall,
>
> On Tue, 02. Jun 2015 at 13:34:56 +1000, Nyall Dawson wrote:
>> I'd be in favour of a hardline "no breaks allowed" approached. You break
>> travis, your commit gets reverted.
> -1
>
> A general rule doesn't make sense to me - it depends on the break.
> If you break travis, you should fix it.

For the record, can you elaborate which types of breaks should not be
fixed and/or why certain breaks should be fixed in master instead of in
a separate branch/pull request?

Regards
Matthias