Re: Policy concerning 'Patch::review'

[James]

Hello Michael,

On 15/02/2021 09:22, Michael Käppler wrote:
> Hi all,
> I'm not sure if I missed some policy change, but I noticed that recently
> some MRs were
> set to 'Review' without posting the test results. I assume that in such
> cases the tests were unchanged,
> but it would be good to clarify this.

There has never been a 'policy' to say you must state the test results 
(even if there is no change or nothing of any significance) it's just 
been a courtesy. Some developers will bother to put in the same 'nothing 
has changed, except the usual prog error events' but I saw no benefit in 
that.

Obviously if there is a change in the diffs someone will post it, but as 
I get so many emails myself, I figured just changing the label and the 
developers assuming that this means 'nothing of any significance' would 
at make the thread less noisy and perhaps not generate more emails.

This is similar reasoning that we assume no update to a thread by 
another developer infers approval (or at the very least no dis-approval) 
of a MR/Patch, so that it can move on, rather than have to wait for 
someone to OK a patch explicitly.

I check every MR/Patch when I do the countdown anyway, so if I spot 
something unusual or someone changing a label without stating a reason 
(over and above patch new -> review) then I'll question it.

What I would say while it is good form for another person to review the 
tests and change the label to review, it is not mandatory and the 
original developer could do it, for example, they fix a trivial typo or 
something like that and it builds and passes all the tests, there is no 
point (IMHO) to add yet another thread to the MR.

James