What's up with pull request buildbots?

monarch_dodra monarchdodra at gmail.com
Tue Jun 4 14:34:30 PDT 2013


On Tuesday, 4 June 2013 at 17:47:36 UTC, Jonathan M Davis wrote:
> On Tuesday, June 04, 2013 09:52:57 Robert wrote:
>> Pull requests should be closed by maintainers if the pull 
>> request won't
>> get accepted even if improved quality wise or if the submitter 
>> is no
>> longer available for making it ready, but not just because 
>> they are old.
>
> I think that the problem is more that the Phobos maintainers 
> are busy and
> don't necessarily do a good job as a group of reviewing pull 
> requests in
> general, let alone making sure that each and every pull request 
> gets reviewed
> and processed in a reasonable time frame. It's something that 
> we need to work
> on. I think that it's rarely the case that a particular pull 
> request has no
> comments on it because it's undesirable. It just hasn't been 
> looked at yet
> like it should have been yet. If it were truly undesirable, it 
> would at
> minimum have comments saying so if not have been closed.
>
> - Jonathan M Davis

Truth. I've had a pull closed in less than an hour once. On the 
other hand, I have one which has been open for about 7 months now.

It's all a balance of importance vs resource/cost ratio, and 
reviewer interest. I've closed a few of my own ("valid") pulls 
myself: they weren't bad, just that I judge reviewer times would 
be better spent on more important things.

It's a bit frustrating at times, but I find it more than balances 
when you realize that you actually *get* to participate at all, 
or even partake in specs discussion...



More information about the Digitalmars-d mailing list