Review backlog

[Ashod Nakashian]

On Sat, Aug 15, 2015 at 3:42 PM, Thorsten Behrens <
thb at documentfoundation.org> wrote:

> speaking for myself - I tend to wait for jenkins builds to succeed
> before having any closer look, so I'd strive for that first (simply
> rebasing will trigger a new build, should there have been a broken
> master).
>
>
Thanks Thorsten for the response.

This is a point I'd like to address. At certain times jenkins acts up and
fails builds randomly.

If you look at my currently unreviewed patches, most didn't have a clean
build on first try (some not even on the 5th!).

I make every effort to submit only patches that fully build _and_ work. I
spend a tremendous amount of time to test and commit the bare minimum
change (sometimes not all changes are necessary to fix an issue and can be
a distraction in reviewing/bisecting/etc, so I remove them).

After frequently rebasing patched with failed builds, I realized that the
fact that I'm rebasing gives others the impression that it's a work in
progress, and they skip reviewing (probably waiting it to settle down).

So after numerous retries, I gave up and let the patches wait for at least
a first reviewer to comment and give feedback.

Another issue, is that for one review I had a +2 review and flattering
comment, I was disappointed to realize that just rebasing (without any
changes to the patch) clears the review status!

Finally, I have 3 self-contained, unreleated, patches with successful
builds that still await the attention of a reviewer.


Also, if your patches depend on each other, push them for review in
> one go (gerrit detects the depend on each other). This gives context
> to the reviewer, and also avoids frustration when pulling one patch,
> and realizing it does not build etc.
>

Ironically, my 4-commit patch that implements a highly requested feature in
Writer got 2 questions implying that they might be separate works in
progress.
I explained they were broken up to make reviewing easy (one is UI changes,
the other options, etc) and the response was that it made sense. (Still no
review comments.)

So again I'm confused: how should I make it clear that my patches aren't
experiments rather they are of reasonably high-quality and ready for
serious review?


For patches where seemingly all is in order, a reviewer had already
> +1-ed it earlier etc - simply poke that person, she might be busy or
> distracted. You find a list of developers and their irc nicks in the
> wiki.
>
> Cheers,
>
> -- Thorsten
>

In the past I've had a much better response upon submitting patches, so I'm
inclined to think everyone is busy (which I highly appreciate,) even though
my patches are closer to 1 month old as I write this.

Thanks again,
Ash
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/libreoffice/attachments/20150815/808b4149/attachment.html>