Robert Haas wrote:
> Since Itagaki Takahiro is now a committer, I sort of assumed he would
> be committing his own patches.
Maybe, but I wasn't going to be the one to suggest that Tom get cut out
of the loop after he raised a list of issues with the patch already.
I think the situation for EXPLAIN BUFFERS is much simpler, given that
the last round of feedback was only quibbling over the line formatting
and docs. What I think is a reasonable general guideline is to route
submitted patches back to their author to commit only when the patch has
been recently free of code issues during its review. If we've already
had another committer chime in with concerns, they should probably
confirm themselves that the updated version is suitable to commit, and
do so instead of the author. That just seems like a reasonable
risk-reduction workflow approach to me, similar to how the "sign-off"
practice works on some other projects.
--
Greg Smith 2ndQuadrant Baltimore, MD
PostgreSQL Training, Services and Support
greg@2ndQuadrant.com www.2ndQuadrant.com