0 for PR - I feel that both have their benefits and downsides.
* +1 for PR: As Naba stated, the integration is simplified and using
Intellij there is even support to create a PR. Single system to rule
them all.
* -1 for PR: I fear that given the huge amount of incoming PR's we
might loose sight/oversight of any PR's that need to be approved
from external submitters. It is impossible for external submitters
to know who needs to be mentioned in order to have the PR looked at.
With the Reviewboard + Apache Geode PR system, those can be kept
separate and some oversight is possible.
On 6/12/17 11:13, Nabarun Nag wrote:
+1 On using PR. I feel using github PR is more convenient because of the
following reasons
a. Commit messages are automatically made into the header and description
in PR, whereas in reviewboard, you upload the patch and do all the writing
- manually. It feels redundant.
b. After changing the code as per some review comments, we need to upload
the new patch back to reviewboard but in case of a PR we just need to do a
git push.
c. A single system for review for all types of contributors - committers
and non committers.
d. Travis CI runs on the PR - detects missed out spotlessApply or merge
issues.
e. Also in case someone is AFK for a long time, one can request a fellow
committer to merge the PR without losing commit credit.
After some google search, I could see that a lot of projects are moving to
Github PR system over Review Board. Apache Spark, Apache CloudStack ,
Apache SystemML are a few examples.
Regards
Nabarun
On Mon, Jun 12, 2017 at 9:37 AM Dave Barnes <dbar...@pivotal.io> wrote:
+0
Proposal as written says "...for changes that would require peer review
before committing...".
If this means that minor changes (in my case, typo repairs are the common
case) can be checked in without going through a PR process, I can go with
the group decision.
Still see no compelling need to retire the Review Board, which (as pointed
out in the discussion thread) has its uses and advantages.
On Mon, Jun 12, 2017 at 7:57 AM, Bruce Schuchardt <bschucha...@pivotal.io>
wrote:
-1
It places an unnecessary burden on committers and git history is the
definitive record of changes to the code so github pr history isn't
really
very useful. Also, Review Board is a much better tool for reviewing code
than a github PR.
On 6/11/17 9:51 AM, Jacob Barrett wrote:
After a few days of discussion [1] this thread has quieted down so I
would
like to take it to a vote. The proposal is to modify the workflow of
committers to match that of non-committers such that committers shall:
* Perform all work in progress on branches in their personal forks on
GitHub rather than directly on the ASF Geode repositories.
* Submit GitHub pull requests, following all current rules for pull
requests, for changes that would require peer review before committing
to
the production branches.
* Register their GitHub accounts with the ASF so that committers can be
identified in the GitHub mirror repositories.
Please cast your vote.
[ ] +1 Approve
[ ] +0 No opinion
[ ] -1 Disapprove (and reason why)
[1] [DISCUSS] Using Pull Requests over Review Board
<http://geode.markmail.org/message/row5hgz5zw2ooadl?q=list:
org%2Eapache%2Egeode%2Edev+discuss+pull>
-Jake
