Cool. I just wanted to make sure the base of your diff was a valid revision. Your git diff command with three dots will generate a diff of your HEAD against that revision.
That's definitely a valid revision, it's still no develop (no one forced pushed it!) And I was able to create a review request with the webui using your command and a branch based on that revision. One more stupid question - you are creating the request against the geode repository, right? I don't think it would be that bad to submit a PR, actually. I've been wondering if we should start using PRs instead of review requests so it's easier for contributors to make the transition to committers. PRs also have the nice property of commenting on related tickets, running travis, etc. -Dan On Thu, Dec 8, 2016 at 2:00 PM, Kirk Lund <kl...@pivotal.io> wrote: > I'm not sure what it should be but here's the output... > > $ git merge-base HEAD origin/develop > ef7418140b903e73bc3a38d3a86c01349924bd09 > > > On Thu, Dec 8, 2016 at 12:49 PM, Dan Smith <dsm...@pivotal.io> wrote: > > > What does git merge-base HEAD origin/develop tell you? That's the > revision > > it's generated a diff against. You could try fetching from origin and > > rebasing your branch on origin/develop to help. > > > > -Dan > > > > On Thu, Dec 8, 2016 at 9:23 AM, Kirk Lund <kl...@apache.org> wrote: > > > > > I'm unable to post any reviews for Apache Geode on > > > https://reviews.apache.org > > > > > > I've updated .git/config in my checkout to match the non-incubator URL > > for > > > asf git. I've also rebased on latest origin/develop, but every time I > try > > > to upload my diff file, I get this error... > > > > > > The specified diff file could not be parsed. > > > Line undefined: error: unable to find > > > 390fb894c9a9095c87284f7f4389b2aeda090408 fatal: git cat-file > > > 390fb894c9a9095c87284f7f4389b2aeda090408: bad file > > > > > > I generated my diff for feature/GEODE-1027 with this command: > > > > > > $ git diff --full-index origin/develop...HEAD > GEODE-1027.diff > > > > > > Any ideas what's wrong? > > > > > > I could switch to using PRs to review my changes, but that's silly. > > > > > > -Kirk > > > > > >
