If there are things missing from Renovate that mean we can't configure the commit messages in the way you'd like that's a shame, but I'm not totally sure that needs to be a blocker here. Does the change just need to go to a vote?
--EM On Thu, Jul 10, 2025 at 3:28 AM Robert Stupp <sn...@snazy.de> wrote: > That would remove the useful information like links > release-notes/changelog, effective Renovate config for the PR, ability > to re-create the PR, link to Renovate logs, etc > > On Wed, Jul 9, 2025 at 6:25 PM Eric Maynard <eric.w.mayn...@gmail.com> > wrote: > > > > Is it possible to adjust the PR description generated by renovate to > better > > align with what you're looking for? The docs > > <https://docs.renovatebot.com/configuration-templates/#pr-body> seem to > > suggest that it can be done with e.g. {{{commitList}}} > > > > --EM > > > > On Wed, Jul 9, 2025 at 9:08 AM Robert Stupp <sn...@snazy.de> wrote: > > > > > My biggest concern is the Renovate PRs. The commit message of those is > > > suitable for being merged "as is" - but the PR description is not. > > > Given the amount of Renovate-PRs it's a lot of cognitive work to > > > adjust those - or the commit log would look really really awful with a > > > ton of not useful information and links - leaking into (automatically > > > generated) changelogs for releases. > > > > > > All "human originated" PRs need manual attention anyway. > > > > > > > > > On Wed, Jul 9, 2025 at 5:54 PM Eric Maynard <eric.w.mayn...@gmail.com> > > > wrote: > > > > > > > > Hi Robert, > > > > > > > > > Having the list of commits in the merge commit message proposed by > > > GitHub > > > > helps there. > > > > > > > > Can you elaborate on this? How exactly does it help? You can always > view > > > > the commits on the original PR. > > > > > > > > Beyond that, I actually don't see these commit messages being added > > > > currently when I look at recently merged PRs like 1 > > > > < > > > > https://github.com/apache/polaris/commit/a0e31b0584437602f30a97ed5fc110c50802d9c1 > > > > > > > > or 2 > > > > < > > > > https://github.com/apache/polaris/commit/a0e31b0584437602f30a97ed5fc110c50802d9c1 > > > >. > > > > They are indeed there by default right now, as you can see in this > recent > > > > commit > > > > < > > > > https://github.com/apache/polaris/commit/145ad3e2715ccb4630e6537de3c077561c86a129 > > > > > > > > which > > > > includes each commit's message. But my reading of your linked > comments > > > > suggests that committers are meant to *change* the default commit > > > message. > > > > > > > > Ajantha's proposed change would make the default commit message more > > > > closely resemble what you suggest. In fact, for those two linked > PRs, it > > > > looks like the default commit message would have been exactly the > same as > > > > what ultimately got merged. > > > > > > > > --EM > > > > > > > > On Wed, Jul 9, 2025 at 3:09 AM Jean-Baptiste Onofré <j...@nanthrax.net > > > > > wrote: > > > > > > > > > Hi > > > > > > > > > > From technical standpoint, the commit template can be updated via > > > asf.yaml. > > > > > It’s a change we did on the parser a few months ago. > > > > > > > > > > Regards > > > > > JB > > > > > > > > > > Le mer. 9 juil. 2025 à 04:30, Ajantha Bhat <ajanthab...@gmail.com> > a > > > > > écrit : > > > > > > > > > > > Hi all, > > > > > > While adding the recommendation to working with the PR on site > docs, > > > > > > we have opened an issue > > > https://github.com/apache/polaris/issues/1656 to > > > > > > update the settings. > > > > > > > > > > > > I have discussed with ASF Infra and they mentioned, we can > update by > > > > > > ourselves. > > > > > > Hence, I opened a PR for the same: > > > > > > https://github.com/apache/polaris/pull/2016 > > > > > > > > > > > > But it looks like we don't have consensus on this change (based > on > > > the > > > > > > comment on the PR). Hence this discussion. > > > > > > > > > > > > I personally think it is good to have these settings as default > as > > > the > > > > > > committer work will be reduced if the PR title and description is > > > proper. > > > > > > I don't see the drawbacks of this approach yet. Please feel free > to > > > > > mention > > > > > > your opinion on this. > > > > > > > > > > > > - Ajantha > > > > > > > > > > > > > > >
