Derek Chen-Becker <[email protected]> writes: >> I guess the question is who is the target auditory for these hooks. >> Should it be occasional contributors? Should it be regular contributors? >> If the former, we should include the hooks as a part of make all. >> If the latter, maybe not. > > My humble opinion is to make it opt-in and give people instructions for > installation via a `githook' target instead of making it automatic for > everyone. I want these to be a useful tool for folks, but not necessarily a > requirement, since we validate the commit message as part of patch review > anyway (we can always mention them in reviews if we see problems with the > message)
Yeah. After thinking more about this, it is probably disruptive to force occasional contributors run the hooks. At the end, the overall goal is making contributions easier, not harder. For people who contribute regularly, the hooks might be of use though. So, opt-in sounds like a better idea. However, if it is opt-in, the hooks should certainly be documented in org-contribute.org on WORG. A simple dedicated target that needs to be built manually will do. >> P.S. I tested the hooks a bit, and it looks like they have >> false-positives when inserting long Urls - complain about "too long >> word" and "too long line". > > I can either disable those or we can try to tune them to better align with > Org's rules for line length. Are you saying that they're too short, or that > long URLs should be exempt from line length requirements? Long Urls should be exempt. Example: https://cgit.git.savannah.gnu.org/cgit/emacs/org-mode.git/commit/?id=2732be62368f586b36fd9cb894d124ff521f64dd You can see that message IDs can get pretty long and there is no practical reason to reject commit messages like in this commit - such long URL cannot be wrapped. -- Ihor Radchenko // yantar92, Org mode maintainer, Learn more about Org mode at <https://orgmode.org/>. Support Org development at <https://liberapay.com/org-mode>, or support my work at <https://liberapay.com/yantar92>
