Noam Postavsky <npostavs@gmail.com> writes:

> It's best to avoid using hashes in commit messages, as they're
> translated to ChangeLog files which might read from the tarball (i.e.,
> without a git repo to hand).  CONTRIBUTE talks about using "action
> stamps" but I think date+title is more readable.  Which would be:
> 2019-04-29 "Refrain from splicing anonymous faces in text properties".

Ah, right, didn't think of the children^WChangeLog.  I went for
date+title.
  
Can e.g. git-show(1) understand action stamps?  I glanced at
gitrevisions(7) but nothing suggests Git knows anything about this
format.  It sure would make a hypothetical vc-revision-at-point command
easier to implement…

(Or, going the opposite route, maybe the git-log-to-ChangeLog machinery
could translate hashes to action stamps?)

>> +(provide 'font-lock-tests)
>
> I don't think there is any reason to `provide' a feature in a test file
> (I know some other test files do that, but I don't see why), test files
> are generally not loaded via require.

Done.