Shame on me that I never cared to look at common/Makefile.common.
There are a lot of cool things there. :-)
However, since we already have 'make clog', what is stopping us from
having a 'make commit'? In my limited view, if one is creating a clog
then he is almost certain to commit to the VCS. It is almost as if
clog is redundant. Why would I want to have the topmost entry of the
changelog in a separate file, even when I have the entire changelog in
the Spec, and not make a commit?
So modifying Colin's original use case scenarios:
edit foo.spec
make clog
cvs commit -F clog
make tag build
# discover error
edit foo.spec
make force-tag build
Now:
edit foo.spec2
make build
# discover error
edit foo.spec2
make build
...we can have:
edit foo.spec
make commit tag build
# discover error
edit foo.spec
make commit force-tag build
Till now we have assumed that every commit will have a new changelog
entry which can be associated with that commit. Have we considered the
odd commit that a packager might make to fix some trivial issue (eg.,
tabs/spaces) that does not warrant a tag, build, update or a separate
entry in the %changelog?
Maybe we can have a 'make fix' for that? I don't know.
Happy hacking,
Debarshi
--
"From what we get, we can make a living; what we give, however, makes a life."
-- Arthur Ashe