On Mon, Jun 10, 2013 at 2:48 AM, Ramkumar Ramachandra
> Felipe Contreras wrote:
>> git-rebase.rb | 2056
>> 1 file changed, 2056 insertions(+)
>> create mode 100755 git-rebase.rb
> I suggest putting this in contrib/ and cooking it. As usual, my
> mantra is: let the patches decide what to do. I'll help review and
> improve this soon.
What would be the point? There's only so much this script could do. I
was already relying on an improved 'git cherry-pick' that's never
going to happen.
I think it would be more interesting to fork Git, replace all the main
builtin commands with Ruby scripts that access libgit.a, and copy the
code to a libgit-ng library that has stuff that is worthy of being on
a library until the scripts don't access libgit.a any more. All the
time passing all tests in the test framework of course.
Then there would be a truly useful and stable Git library, and we
would have scripts that work on all platforms, including Windows,
without any forks, and the commands would be more maintainable, and
would gather the potential of so many Ruby developers, which if GitHub
is any indication; a lot of them love Git.
I think that's the only way forward, since the Git project doesn't
wish to be improved.
There's only one last patch series that I'll try to push to Git that
I've been cooking for years. Sadly, I don't think it's going to land,
despite it being clearly good, and a simple single patch doing the
trick, and it would be immensely useful for our users, who have
complained about that for years, and most Git developers have agreed
it needs to be done. Yet, if I send it, it would not land.
Perhaps it's time for me to create a pseudonym, and when you have to
do that to land clearly good patches, you know something is *REALLY*
wrong with the project.
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html