Subject: Re: [boost] Git Modularization Ready for Review
From: Dave Abrahams (dave_at_[hidden])
Date: 2013-05-22 01:26:01
on Tue May 21 2013, Vladimir Prus <ghost-AT-cs.msu.su> wrote:
> On 20.05.2013 10:08, Dave Abrahams wrote:
>>> I would be fine with /dev/null, to be honest, unless Rene and Steven have any
>>> concerns. That is part of the history that is either not used at all, or has been
>>> rewritten over and over, so it won't be much useful in future.
>> You're free to delete any branches you want after the conversion is
>> over (it's really easy), but until then, we want to make sure every
>> commit is accounted for. Any commits that aren't caught by the ruleset
>> end up in https://github.com/boostorg/svn2git-fallback and the build at
>> http://jenkins.boost.org fails.
>> You can make an arbitrarily obscure branch name
>> (e.g. old-branches/deleteme) and send unwanted history there, if you
>> like. Please try to submit a pull request with edits to
>> repositories.txt. An explanation of the semantics is at:
> I have spent around 30 minutes trying to grok this format, and I cannot
> quite figure how how to sent tools/build/v2 directory to master branch
> of "build" repo, while sending everything else from trunk/tools/build
> into some other branch.
That doesn't sound like a very accurate reflection of history. There
used to be code in tools/build/, directly. What "other branch" should
that go into, and at what path would you like it to be placed?
And there are lots of changes outside trunk/ that should get similar
treatment, whatever you decide to do. See:
One possibility is that you keep everything under tools/build (other
than tools/build/v2) in the branches where it currently resides, but put
it in a subdirectory called v1/ rather than at the top level. However,
there's probably a point in history where that stuff would *belong* at the
top level because, e.g., v2 didn't exist.
> If the change is easy, as you make it sound, can either you or Daniel
> make it?
We could... but I am a little concerned about making changes to history
that diverge too far from reality. Remember, you are free to move
things around and rename branches after modularization is
complete... but I'm not sure we should do too much revision of the
actual history. Thoughts?
-- Dave Abrahams
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk