Private branch grafts

Private branch grafts

[Eric S. Raymond]

I have built and tested some machinery to solve the Aquamacs
problem. I am describing it on the Emacs list because other
owners of private branches will need to know how it works.

The new feature involved here is "callouts". They solve a problemm
with partial repository writes, in which commits contain mark
references for which the corresponding parent commits are not in the
selection set written out.

Previously, the write code would simply have dumped unaltered mark
values, some of which would become meaningless outside the original repo
context.  What reposurgeon now does in that case is replace any
unresolved marks with callouts (which are just action stamps
identifying a date and committer).

An import stream containing callouts cannot be loaded by
git-fast-import.  What can happen, though is this: when a stream with
callouts is grafted to another repo, the code tries to resolve all the
callout links in the context of that repo. It looks for a committer
and committer-date match for each callout and, in effect, replaces
it with the corresponding parent mark.

The match attempt (and the graft) will fail if there is no matching 
committer/committer-date pair in the repo, or more than one. In that
case (which I expect to be rare and might not occur at all) the 
offending callouts will need to be removed or hand-patched with ordinals,

What I will is locate each branch you want to preserve, write it
out as a partial dump with callouts, and graft iyit onto the main
repository in this way. 

-- 
		<a href="http://www.catb.org/~esr/">Eric S. Raymond</a>

The conclusion is thus inescapable that the history, concept, and 
wording of the second amendment to the Constitution of the United 
States, as well as its interpretation by every major commentator and 
court in the first half-century after its ratification, indicates 
that what is protected is an individual right of a private citizen 
to own and carry firearms in a peaceful manner.
         -- Report of the Subcommittee On The Constitution of the Committee On 
            The Judiciary, United States Senate, 97th Congress, second session 
            (February, 1982), SuDoc# Y4.J 89/2: Ar 5/5

Re: Private branch grafts

[David Reitter]

[-- Attachment #1: Type: text/plain, Size: 2500 bytes --]

Eric,
Thanks for your work.  Is there a version of the converted repository that we can inspect?

Thanks,
David

On Mar 9, 2014, at 5:32 PM, Eric S. Raymond <esr@thyrsus.com> wrote:

> I have built and tested some machinery to solve the Aquamacs
> problem. I am describing it on the Emacs list because other
> owners of private branches will need to know how it works.
> 
> The new feature involved here is "callouts". They solve a problemm
> with partial repository writes, in which commits contain mark
> references for which the corresponding parent commits are not in the
> selection set written out.
> 
> Previously, the write code would simply have dumped unaltered mark
> values, some of which would become meaningless outside the original repo
> context.  What reposurgeon now does in that case is replace any
> unresolved marks with callouts (which are just action stamps
> identifying a date and committer).
> 
> An import stream containing callouts cannot be loaded by
> git-fast-import.  What can happen, though is this: when a stream with
> callouts is grafted to another repo, the code tries to resolve all the
> callout links in the context of that repo. It looks for a committer
> and committer-date match for each callout and, in effect, replaces
> it with the corresponding parent mark.
> 
> The match attempt (and the graft) will fail if there is no matching 
> committer/committer-date pair in the repo, or more than one. In that
> case (which I expect to be rare and might not occur at all) the 
> offending callouts will need to be removed or hand-patched with ordinals,
> 
> What I will is locate each branch you want to preserve, write it
> out as a partial dump with callouts, and graft iyit onto the main
> repository in this way. 
> 
> -- 
> 		<a href="http://www.catb.org/~esr/">Eric S. Raymond</a>
> 
> The conclusion is thus inescapable that the history, concept, and 
> wording of the second amendment to the Constitution of the United 
> States, as well as its interpretation by every major commentator and 
> court in the first half-century after its ratification, indicates 
> that what is protected is an individual right of a private citizen 
> to own and carry firearms in a peaceful manner.
>         -- Report of the Subcommittee On The Constitution of the Committee On 
>            The Judiciary, United States Senate, 97th Congress, second session 
>            (February, 1982), SuDoc# Y4.J 89/2: Ar 5/5


[-- Attachment #2: smime.p7s --]
[-- Type: application/pkcs7-signature, Size: 4151 bytes --]

Re: Private branch grafts

[Eric S. Raymond]

David Reitter <david.reitter@gmail.com>:
> Thanks for your work.  Is there a version of the converted
> repository that we can inspect?

Not yet. but I'm going to do a conversion run tonight that will produce one 
I can push to gitorious for public inspection tomorrow or Saturday.
-- 
		<a href="http://www.catb.org/~esr/">Eric S. Raymond</a>