Re: [Gnu-arch-users] Re: [BUG] FEATURE PLANS: "perfect" summary deltas

[Tom Lord]

    > From: Aaron Bentley <address@hidden>

    >> A purely "commit --continuation" branch fits the bill
    >> precisely --- all that's needed (at core) is to remove the restriction
    >> that currently prohibits "commit --continuation" revisions.

    > Maybe that's the only core issue, but there are a number of others:
    > - builder changes
    > - branch-naming changes
    > - name-mangling


The builder changes should be utterly trivial.  They amount to this:

    new_build (rev)
    {
       if (success == old_build (rev:summary))
         {
           remove patch log for summary branch
           return success
         }
       else
         {
           return old_build (rev);
         }
    }

plus modifying old_build to call new_build whenever it 
crosses a version boundary.

That's one reason why the "mathematical neatness" of summary deltas is
cool: because it interacts with the existing build algorithm in a
useful way.  (Another reason is that it's deterministic and
predictable, and hence a reliable tool.)

The branch-naming changes and name mangling have been a long time
coming.

In the future, we should wind up with high-level meta-commands for
managing archives and families of related archives.  These will take
care of mirror creation and maintenance.

The summary-delta plan as presented will allow such higher-level tools
to easily give users very clear choices (and good defaults) about
trading space for time.


-t