Will re-basing support be added into Bazaar core ?

[Ben Finney]

"Stephen J. Turnbull" <stephen at xemacs.org> writes:

> What Linus pointed out is that if you have a divergent pair of
> branches where C (the common ancestor) is tested and the branches A
> and B have both been tested, it's easy to forget that when you rebase
> B on top of A, all of your tests for A, B, and C have been invalidated
> and you have to run the whole suite again.[1]
> 
> This is just as true of looms but worse, because of the combinatorial
> explosion of "realizations" of the loom. (By "realization" I mean a
> workspace containing a specific list of threads in their various
> states.)

I don't see how that's true. The loom has the crucial difference that
none of the existing revisions are altered, nor are their relationships
in the history altered.

I am still committing every change to the working tree, even those that
are generated by updating a thread; so I am still able to test every
working-tree state that I commit, and once committed that fact (that the
committed working-tree state has a particular test result) won't change
in future.

This seems to be to be the crucial difference between a loom update and
a rebase, so I'm still not seeing how you conflate the two (“just as
true of looms but worse”) in this regard.

-- 
 \      “For my birthday I got a humidifier and a de-humidifier. I put |
  `\  them in the same room and let them fight it out.” —Steven Wright |
_o__)                                                                  |
Ben Finney