Bazaar still below the radar when evaluating VCS tools

[Matthieu Moy]

Ben Finney <ben+bazaar at benfinney.id.au> writes:

> "Stephen J. Turnbull" <stephen at xemacs.org> writes:
>
>> Andrew Bennetts writes:
>>
>>  > I'm going to guess though: I think Stephen might be saying that none of
>>  > the systems automatically capture a 'mv foo bar' done in their working
>>  > trees, when ideally they would.
>>
>> Precisely.
>
> In that case, I disagree :-) It's not necessarily the case that the VCS
> should automatically pick up on that, and I prefer that it not
> guess.

But almost all VCS do "guess" changes within files. The output of "bzr
diff" is just a guess, it does not necessarily match your intention
when you did the changes. As a proof that it's a guess, bzr did change
its diff algorithm in the past to use patience diff, which should
match the programmer's intention in more cases.

Version-control systems recording in-file changes as they happen
exist, an example is MS-Word.

The philosophy behind Git's management of renames is just to use a
heuristic not only for in-file changes, but also for changes accross
files.

No solution is perfect, but I never had to complain about Git's
management of renames. And by experience, most complain about Git's
rename management come frome people who never actually tried it.

-- 
Matthieu Moy
http://www-verimag.imag.fr/~moy/