hg rollback, hg qimport -r tip different behaviour

[Martin Geisler]

"rupert.thurner" <rupert.thurner at gmail.com> writes:

> what is the reason why "hg qimport -r tip" does not work on a merge
> changeset, while "hg rollback" does?

The two commands solve very different things. It's a bit like when I
explained the difference between rollback and strip:

  http://markmail.org/message/w3nztbrncc6damwx

As I'm sure you know, the qimport command does a third thing: it turns a
changeset into a MQ patch.

It is actually a shame that you cannot qimport merge changesets. Such a
changeset could in principle be turned into a patch against the first
parent with a meta data line that says what the other parent is.

That meta data line would of course quickly become outdated if you
modify both parents further -- that suggests that the line could instead
refer to a patch name... So the real reason is probably that it's
difficult to come up a good and robust user interface for this.

-- 
Martin Geisler

VIFF (Virtual Ideal Functionality Framework) brings easy and efficient
SMPC (Secure Multiparty Computation) to Python. See: http://viff.dk/.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 196 bytes
Desc: not available
Url : http://selenic.com/pipermail/mercurial/attachments/20090813/1e26c142/attachment.pgp