[PATCH] win32text: be more careful about rejecting violating changesets

[Dirkjan Ochtman]

On Thu, Apr 23, 2009 at 21:19, Bryan O'Sullivan <bos at serpentine.com> wrote:
> This should prevent us from rejecting a changegroup that contains an
> unacceptable commit followed later by a commit that fixes the problem.

This sounds useful, but I'm not sure that no one would ever want the
other/old behavior (e.g. because all csets should be testable, or
something like that).

Does that sound silly (it sure might, I've never used win32text)? Or
should this maybe be optional in some way?

Cheers,

Dirkjan