Vladimir Prus <ghost@cs.msu.su> writes:
> I do:
>
> svn add foo
>
> and send "foo" to somebody else, who does:
>
> svn add foo
> <some minor edits of foo>
> svn ci
>
> Then I do:
>
> svn up
>
> and get the "obstructed update" error message.
Ah, I understand, thanks.
In which case, I disagree with your plan of action. In this
situation, I do not think the svn client should attempt to merge
changes into your copy of foo, because there's no reason to believe
(and no way to know) that the two files are related. It would be very
bad to merge a patch from an unrelated file... especially when your
copy of foo isn't really under version control yet. Your copy of foo
isn't in the repository yet, and there's no text-base of it yet, so
the contextual merge has the potential to destroy your data: there's
no way to simply revert or undo the conflict markers.
You claim that this is a common situation. Is that true? In a year
and a half of using Subversion every day, I've never seen this
scenario happen before. Why does it happen to you so often?
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Mon Nov 25 15:39:28 2002