Ok, I'll clarify what happened.
Repository is trunk at rev 205. I rename a directory. That's rev 206. I
move a few files, copy a file, and modify the copy. That's rev 207. I
make another miscellaneous change. That's rev 208. So now most of my
repo is 205 (which is still up to date as that's at least the last
committed rev), and part of it is between 206 and 208. I modify 2
files, then realize it should be in a new branch instead. I do `svn
copy . <url>/branches/resizable`. I then do `svn switch
<url>/branches/resizable` and I see a bunch of adds/deletes/mods happen
- i.e. it was essentially reverting my changes back to rev 205. The
only difference is the local changes I had in my WC are present in the
branch. This makes me think that subversion created a transaction on
the server which contained all of revision 205, then applied my local
uncommitted changes to this, then saved that txn into the branch.
I will attempt to reproduce this.
On Aug 11, 2004, at 2:03 PM, Ben Collins-Sussman wrote:
>> My WC was mixed-revision but every revision was at least at the
>> last-committed revision (plus the couple of modifications in files,
>> which is why I was doing a copy of my WC instead of just the trunk on
>> the server),
> There's no way to speculate about what was going on here; I've been
> assuming that your working copy was all at r205, with exception of a
> files at r208, plus a few local mods. But do we know that's true?
>> Instead, what showed up in the branch was entirely revision 205.
> Are you saying that your local mods weren't uploaded to the repository?
> If you think there's a bug here, try doing a test and show us a
Received on Wed Aug 11 20:10:59 2004
- application/pkcs7-signature attachment: smime.p7s