On Thu, Mar 10, 2016 at 1:10 AM, Johan Corveleyn <jcorvel_at_gmail.com> wrote:
> [ Please, if possible, use plain-text on this mailing list, and use
> bottom-posting (putting your reply at the bottom, or inline to the
> thing you're replying to). More below ... ]
>
> On Wed, Mar 9, 2016 at 3:08 PM, <webster.brent_at_rogers.com> wrote:
>> These files were never changed in the first place, that's the weird part.
>> We aren't using any local locks in the repo.
>
> Okay, so definitely a "non-content-changing" revert then. So we're
> talking about needless notifications by revert, possibly scaring
> users.
>
> I've quickly tried to reproduce it, based on Bert's explanation of
> read-onlyness changes. No need for locks or the svn:needs-lock
> property. Just make a file read-only, and run a recursive revert. The
> file is made writable again, and revert notifies:
>
> [[[
> C:\Temp\svntest>svn --version -q
> 1.9.3-SlikSvn
>
> C:\Temp\svntest>svnadmin create repos
>
> C:\Temp\svntest>svn co file:///c:/Temp/svntest/repos wc
> Checked out revision 0.
>
> C:\Temp\svntest>cd wc
>
> C:\Temp\svntest\wc>echo This is file 1 > file1.txt
>
> C:\Temp\svntest\wc>svn add *.txt
> A file1.txt
>
> C:\Temp\svntest\wc>svn ci -mm
> Adding file1.txt
> Transmitting file data ..done
> Committing transaction...
> Committed revision 1.
>
> C:\Temp\svntest\wc>attrib +R file1.txt ### (making read-only)
>
> C:\Temp\svntest\wc>svn st
>
> C:\Temp\svntest\wc>svn revert -R .
> Reverted 'file1.txt'
> ]]]
>
>
> Is this similar to what could have happened, Brent? I'm assuming it's
> not Windows-specific, because you reported the problem for a 1.8.14
> svn client on Centos6.5 Linux. Maybe other variants of changes in
> permissions, executability or file ownership also give the same
> behaviour.
>
> I agree this is an issue, but I'm not sure if there is only one issue
> or two :-).
>
> 1) Is it OK for revert to change file metadata? Is that intended
> behaviour? Maybe it's OK for revert to change file metadata if the
> file also has content-changes that need to be reverted, but not if the
> file only has metadata-changes?
>
> 2) If we revert metadata-only-changed files, should revert notify
> about this? If it notifies for a metadata-only change, maybe the
> notification should indicate this?
FYI:
Lost track of this for a while, but now I've filed an issue for this,
so we don't forget about it:
https://issues.apache.org/jira/browse/SVN-4637 (Revert affects
unchanged files with changed permissions)
I think it's wrong for revert to change the permissions of an
unchanged file. I think 'revert' should not care about such files,
since 'status' doesn't report them either.
--
Johan
Received on 2016-06-20 23:34:58 CEST