On Fri, 14 Sep 2007 10:10:05 -0700
"Hari Kodungallur" <hkodungallur@gmail.com> wrote:
> On 9/14/07, John Gateley <gateley@jriver.com> wrote:
> > Today I got a conflict in a file that I had not changed. To be
> > specific:
> >
> > Wednesday: merged the main branch (r10:20) into the experimental
> > branch working copy, committed the experimental branch (r21).
> > This included a change to Foo.cpp.
> >
> > Wednesday-Friday: many changes to main branch including some
> > to Foo.cpp. Many changes to experimental branch, but NONE to
> > Foo.cpp. Commited the main branch (r30) and the experimental
> > branch (r31).
> >
> > Friday: merge the main branch (r22:30) into the experimental branch
> > working copy. I get a conflict on Foo.cpp.
>
>
> Off the top of my head, I can not think any reason this should happen.
> Will it be possible for you to show the conflict in the email? It might be
> helpful to see if it is becasue of any special characters or somethign like
> that. Any chance the main branch and the experimental branch are checked in
> from different systems with different EOL or EOF schemes (Windows and Linux,
> typically)?
Thanks for the response. It happened again today. Both branches are
updated only from Windows. The conflict is not because of special characters.
I think it is because a previous merge did not work correctly.
Here's a small snippet of the file showing the only conflict:
class CTSFileReader : public CFileReader, public ITSReaderCallback
{
public:
CTSFileReader();
virtual ~CTSFileReader();
<<<<<<< .working
=======
virtual void OnFilesStoppedGrowing();
virtual void SetCancelForwardReadLoop(bool bCancel);
virtual void SetCancelReadAhead(bool bCancel);
>>>>>>> .merge-right.r410
virtual HRESULT SetFastPlay(long lFastPlayMode);
The problem is that the change from the main branch was only
to add a single line (OnFilesStoppedGrowing).
The other two lines had been added (in the main branch) a long
time ago, and somehow my merging missed them. They were added in
revision 331, and I merged revision 272 to 332. (ignore the revision
numbers in my example above, they are made up).
So, the question is: did I make a mistake entering the revision
numbers or is there a bug.
How would I find other places this has happened?
Thanks,
j
--
John Gateley <gateley@jriver.com>
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Received on Tue Sep 18 17:24:08 2007