[PATCH][merge-tracking]Fix for merge-tests 24 failure

[Kamesh Jayachandran <ka...@collab.net>]

Hi All,
Find the attached patch and log.

With regards
Kamesh Jayachandran

Re: [PATCH][merge-tracking]Fix for merge-tests 24 failure

[Daniel Rall <dl...@collab.net>]

On Fri, 22 Sep 2006, Kamesh Jayachandran wrote:

> Daniel Rall wrote:
> >On Thu, 21 Sep 2006, Kamesh Jayachandran wrote:
> >
> >>>Kamesh, thanks for the test fix -- I've committed to the branch as
> >>>r21577 with some tweaks.  While correct handling of merge info for the
> >>>3-way merge really doesn't work yet, this is an incremental step in
> >>>the right direction (documented in the TODO file).  It's great than
> >>>you were able to make forward progress and fix a test at the same
> >>>time!
> >>>      
> >>Should we change our 'svn:mergeinfo' format a bit take care of '3 way 
> >>merges'?
> >>Like 'pathname:revlist'|pathname1@rev1-pathname2@rev2'?
> >>    
> >
> >I don't understand this.
>
> Like every other merge just record the 'svn:mergeinfo' with the value of 
> the form 'path1@rev1-path2@rev2' for 3 way merges so that future such 
> merges might start from rev1+ and rev2+ for those corresponding paths. 
> This would be consistent with our way of maintaining the mergeinfo.(Of 
> course it needs a tweak to 'sqlite database schema'.)

DannyB nixed this as unnecessary.

Re: [PATCH][merge-tracking]Fix for merge-tests 24 failure

[Kamesh Jayachandran <ka...@collab.net>]

Daniel Rall wrote:
> On Thu, 21 Sep 2006, Kamesh Jayachandran wrote:
>
>   
>>> Kamesh, thanks for the test fix -- I've committed to the branch as
>>> r21577 with some tweaks.  While correct handling of merge info for the
>>> 3-way merge really doesn't work yet, this is an incremental step in
>>> the right direction (documented in the TODO file).  It's great than
>>> you were able to make forward progress and fix a test at the same
>>> time!
>>>       
>> Should we change our 'svn:mergeinfo' format a bit take care of '3 way 
>> merges'?
>> Like 'pathname:revlist'|pathname1@rev1-pathname2@rev2'?
>>     
>
> I don't understand this.
>
>   
Like every other merge just record the 'svn:mergeinfo' with the value of 
the form 'path1@rev1-path2@rev2' for 3 way merges so that future such 
merges might start from rev1+ and rev2+ for those corresponding paths. 
This would be consistent with our way of maintaining the mergeinfo.(Of 
course it needs a tweak to 'sqlite database schema'.)

With regards
Kamesh Jayachandran

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: [PATCH][merge-tracking]Fix for merge-tests 24 failure

[Daniel Rall <dl...@collab.net>]

On Thu, 21 Sep 2006, Kamesh Jayachandran wrote:

> >Kamesh, thanks for the test fix -- I've committed to the branch as
> >r21577 with some tweaks.  While correct handling of merge info for the
> >3-way merge really doesn't work yet, this is an incremental step in
> >the right direction (documented in the TODO file).  It's great than
> >you were able to make forward progress and fix a test at the same
> >time!
> 
> Should we change our 'svn:mergeinfo' format a bit take care of '3 way 
> merges'?
> Like 'pathname:revlist'|pathname1@rev1-pathname2@rev2'?

I don't understand this.

The current plan for handling 3-way merges is described in the TODO
file:

  * A better solution requires that either the client or server know
    the merge info for both the WC and repos.  This would be
    implemented by transfering the merge info for the versioned
    sub-tree(s) involved in the merge to or from the repository via a
    (new?) RA operation, allowing the calculation of which merges
    remain.  The plan is to do this processing on the server-side, an
    unfortunate difference from the more typical 'merge -rX:Y' style
    of operation (which will be kept client-side to offload more
    processing from the server).

Re: [PATCH][merge-tracking]Fix for merge-tests 24 failure

[Kamesh Jayachandran <ka...@collab.net>]

> Kamesh, thanks for the test fix -- I've committed to the branch as
> r21577 with some tweaks.  While correct handling of merge info for the
> 3-way merge really doesn't work yet, this is an incremental step in
> the right direction (documented in the TODO file).  It's great than
> you were able to make forward progress and fix a test at the same
> time!
>   

Should we change our 'svn:mergeinfo' format a bit take care of '3 way 
merges'?
Like 'pathname:revlist'|pathname1@rev1-pathname2@rev2'?

With regards
Kamesh Jayachandran

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: [PATCH][merge-tracking]Fix for merge-tests 24 failure

[Daniel Rall <dl...@collab.net>]

On Wed, 20 Sep 2006, Kamesh Jayachandran wrote:
...
> Comparing 'range->start' and 'range->end' on two different URLs is
> meaningless.
> 
> * subversion/libsvn_client/diff.c
>   (grok_range_info_from_opt_revisions):
>    Accepts URL1 and URL2 to make a meaningful comparison of 
>    range->start and range->end.
>   (do_merge): Calls new 'grok_range_info_from_opt_revisions'.
>   (do_single_file_merge): Calls new 'grok_range_info_from_opt_revisions'.

Kamesh, thanks for the test fix -- I've committed to the branch as
r21577 with some tweaks.  While correct handling of merge info for the
3-way merge really doesn't work yet, this is an incremental step in
the right direction (documented in the TODO file).  It's great than
you were able to make forward progress and fix a test at the same
time!