[Svnmerge] [PATCH] Bidirectional merging patch for svnmerge.py

Giovanni Bajo rasky at develer.com
Tue Feb 21 16:39:36 PST 2006


Blair Zajac <blair at orcaware.com> wrote:

>> This patch differs from my bidi patch to the original svnmerge by not
>> using dot-files to store the merge revisions. It instead uses
>> Archie's suggestion of checking the svnmerge integrated property.
>> The upside to
>> this is that no dot-files are required, the downside is that it is
>> slow because it requires a remote svn diff call for each possible
>> revision to
>> be merged.
>>
>> Because it can be slow, and because it is only required in special
>> situations (bidi merging), by default the reflected revision check is
>> not done. I have added an option "-b" or "--bidi" to the avail and
>> merge commands that must be specified if the user wishes to check for
>> reflected revisions.
>
> The current behavior could be thought of as sub-optimal from a merging
> standpoint (not from a performance point of view), so maybe we should
> enable it by default and add an option to disable the bi-directional
> support could be made as an optimization, for those people that know
> they don't need it.

I'm a little behind with patch and mail reading (I'll try to get in sync
tomorrow), but let me state I'm -1 on turning this on by default if we don't
find a way to implement it faster (there ought to be one!). I don't want to get
to the point where people read the svnmerge property and run a manual "svn
merge" because svnmerge is too slow and does too many things they don't
understand/know/want.

Giovanni Bajo




More information about the Svnmerge mailing list