[Svnmerge] [PATCH] Performance patch for merging

[Giovanni Bajo]

Raman Gupta <rocketraman at fastmail.fm> wrote:

> No, I don't believe it is wrong -- svnmerge merge is not called any
> more times than it was before because the minimal_merge_intervals
> adds the phantom revisions back in.

Hmmm it looks like you're right. This also means that the code in
minimal_merge_intervals was never really excercised before, as the revisionset
it was taking was already containing the phantom revisions. If I can find a
little time I shall add that testcase, but I'm very occupied in this period and
I can barely find time to review your patches (many thanks again, btw!).

>> As for 2), I know that log construction is very slow, but the
>> problem should be fixed otherwise. You should use
>> minimal_merge_intervals() in construct_merged_log_message(), so to
>> call "svn log" for ranges of revisions (instead of one revision at a
>> time). Even here, the idea is to call "svn log" less times as
>> possible, so it doesn't really matter if we call it once for a solid
>> range like 100-200 including phantom revisions: those phantom
>> revisions will not appear in the "svn log" output and thus will not
>> be part of the commit message.
>
> Yes, this is another optimization we can make. Once this optimization
> is made, my simpler patch can be reverted as I think it will be
> unnecessary.


Agreed. Let's keep your patch in as-is. Thanks!

Giovanni Bajo