>>
>> if (notify_b.same_urls)
>> {
>> + /* ### TODO: Should we use from_same_repos? I don't like its
>> behaviour
>> + ### for dry-run case. */
>> + const char *src_repos_root;
>> + SVN_ERR(svn_ra_get_repos_root(ra_session, &src_repos_root,
>> pool));
>> + if (svn_path_is_ancestor(src_repos_root, entry->url))
>> + is_same_repos = TRUE;
>> + }
>>
>> is_rollback = (merge_type == merge_type_rollback);
>> SVN_ERR(svn_client__path_relative_to_root(&rel_path,
>> initial_URL1, NULL,
> ...
>
> Kamesh, can you say more about what's objectionable about
> from_same_repos? Could is_same_repos and from_same_repos be merged
> into a single variable? If not, it would be good to rename
> is_same_repos to something which better differentiates itself from
> from_same_repos.
>
Dan,
'from_same_repos' was giving 'FALSE' for '--dry-run', which I was not
sure why, So put a 'TODO' marker to take care of it.
I started using 'merge_cmd_baton->same_repos' instead of custom code, I
got 7 tests failing, complaining 'merge --dry-run' and 'merge' outputs
are not matching.
The cause of those failures was the optimization introduced at 'r25424'
based on your review comments at
'http://subversion.tigris.org/servlets/ReadMsg?list=dev&msgNo=127292'.
With the current state of things(fix to issue 2889), we need to know the
whether target and source are in the same repos in all cases including
--dry-run and --record-only.
Committed the same at r26362.
Thanks for the review.
With regards
Kamesh Jayachandran
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Aug 28 15:14:41 2007