some notes about svn_client__get_copy_source

I was a little confused by the change in r25180. dlr cleared up my
confusion on IRC; svn_client__get_copy_source is a work in progress,
mostly intended to support svn_client__suggest_merge_sources for now.
I had a few suggestions for future improvements to
svn_client__get_copy_source; he asked me to mail the list so they're
not forgotten once it's time to work on svn_client__get_copy_source
itself again.

First, we're currently exposing svn_client__get_copy_source publicly
in JavaHL (as SVNClient.getCopySource); we probably want to make it
public in C as well, or private in JavaHL, instead of being
inconsistent.

Also, The basic purpose of svn_client__get_copy_source is very similar
to svn_fs_closest_copy. There are several edge cases that
svn_fs_closest_copy handles that r25180 doesn't, such as nested
copies:

    $ svn cp branches/foo branches/baz
    $ svn cp trunk/bar branches/baz/bar

and adds within copies:

    $ svn cp branches/foo branches/baz
    $ svn mkdir branches/baz/bar

(In neither of these cases should branches/foo be the copy source of
branches/baz/bar; in the latter case, it has no copy source.)

Perhaps we should just expose svn_fs_closest_copy over RA instead of
reimplementing its logic in libsvn_client?

(But svn_client__get_copy_source does work in the most basic cases
right now and it seems reasonable to wait to clean this up until after
its consumers are ready.)

--dave

-- 
David Glasser | glasser_at_mit.edu | http://www.davidglasser.net/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org