We are probably using Subversion and SQL Developer differently than you envision.
We work in a data warehouse that contains over 6 terabytes of data. There are extensive foreign-key relationships that we enforce via Constraints. Accordingly, dropping and recreating tables is not a desirable approach.
We work on Oracle objects in our Development environment. When they are ready for Integration Testing, we export them to a Working Copy (on the developer's PC), which we then Commit to the Subversion Trunk. We then Tag that object as a member of a "release", which is a folder containing two layers of sub-folders: Schemas, and under each Schema, Object Types. In addition to the standard Object types generated by a SQL Developer Tools->Database Export (Tables, Views, Indices, Packages, Package_Bodies, et cetera), we have two additional types: DDL_Changes and Adhocs (which contains DML for changes to our Dimension tables). Normally we have an iteration (release) every 3 weeks. This approach has worked well for us, and has been supported by SQL Developer's Subversion interface up until this (Early Adopter) release. It was easy to Branch/Tag within the Repository by using the Copy URL and Branch/Tag options on the right-click context menus.
We can certainly consider modifying our approach to using Subversion, but this is a change from the current options available in SQL Developer's Subversion interface.
It has belatedly occurred to me why we use this particular process. In both SQL Developer versions 3.2 and 4, the Branch/Tag window for a file in my working copy pops up with a default "FROM" path that points to the parent directory. This is different from that which pops up when I open the Branch/Tag window for the same object from the trunk, where it points to the individual file.
For example, when I open the working-copy SQL file for a package body for a module named PKG_LOAD_DIM, the Branch/Tag window default for the FROM path is "svn://clcyessvnp01/datawarehouse/trunk/Oracle/stnd_rpt/PACKAGE_BODIES". When I open the Branch/Tag window for that same object from the trunk (via the "Team -> Versions" window in SQL Developer version 4, and via the "View-> Team -> Versioning Navigator" window in SQL Developer version 3.2), the Branch/Tag default for the FROM path is "svn://clcyessvnp01/datawarehouse/trunk/Oracle/stnd_rpt/PACKAGE_BODIES/PKG_LOAD_DIM.sql".
If that were to be changed (so that the "working-copy Branch/Tag FROM default" pointed to the individual file rather than the whole directory) it would make more sense to Branch/Tag directly to the release directory, and then merge the release into the Trunk, which is, if I understand correctly, the more normal approach to using Subversion.
We realize that our current approach is unusual, and would not work well for a large team. It works because we have a very small team.
We plan to have the branch/tag menu available on the file in the version navigator (like 3.2).
We plan to have the default From path to be the file (not the parent folder) from the file navigator.
The paths should be the working copies/repository source location. That is where they were checked out from.
From the windows tortoise svn doc:
"The default destination URL for the new branch will be the source URL on which your working copy is based.
You will need to edit that URL to the new path for your branch/tag."