A comment on an earlier post has prompted me to clarify the way that Subversion handles certain tasks. The commenter is using SQLDeveloper, that uses JDeveloper's SVN implementation, and wants-
1. Subversion navigator to indicate if others are working on the same code
Subversion uses the copy-modify-merge paradigm. This means that any user with the correct authorization can check out a copy of the code from the repository to a local file system. This local copy can be manipulated (using JDEV, Tortoise, command line etc) so that updates and commits can be carried out from it. However, the Subversion repository does not have any understanding of how many users are working on or have checked out the same code. Updates and Commits are instigated from the local copy. This also means that any local copy checked out from Subversion may never be checked back in.
2.When double clicking on a package in the database or in the versioning tree and have the option to load the local copy (linked to subversion).
With respect to the request to double click on the versioning tree this comes back to Subversion having no knowledge of the local copies. I think that he is asking that the local copy be updated through using the Subversion navigator - but updates are driven from the local copy, not the repository.
As for the database package, that would be another level of complexity. Presumably the single point of truth is the package definition held in the SVN repository. To ensure that the database holds the latest version the user would have to checkout a the latest version as a local copy from the repository and update the DB. I'm not sure that somehow automating this process would be desirable - it would need links from the DB to the tool to the correct local copy location and through this to the SVN repository - sounds error prone to me.
3. Be able to do a compile and see the log window
This is a SQL Developer question, rather than SVN oriented, so I'll leave that to my colleagues with SQLDeveloper to answer.