Just curious, why do you want to do Subversion commands in Ant?
Sometimes it's necessary. For example, if you use CruiseControl, you need to do the checkout for CruiseControl in your build.xml because CruiseControl doesn't do it.
However, most of the time, it's not a good idea. For example, you might use your build.xml to checkin the resulting built artifacts. This is not a good idea. You should be storing built artifacts in a release repositor where they can be available for other projects. Checking in large binary files into your Subversion repository will make your Subversion repository grow to tremendous size with little benefit. These built binaries take a lot of room and become obsolete in a matter of months.
We use Ant with Ivy. Our builds are done with Jenkins, and we use the mvm deploy:deploy-flle command in Jenkins to deploy our files into our Maven repository where Ant w/ Ivy can check them out for other projects. The stuff we deploy to other machines are kept in Jenkins.
Martin Clayton has answered you where you can find the files you need to get the Ant Subversion tasks working. However, my experience has been that it's simply easier just to use <exec> and execute Subversion commands directly. This way, you don't have to make sure that the Subversion API is installed and that it matches the Subversion client and server.