Hi David,
If you look at the default configuration of the tck-webapp, which
basically targets test runs on jackrabbit, you can see that level 1
tests are all performed on the /testdata tree. Of course you can change
this configuration to something that better suits your repository or
make sure that /testdata contains enough nodes in order to run the test
cases.
What data you have to place under that node is actually quite simple.
Whatever your repository supports should be made visible. E.g. if your
implementation supports same name siblings, create a node with same name
siblings somewhere under /testdata.
You may also look at the init test cases in jackrabbit as a starting
point. That's in: core/src/test/org/apache/jackrabbit/init
Jackrabbit executes those test cases first to prefill the repository
before any other test cases run.
regards
marcel
David Caruana wrote:
I would like to bootstrap my implementation of JCR with the test data
required for the TCK. However, it's taking more time than necessary to
determine what the test data should be. Is there any documentation that
tells me the nodes, properties, values etc I should create for the TCK
tests. For now, I only have level 1 api support, but can easily create
the data through my own means, if I knew what it should it be.
