This is probably a rookie question, but nobody here seems to be able to help me out. I’ve got some tables in an Oracle database and we know that there is one application that uses BO reports that come from these tables. We are considering some changes to this data and need to know if it impacts other BO reports.
I have attempted to write a report from a universe called BO Metadata. Looks like something that was created by BO or some BO user out there. But I do not see where Universe and Document ids intersect in this universe. Furthermore I have run some queries against the repository tables, but they are so freakin’ ugly its hard to discern anything from them unless you have a PHD in Repository.
The question I want to answer is which ( if any) BO reports use a particular universe, and it would also be nice to know which objects within that universe.
Hi, Shaun… you might want to take a look at a utility I posted a little while ago. It allows you to scan all documents in a particular directory and get a list of which objects are used. With a little work, you can also get a list of which universes are used.
If you just want to know which universes are used in a document, you can get that from the repository looking at the data provider table. I’ll have to review it and get back with you as to which table that is (or someone else can post it). But if you need the information down to the object level, you need to do that outside of the repository.
These two solutions don’t help me much. I have 831 reports in my repository. I don’t have time to import each one and check out each of the data providers.
Dave’s suggestion isn’t going to work for me either, because I do not have a general directory that holds all the reports that I need to look at.
It would be real helpful if somebody could post some SQL to pull the info I need straight out of the Repository.
Well, you can’t get object level information from the repository. And if you have the same universe in multiple domains, then you can’t tell which universe is used (which version) either. But if you want to make assumptions…
select b.m_doc_c_name doc_name,
a.m_docatv_c_dpname dp_name,
a.m_src_c_name source_name
from obj_m_docatvar a,
obj_m_documents b
where a.m_doc_n_id = b.m_doc_n_id
That will pull the document name, the data provider name (if that is useful to you), and the source name out of the repository. The source name will be the short name of the universe file. You could join that to universe table to get the long universe name if that would be useful.
I have an enhancement planned for my utility that would ask you for a domain, then retrieve every document that you have access to in that domain do a specific folder. That would eliminate the need to import 831 documents on your own. But that’s just one of many things on my “to do” list.
(BOB member since 2002-06-06)
