Helpful dummy objects
Example: You want to update many objects with the same attributes, but some of your result sets should not be updated. Or you must update a selection of objects with different attributes.
One of the possible solutions is to create many unique update statements or WHERE conditions. Or you create a big IN or NOT IN condition with many object_ids.
The best practice is to create a dummy dm_document object in your repository (i.e. with the DQL Tester). The included or excluded objects_ids are stored in the keywords of the dummy dm_document object. The dm_document object can store a huge number of keywords (> 10 000). In practice I have saved more than 30k object_ids in one document. The SELECT or UPDATE statement is very easy and is issued like this:
SELETCT r_object_id, r_lock_owner, object_name, r_lock_date FROM r_and_d_document (ALL) WHERE r_object_id NOT IN (select keywords from dm_document WHERE object_name=<dummy name>) ORDER BY 4 DESC
This statement selects all checked out documents except a number of predefined documents. You can execute many select statements to select the excluded documents. This has the advantage that all of your results are documented.