In a Data Module a developer creates a joined view from two tables (a and b) and then joins this to another table (c). What would be the execution logic in the database?

Question

Options

• AAll queries are collapsed and a standard join across all three tables (a, b, c) is performed.
• BThe columns used in table (c) are are joined on the key to table (b) first and then the results are
• CIt makes no difference to the execution as a joined view is only used to group data items in a
• DThe joined view (a and b) is evaluated and all results are then post-joined to the other table (c).

• A
  8% (4)
• B
  31% (16)
• C
  16% (8)
• D
  45% (23)

Explanation

Option D is correct because in a Data Module, a joined view is treated as a derived table (subquery) - the database first fully evaluates the join between tables a and b, materializes those results, and only then joins that intermediate result set to table c. This two-phase execution is the defining behavior of joined views in Data Modules.

Option A is wrong because the database does not flatten or collapse the joined view into a single three-table join; the view boundary is respected, preventing the query optimizer from reordering all three joins freely. Option B is incomplete and describes an incorrect execution order that does not reflect how the view boundary is honored. Option C is wrong because joined views are not merely cosmetic groupings - they have real, tangible impact on how the database executes the query, as D demonstrates.

Memory tip: Think of a joined view as a "sealed box" - the database must open and process everything inside the box (a + b) before it can use the box's output to connect to anything outside it (c). Sealed box first, then connect.

No community discussion yet for this question.