The major difference between the standard query request and the Structured Query request is that the structured query consists of multiple search input texts. Each search text is treated as a separate query with some of its own settings, including which fields to search within. This allows a structured query to search differently in different fields, taking advantage of the structure.
Note that although ATG Search performs multiple queries of structured content, only one result is returned to the user. Because each query can be on different text fields, there is no way to intersect the results at the statement level; instead, the structured query intersects results at the index item (document) level, aggregating statements from the same index item. Therefore, although each individual query retrieves statements, the final result list is focused on the retrieved index items.
The response has the same structure as the standard Query request (described in the Standard Query chapter), but each result contains all matching statement texts from each retrieved index item, and the relevancy is a weighted average of the individual relevancies.
Aggregation is controlled by a special operator on each individual query (see the Query Aggregation Operator section). Individual queries must be sorted by the precedence of these operators, as shown below:
Required queries – final aggregate results (by index item) must have matched these queries.
Optional queries – final aggregate results may have matched these queries, and if so, their relevancies are affected.
Negative queries – final aggregated results may NOT have matched these queries. These queries effectively subtract results from the required matches.
Requests with statements in a different order will not return expected results.
The Structured Query request consists of one or more individual queries which are executed separately. Each query has a search text, plus the following settings that control its execution. There are also global settings for all queries.
