Introduction
IndexSearch Setting is primarily used in Drillthrough and Spreadsheet datablocks.
Using IndexSearch Setting
Index search in Drillthrough allows users to make complex queries to a Fact Query via the use of the Elastic Search Query (DSL Query String) feature.
Click here for details about the Elasticsearch query syntax. You can also see the example query term when clicking on the Help icon.
1. Click on the IndexSearch Setting icon from the Drillthrough header.
2. Under Query tab, select an Index Group and any corresponding Factquery (or Factqueries) name to view data from the Factquery.
Multiple Factqueries can be selected in order to query from them. The result of selecting multiple Factqueries will be a union of the data in both the Factqueries in the Drillthrough.
However, if both the Factqueries have different schema then the data of the other Factquery will show up blank.
3. Filter by a selected date with various date options.
The Default date option corresponds to the Date specified in Global settings. It may not work if your fact query does not have the same Date as the Default Date in Global settings.
4. Click on the Grid icon on the right side of the IndexSearch Configuration window to build complex queries. This allows you to generate a query sent to Elasticsearch to further drill down their searches.
The corresponding query for the above selected options will look like the one below:
You can also modify queries to get the appropriate result. For example, use an OR clause instead of an AND clause.
A wrong query may result in an incorrect or no results in the Drillthrough.
Enter dates following the standard format of yyyy-MM-ddThh:mm:ss for the Query String search to work properly, otherwise the Drillthrough will return an incorrect result. This is the same format with dates stored in Elasticsearch.
Below is an example of a date search.
If eMite is set in a timezone other than UTC, the Date search has a higher chance of failing.
IndexSearch Setting for Spreadsheet is exactly the same with Drilthrough.