Search Implementation Steps
Creating a Search Engine Interface
The Search Engine is used to connect Fabric to the Elasticsearch engine when running Search commands. During a search, the Search Engine's interface is populated with the Elasticsearch engine's connection details.
To create a new Search Engine interface:
- Go to Project Tree > Shared Objects, right click Interfaces and select New Interface.
- Select Search Engine from the dropdown menu.
- Populate the connection settings.
Search Engine Interface Settings
Fabric Studio - Defining Search Fields
Search fields must be defined on the selected LU table's columns to enable a cross-instance search based on these columns. For example, to enable a search of all customers named "John Doe", define the FIRST_NAME and LAST_NAME columns of the customer schema as search indexes.
To create a search index on a column, do either:
- Right click the selected column > Create Search Index From Selected Columns > select the index type.
- Open the Search tab and populate the Column Name and Type.
Note that Fabric Studio does not enable defining more than 63 columns in the same LU table as CDC fields, assuming that all columns are positioned according to 1 to 63 in the LU table.
Search Field Types
The default built-in types of Search fields are:
- keyword, enables a search by this column. In the background Fabric creates two search indexes for the keyword field, a keyword index and an additional index based on the type of field. The keyword index enables searching for an exact match (case sensitive) of the searched value.
- date, enables a search on a date column. Date fields must be populated by a date format and identified by Elasticsearch. Click to see the list of the date formats supported by Elasticsearch:
- data, can be returned by the search, a search cannot be initiated by this column.
- search templates, Fabric supports adding templates for Search fields to support specific cases.
Example:
- Set the search indexes of the Customer LU as follows:
- First Name and Last Name are keyword fields.
- Customer ID is a data field.
- Run a search to get the list of all customer IDs called “John Doe”.
Search Implementation Steps
Creating a Search Engine Interface
The Search Engine is used to connect Fabric to the Elasticsearch engine when running Search commands. During a search, the Search Engine's interface is populated with the Elasticsearch engine's connection details.
To create a new Search Engine interface:
- Go to Project Tree > Shared Objects, right click Interfaces and select New Interface.
- Select Search Engine from the dropdown menu.
- Populate the connection settings.
Search Engine Interface Settings
Fabric Studio - Defining Search Fields
Search fields must be defined on the selected LU table's columns to enable a cross-instance search based on these columns. For example, to enable a search of all customers named "John Doe", define the FIRST_NAME and LAST_NAME columns of the customer schema as search indexes.
To create a search index on a column, do either:
- Right click the selected column > Create Search Index From Selected Columns > select the index type.
- Open the Search tab and populate the Column Name and Type.
Note that Fabric Studio does not enable defining more than 63 columns in the same LU table as CDC fields, assuming that all columns are positioned according to 1 to 63 in the LU table.
Search Field Types
The default built-in types of Search fields are:
- keyword, enables a search by this column. In the background Fabric creates two search indexes for the keyword field, a keyword index and an additional index based on the type of field. The keyword index enables searching for an exact match (case sensitive) of the searched value.
- date, enables a search on a date column. Date fields must be populated by a date format and identified by Elasticsearch. Click to see the list of the date formats supported by Elasticsearch:
- data, can be returned by the search, a search cannot be initiated by this column.
- search templates, Fabric supports adding templates for Search fields to support specific cases.
Example:
- Set the search indexes of the Customer LU as follows:
- First Name and Last Name are keyword fields.
- Customer ID is a data field.
- Run a search to get the list of all customer IDs called “John Doe”.
