Configuring the attributes in a data source

On the Configure Data Source page of the data source configuration dialog, you configure the attributes that are included in the data source.

The configuration for data source attributes includes most of the options available when users create new data sets. This allows you to provide default values that should in most cases be correct, so that users do not have to change them.

When the page is first displayed, it shows the Basic Options view.

To display additional configuration options for the attributes, click Advanced Options.

To configure the data source attributes:

1. From the Default Search Language drop-down list, select the default language to use for keyword searches against the data.
2. Use the Include column to determine whether each attribute is available to users when they create a data set.

   If the checkbox is checked, then the column is included in the data source.

   By default, the checkbox is checked for all columns for which the Provisioning Service can retrieve data.

   If the data is not accessible, for example because the column is a constant or a reference to another column, then the checkbox is unchecked by default. If a checkbox is unchecked by default, it is recommended that you leave it unchecked.

3. To allow users to use the attribute to filter the data that is included in a data set, check the Available to Filter by checkbox.

   If you do not specifically enable any of the attributes for filtering, then users cannot filter the data at all. If the source data contains a very large number of records, then it is recommended that you enable filtering for some of the attributes.

   Note that only numeric, string, and date/time fields can be enabled for filtering. Users can never use time or duration fields for filtering.

   You also should be careful about allowing users to filter by database columns that use CHAR or NCHAR as the original column type. If the columns contain values that are padded with spaces, users may have difficulty getting values to match when they go to filter the data. For example, if the value is actually "abc" followed by two spaces, there won't be a match when users type "abc" without the spaces.

   When you check the checkbox, an edit icon is added to the column.

   
   
   

   To display a hint to users who are using the attribute to filter the data for a data set:

   1. Click the edit icon.
   2. On the filter hint dialog, in the Help text field, type the hint.
   3. Click Save.
4. The Attribute Name column contains the default display name for the attribute. You can use the field to change the default name.
5. The Data Type column indicates the data type for the attribute as specified in the source data. To use a different data type in the data source, from the drop-down list, select the data type.

   Note that if an attribute is not initially identified as a date, you cannot change the data type to make it a date.

6. In the Advanced Options view, you can use the Adjustment Rules column to provide rules for setting the attribute value if the original value is either invalid (does not match the data type) or empty.
   1. Click the Edit Rules button.
   2. Under Fix non-matching data type values, click a radio button to indicate how to handle values that cannot be converted to the selected data type. You can choose to either:
      • Replace the invalid values with a blank value
      • Replace the invalid values with a custom value
      • Remove records that have invalid values

      
      
      

      Note that because string values never have non-matching values, you cannot configure rules for adjusting invalid values.

   3. Under Fix missing values, click a radio button to indicate how to handle empty values.
      You can choose to either:

      • Leave the blank values as is
      • Provide a custom value to use wherever a value is missing
      • Remove records that have empty values
   4. To display a sample list of values for the attribute, expand the Sample Values section at the bottom of the dialog.
7. In the Advanced Options view, the Split and Merge columns allow you to create multi-value attributes.

   The Split option is used when the attribute value is actually a list of values. For example, for a Color attribute, you can indicate that "blue;red;white" are actually three separate values delimited by a semicolon.

   To split the value for an attribute:

   1. Check the Split checkbox for the attribute.
   2. On the split value dialog, under Values are delimited by, specify the delimiter used to separate the values.
      
      
      

      Select the delimiter from the drop-down list. If the delimiter used is not in the drop-down list, click the other radio button, then type the delimiter in the field.

   3. To save the split configuration, click Save.
   4. If you need to change the split configuration, click the edit icon in the Split column.
   5. To not split the attribute value, uncheck the Split checkbox.

   The Merge option is used to combine values from other attributes into the current attributes. For example, you could add the values from Color2 and Color3 to the Color1 attribute in order to generate the full list of colors.

   To merge values from multiple attributes:

   1. Check the Merge checkbox for the attribute.
   2. On the merge dialog, from the Attribute 2 drop-down list, select the first attribute to merge with the original attribute.
      

      The merged attributes must be of the same type as the original attribute.

   3. To add the attribute, click the Add Attribute button.
   4. To remove an attribute, click its delete icon. You cannot remove the original attribute.
   5. To save the merge configuration, click Save.

      On the Configure Data Source page, you can no longer edit the merged attributes. They automatically inherit all of the configuration options you specify for the original attribute.

      The Merge checkbox is also checked for those attributes.

   6. If you need to change the merge configuration, click the edit icon in the Merge column.
   7. To remove the merge, uncheck the Merge checkbox.
8. In the Advanced Options view, from the Refinement Behavior drop-down list, select how refinement works for the attribute.

   The refinement behavior indicates whether users can refine by multiple values, and whether multiple values use AND or OR.

   The available options are:

   Option	Description
   Multi-Or	Indicates that end users can refine by more than one value at a time. For multi-or, a record matches if it has at least one of the selected values. So if an end user selects the values Red, Green, and Blue, then matching records only need to have one of those values (Red or Green or Blue). For most attributes, this is the default and the recommended value. Date attributes are always multi-or.
   Multi-And	Indicates that end users can refine by more than one value at a time. For multi-and, a record matches only if it has all of the selected attribute values. Multi-and should only be used with multi-value attributes. So if an end user selects the values Red, Green, and Blue, then matching records must have all of those values (Red and Green and Blue).
   Single	Indicates that end users can only refine by one value at a time. The Single refinement behavior is recommended for string attributes with long values, and numeric and geocode attributes with a large number of unique values. It is not recommended for multi-value attributes, including attributes that you have split or merged.

9. In the Advanced Options view, the Searchable column indicates whether the attribute can be used for text searches.

   Text searches are when end users use the Search Box component to search for a specific search term.

   The Searchable checkbox is only displayed and checked by default for all string attributes.

   Other types of attributes do not support text search.

   To exclude the attribute from text searches, uncheck the Searchable checkbox.

   By default, the text search for the attribute uses the default search language. If the attribute values are in a different language than the default, then use the drop-down list to select the search language.

10. To save the new or updated data source, click Save.