Crawler field mappings
After you crawl a data source, data source fields are automatically mapped to dataset fields, if possible. This tab shows the results of this mapping. You can modify the mapping of individual data source fields, as well as adding or removing data source fields.
After you create a crawler and the crawler makes an initial run, the discovered native fields are listed in the Field mappings tab. The data source field to dataset field mappings are automatically created. You can change the dataset field name and type here, and also delete data sourced fields from the mapping. You can also add new mappings.
- Changing a dataset field name, type, or mapped dataset field
- To modify an existing mapping, click Edit... You can now modify the field name and type. To change the mapped dataset field, select a new dataset field from the drop-down list.
- Deleting a mapping
- To delete an existing mapping, click
and then click Delete data source field.
- Adding a mapping
- To add a new mapping, click
and then click Add data source field. Specify the crawl space, field name, and type, and then click Save. Select the mapped dataset field from the drop-down list.
After you have updated the field mappings, click Save and run the crawler to reflect the changes.
Dataset Field Name conventions
Field names must follow restrictions based on the Apache Lucerne field name conventions. These restrictions are described here.
- When the native field name is
id
- When the native field name is
id
, the dataset field nameid_
is generated. This rule is case-insensitive. - When native field names are non-alphanumeric
- Non-alphanumeric native field names are not allowed and automatically converted by the following
rules applied in sequence..
- URL encode native field names.
- Replace
%
by_
.
- When native field names are longer than 64 characters
- When native field names are longer than 64 characters, field names are truncated to 64 characters. This applies to converted non-alphanumeric names.
- Native fields generated by crawlers
- Some crawler types generate native fields with the prefix
__$
and the suffix$__
. In these cases, the dataset field names are generated by removing the prefix and suffix of the native fields. For example,__$Title$__
is changed toTitle
. - Document date field
- All crawlers specially handle the document date values. These values are stored as a data set
field named
Date
. - Document body field
- All crawlers specially handle the document body values. These values are stored as a data set
field named
Body
.