Metadata Extraction from CSV File
Metadata extracted from the CSV file has priority over the metadata extracted from native files.
For any metadata that you want to extract from the CSV file instead of the native file, create a mapping between the CSV field and a CORE field.
You may have to create additional fields in CORE Administration.
Select the appropriate Type for mapped fields, as described for field mappings configuration.
Note: You may protect CSV metadata against modification. Then specific fields are not changed by the crawler framework or the index engine, provided they are mapped to a CSV file field in the field mappings configuration.
Example:
Assume a CSV file containing this information:
ID; native file; Last Modified Date 1; marges.msg; 2015-12-01 2; constraints.msg; 2015-12-04
Further assume that you fill in the mappings table like this:
| Field name | Mapped to | Type |
|---|---|---|
| ID | xxx | ID |
| native file | xxx | File |
| Last Modified Date | rm_lastmodifieddate | Date |
Then the system processes the native files like for a standard file data source:
- It extracts metadata from the native file, but overwrites the last modified date information of the native files with the last modified date information of the CSV file.
- It extracts text from the native files.
- It extracts attachments, if there are any in the MSG files.
- It copies the files to the default native file storage. Files are tagged with the Copied - Only Crawl time copy supported value of the Native Status Smart Filter.