Extract Metadata from Files

The ExtractMetadata task extracts metadata from the file associated with a document. This task extracts a subset of the metadata obtained by standard KeyView filtering. It is faster than standard KeyView filtering and does not extract the file content.

TIP: When documents are ingested, CFS automatically extracts metadata. Do not use this task unless you have set the fields AUTN_NO_FILTER and AUTN_NO_EXTRACT on a document and want to extract basic metadata only.

The ExtractMetadata task is configured as a Pre task. Specify the name of the section that contains settings for the task. For example:

[ExtractMetadata]
Pre0=Lua:scripts/nofilter.lua
Pre1=ExtractMetadata:ExtractMetadataSettings

[ExtractMetadataSettings]
FieldnamePrefix=FIELD_
ReservedFieldnames=Reserved1,Reserved2

The Pre0 task runs a Lua script that adds the fields AUTN_NO_FILTER and AUTN_NO_EXTRACT to documents. Adding these fields prevents KeyView from filtering the documents and extracting subfiles.

The Pre1 task runs the ExtractMetadata task using the settings contained in the [ExtractMetadataSettings] section of the CFS configuration file.

The FieldnamePrefix parameter specifies a prefix for the names of the metadata fields that are added to the document. The ReservedFieldnames parameter specifies a comma-separated list of field names that the task must not use. If the task needs to add a metadata field with one of the specified names, it prefixes the name with an underscore. For example, with the settings specified above, the task would not add a field named FIELD_Reserved1. Instead, the task would add _FIELD_Reserved1.