View through Connectors

Viewing with connectors allows you to view data from all repositories without keeping an additional store.

View through Connectors

This method uses the repository access that the connectors already have. You send View actions to the Distributed Connector, rather than sending a View action to the View component directly. This connector distributes the View action to the appropriate connector. The connector retrieves the file, and uses a connected View component to convert the file to HTML, before returning it.

Advantages:

  • You can view documents from all repositories that have a connector that supports the view action.

Disadvantages:

  • View server caching is not effective with this method, and it is not possible to highlight documents, or view embedded images.

  • The architecture and configuration is more complicated, with more components.

  • This method uses AUTN_IDENTIFIER for viewing rather than DREREFERENCE, which may be incompatible with older user interfaces.

Configuration

This method has the following setup requirements for indexing:

  • Documents must have an AUTN_IDENTIFIER, set to a reference field in IDOL.

  • Subfiles must use an identifier string in the form AUTN_IDENTIFIER|x.y.z where x, y, and z are the subfile indexes for each level. You build the subfile identifiers in Lua (Connector Framework Server includes an example script to do this).

  • Connectors must each set a unique AUTN_GROUP (controlled by the ConnectorGroup configuration parameter).

The method has the following setup requirements for viewing:

  • The connector must support the View action.

  • In the connector [ViewServer] section, you must set the Host and Port parameters to the host and port of your View server.

  • You must set the following parameters in the connectors:

Parameter Value Notes
EnableViewServer True  
EnableExtraction True You must also have filters present to view subfiles.
RegisterConnector True  
ConnectorGroup   Must have a unique value for each connector.
SharedPath   A path that the connectors and View server can read from and write to.

TIP: It is generally fine to use the same instance of the connector for viewing as you use for fetching documents.

If required, you can use a separate instance of the connector for viewing, as long as it has the same configuration to allow it to connect to the repository and interpret the identifiers correctly.