Document Object Management
The Repository API is used by Arbortext Editor to access common document management services provided by the underlying repository software.
The supported operations include:
• Access Control—The Repository API does not provide any access control scheme; it simply provides the hooks for a string describing a document's access privileges to be passed back and forth. The actual granting or denial of access to a particular document by a particular user is the responsibility of the repository. (Arbortext Editor does not support any modifications of repository permissions from Arbortext Editor; access control specifications must be made in the repository directly.)
• Revision Control —The Repository API revision control system scheme is general enough to encompass most revision control schemes. It is inspired by RCS, but has the major difference that it creates a persistent working version of an object as soon as a parent revision is locked. This frees you from the necessity of managing a local version of a locked object until it is checked in. Instead, the working version of an object is immediately created, all subsequent save operations will update the working version in place, then the Object check in operation makes the working version the latest version in the revision list.
Document objects may be checked out and in from Arbortext Editor. When a document object is checked out, a new revision is created in the database. Subsequent saves will update this new revision; and the object may be checked in to make the new revision permanent, or unlocked to remove it. Each revision of a document object has a unique POID, so the Arbortext Editor user can access any version of a document object.
• Document I/O and management—Document objects may be loaded and saved from within Arbortext Editor. Document objects may also be copied or deleted with the Repository API. Repository API routines do not attempt to provide a full range of document object manipulation and administration facilities; those tasks should be handled by the utilities in the underlying repository.
• Accessing Document Object Metadata—The Repository API requires a base set of metadata fields. Additional metadata fields may be defined by customers or Application Developers to the extent that is feasible on a given platform and for a particular Arbortext Editor application. The Repository API and the Arbortext Editor ACL allow a passthrough between a custom user interface and a custom database to store and retrieve any value of any metadata field.