Introduction to the Vidispine Data Model
The Vidispine data model is constructed to be as flexible and efficient as possible. This will enable you to implement your workflows as you want, while still being ready for any future changes.
This short introduction will give you an overview of the entities in the data model that builds the assets in Vidispine, and are the base for everything that can be done in Vidispine. The main entities are; Item, Shape, Component, Files/Storage, Library and Collection, and they can be seen in the figure below.
The item is the most important entity in the data model, and corresponds to what usually is called an asset in other systems. The item is an abstraction of the physical content (essence) and holds information about the content (metadata).
Other entities, further down in the hierarchy, may also hold metadata. The item has the richest functions for how metadata can be stored, searched, and indexed. The item also holds information about which users that are allowed to read and modify information (access control).
A shape is a physical rendition of an item, and it can be different things depending on what kind of asset it is, e.g.:
- For a video – it can be a low-resolution editing version, a web version, etc
- For a document – it can be the pages as images, extracted text, etc
A shape can have one or several shape tags. The shape tags are used when Vidispine selects which files that are being transcoded, exported, thumbnailed, etc. A special shape tag is original, a shape tag that the imported source file gets. The shape tag also contains the recipe for how to create new shapes using the transcoder.
Each shape has one or more components with some content. Examples of content of a media shape are:
- A container component
- Video components
- Audio components
Each component corresponds to one file content. The component contains information (technical metadata) about codecs, resolution, frame rate and more.
File and storage
The file entity represents a physical file on a file system. The file is stored on a storage. Vidispine manages all files, and knows which copies of a file that have been made, and how they relate.
A library is a list of items. A library can be created manually, by adding the items to a library, or dynamically, by adding search results to a library. Libraries are useful when performing batch operations. Libraries can also be used when creating rules.
A collection is a list of items, libraries, and/or collections. Collections may have metadata and access rights, which are applied to the items that belong to the collection. While the library is typically created from a search operation, the collection is often used like a file system folder to organize items.
Further details about the Vidispine data model and how you can use it can be found in the API documentation, and in other posts on this site.