You can think of a content repository item as consisting of content and metadata. For example, if your content repository includes repository items that are news stories, then the metadata might include a story’s byline, dateline, length, and keywords, while the content would include the text of the story itself. You can adopt one of two basic architectural styles when you set up a SQL content repository:
You can store both the content and the metadata in your SQL database. A content repository item would include a property whose value was the content.
You can store the metadata in the database, and the content in your file system. In this case, the metadata would include properties that indicate how to look up the content in the file system. A content repository item would include a property whose value was a pointer to the content in the file system.
As with other repositories, setting up a SQL content repository involves the following steps:
Design the item types you want to include in your content repository. For each type of repository item, decide what sorts of properties you want to have available to you for searching and targeting content in the repository.
Set up a SQL database containing content repository items, to act as the data store of the repository.
Create a repository definition. This is an XML file that describes the repository’s item descriptors and property descriptors, and defines the relationship among these items and the rows and tables of the database. See Creating a SQL Content Repository Definition.
Configure a SQL Repository component that interacts with the data store you set up in step 2 to create, modify, and retrieve repository items. See Configuring a SQL Content Repository.
A repository that contains content items must include item descriptors flagged with the folder
and content
attributes of the <item-descriptor>
tag in the SQL repository definition.