Database

The Archiva database is organized into several key concepts. It is useful to understand the principles when you browse or use the remote data toolbox.

Repositories - The remote data site is organized into several repositories. When you set up the RemoteDataToolbox class in Matlab, you specify which repository you will interact with. Currently, we have repositories for

• ISETBIO
• SCIEN
• VISTASOFT

Remote Paths - You navigate the repositories much as you would navigate through a directory tree. The root directory of a repository is '/', and a subdirectory might be '/resources/scenes'.

Artifacts - Data files uploaded to the database are stored in a remote path. If you upload two files with the same basename within a particular path, they will be stored in a single artifact. Thus, an artifact consists of a collection of files with a common base name (e.g., myFile) but different extensions (e.g., myFile.mat, myFile.jpg, myFile.zip).

Versions - You can store multiple versions of the files within an artifact. For example, the first time you upload 'myFile.mat', you might put it in version '1' (a string). The next time you upload the file, you can either over-write version '1', or you can choose to define a new version '2'. The versions are strings, not numbers, and they can be named anything you like.

Types - The type disambiguates the files within an artifact. The file myFile.mat has type 'mat', and myFile.jpg has type 'jpg'. A list of known types is {LIST GOES HERE}.

Archiva files

The Archiva database creates files that check the validity of the stored data. When using a web-browser to see the repository, these files may be visible to you. In fact, there are several ways to browse the repository, and these are described on the Browsing page.