The Railroad Repository: Managing large files 'around' a CMS
The Railroad Server provides facilities to manage large files from within a CMS, but handles the upload and viewing of the files ‘around’ it. This means a CMS isn't affected by the upload of, say, a 50 GB video. In actuality only a reference to the Railroad resource is stored. This ensures that bandwidth of both upload and public viewing do not interfere with CMS performance.
Railroad is friendly to sysadmins and power users because it uses WebDAV in conjunction with Apache for file management. This means the repository can be mounted as a disk, and files can be managed in bulk. Metadata is attached to WebDAV file properties, but is read/writeable from the CMS.
The Railroad Server has a layered architecture of file system storage, WebDAV, and services, plus a future distribution layer. The services provide search, delivery of indexes, metadata sets, user management, and content manipulation tools such as image and video transformation.
Railroad resource listings and search results for a CMS are represented in XML, enabling each client system to transform data into appropriate presentation.
The Railroad Project developed in the Python community. Server infrastructure is written in Python, and makes use of existing tools such as libxml and PIL (Python Imaging Library). WebDAV properties are stored in a database (Postgres) to facilitate searching. Railroad is designed so installation complexity can be stepped depending on need.
The presentation will include slides and a demonstration of a working Railroad Server integrated into the Silva CMS.
Presenter: Kit Blake