Yes, i saw the repository structure and i saw that the internal storage of the nodes and content is not , i would say' human readable. Anyway i dont have to keep a one to one mapping as you say. I would like the binary data to be kept of an FTP for example, by making a new data store class handling that.
2007/10/15, Jukka Zitting <[EMAIL PROTECTED]>: > > Hi, > > On 10/15/07, henri godron <[EMAIL PROTECTED]> wrote: > > Here is my problem : whenever i start tomcat, i can see jackrabbit going > > through the init() but then if i copy/paste a file through the windows > > explorer (network favorites webdave client), its not going though my > class.. > > Note that Jackrabbit doesn't have a one-to-one mapping between nodes > in the JCR workspace and files in the underlying file system. The > FileSystem abstraction (and the LocalFileSystem implementation) within > Jackrabbit is used for managing internal data structures, some of > which contain information about the JCR nodes but not in a way that > would for example allow you to expose an underlying file system as a > JCR content tree. > > BR, > > Jukka Zitting > -- ________________ Henri Godron 06 80 00 59 71
