One thing that might be a possibility here is the use of the WebDAV protocol.
WebDAV is technically an extension of HTTP, is implemented in the webserver you are already using (although it might need to have the module be enabled...) and requires no active back-end intrepreted language such as PHP, Perl or Python. WebDAV allows the typical HTTP operations for GET, PUT, POST, and HEAD, but adds to those with the additional operations of: PROPFIND, PROPPATCH (used for managing 'properties' but could be database elements in collections), COPY, MOVE (designed to remotely move server-side files), SEARCH, BIND, and some others... WebDAV is designed to work on atomic items remotely, meaning you don't need to "GET" the entire file, you can specify to work with an individual record out of that file. I believe this primarily works with XML elements. It is theoretically possible to have your "database" be a structured set of directories and files, such as: http://server/webdav/dir1/dir2/file-area and the 'webdav' handler in that path would use dir1/dir2/file as the equivalent of the REST-like "dir1=foo&dir2=foo&file=foo" type operands in a GET request. At the file level, you could actually have the file be an XML file, containing numerous records. WebDAV also provides locking mechanisms so that multiple writers are not conflicting or locking each other on writing to that file. While I haven't tried this, it is certainly worth exploring - if you are allowed to run the WebDAV extension (I don't know, your IT administrator might rule that out too.) I also haven't any idea if node.js, jquery, or similar Javascript frameworks have built-in WebDAV client support. Some of the more complex implementations implement WebDAV through custom server-side code, Java, PHP or the like. You'll need to focus on what you get with the base protocol as you would have enabled in Apache. Check out these: * http://www.webdav.org * http://www.webdav.org/specs/rfc4918.html * XDB is open-source and uses native WebDAV to store/retrieve semi-structured, schema-less data. (Probably requires back-end Java or something though...) * Milton (http://milton.ettrema.com/index.html) is a Java process (perhaps not exposed through Apache but runs in it's own external process through Tomcat) which provides WebDAV for remote applications (looks like it's designed for files/folders, but not database files.) Well there might not be anything which automatically fits the bill, but with some ingenuity you should be able to cook up something that exposes your data elements over basic WebDAV-Apache (but would have to be stored in files since you can't use a back-end language). It's also an area I'm interested in, just due to the general 'NoSQL' concept, and distributed content access over a cloud-like system. (A pattern where thousands of current users need to access the same set of content which could be held in files, but only 1 or 2 users would ever update those files remotely - the shared filesystem such as NFS on the backend is the storage - no SQL needed.) DK On 28Feb2011 06:15PM (-0700), Chris Louden wrote: > Totally. The content can be loaded into a div on a page of HTML using > a single call in jQuery that utilizes the load() function. _______________________________________________ LinuxUsers mailing list LinuxUsers@socallinux.org http://socallinux.org/cgi-bin/mailman/listinfo/linuxusers
