That sounds like a great feature of couchdb-lucene! We need to start on lucene integration soon, so it's good to know about that. This is more low-level, I guess. I think the most useful thing we have is a JSON tokenizer adapted to return Rhino ScriptableObjects. It's pretty fast. For instance, you could use the parser in place of the code in json2.js in Rhino.java to make that transformation run faster.
But I guess even more than code, I'm interested in how people are working with Rhino. For instance, we've found it incredibly useful to write unit tests against our reduce functions. I think about developing a mock couch in pure javascript, for use in integration tests. A On Mon, Mar 30, 2009 at 4:03 PM, Robert Newson <[email protected]>wrote: > That would be me... > > The Rhino integration in couchdb-lucene is constrained to providing a > user-defined transformation function when indexing documents, I assume > you're doing something more elaborate? > > All that code in couchdb-lucene is in Rhino.java fwiw. > > B. > > On Mon, Mar 30, 2009 at 11:59 PM, Paul Davis > <[email protected]> wrote: > > Closest I've heard would be the Rhino integration in couchdb-lucene at > [1]. > > > > HTH, > > Paul Davis > > > > [1] http://github.com/rnewson/couchdb-lucene > > > > On Mon, Mar 30, 2009 at 6:48 PM, Adam Wolff <[email protected]> wrote: > >> Hi,We're using CouchDB in a JavaScript server-side environment based on > >> Rhino. Anyone on this list doing the same? Interested in sharing code? > >> > >> A > >> > > >
