I've written a DWR client in Python, which I use for automating tests run against a DWR-based website. Up until now, all the calls I make are single, non-batched requests, and I make several requests, all in a single-threaded manner.
I noticed that when I use a browser, some of the calls are batched up into a single request. Aside from the bundling, are there any differences with how the calls are handled by the server? I believe they're parsed back out and handled serially, and from what I've tested, there are no noticeable differences. However, I'm concerned there might be some hidden pitfalls, and I need my python client to simulate browser usage.
In almost all cases, batching acts the same way as a bunch of synchronous unbatched calls. See the Caveats section of DWR Batching for potential pitfalls.