Spencer Park
Spencer Park
As mentioned in #61, magics are not transformed before evaluating startup scripts which makes adding dependencies in those places much more awkward. The kernel should transform such scripts before evaluating.
The introspection supports rendering richer data than just text and we should take advantage of that to render the javadocs as HTML.
Another small PR but small PR's get merged quicker :) This one prints _all_ arguments passed to `console.log` and also makes `__unicode__` work as expected in py3. `repr` still used...
I'm looking into implementing the kernel side of the debugging protocol and had some questions regarding [the messaging spec](https://jupyter-client.readthedocs.io/en/latest/messaging.html). Specifically regarding the request/reply sequencing and expectations regarding the reply order....
Compatibility is a bit tricky because the current "ipc" implementation isn't ipc, so anyone relying on the jeromq tcp loopback implementation intentionally will break. I think this would only really...
Add infrastructure for building debuggers on top of existing DAP servers
For kernels running on java 16 we should be able to use unix domain socket channels to connect to a jupyter client with `ipc` configured as the transport. Depends on...
Following discussion in SpencerPark/IJava#61, in the api package we can outline the extension loading convention and interfaces for loading via the [ServiceLoader](https://docs.oracle.com/javase/8/docs/api/java/util/ServiceLoader.html). An interface like `JupyterExtension` would be implemented by...