Github user agoodm commented on the issue:
https://github.com/apache/zeppelin/pull/1439
@doanduyhai To be precise, I designed the callback registry system with two
different types of use cases. The first one is for the different interpreter
maintainers to implement in their individual subclasses of `Interpreter`, which
would be used for use cases that are specific for better integrating certain
features / libraries of that interpreter with zeppelin. Right now a prime
example of this is matplotlib integration with python / pyspark as I have
discussed in more detail in the JIRA issue.
The second set of use cases are those in which the API is directly exposed
to notebook users. Right now this is indeed only fully available in spark /
pyspark via the `ZeppelinContext` class. The reason for this is that getting
this sort of interface to work with every interpreter would require some sort
of language specific API that can be used to expose the JVM (as py4j does with
python), hence one reason I labeled it as experimental in the docs. It's a
similar limitation to the `angularBind()` and `angularWatch()` API. Keep in
mind though that the registry can still be applied for other interpreters if
the replName is manually specified, eg:
```scala
%spark
z.registerCallback("post_exec", "ls -l", "sh")
```
Does this help clarify things?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
