Python complex event processing

Question

Are there any Python alternatives similar to Esper (Java and .NET) that deal with complex event processing (CEP)?

Answer 1

Casual browsing indicates that this is not a very common problem domain for Python (although very interesting!). The framework that closest come to my mind is PEAK-Rules or dynrules.

There might be more, but not widely known (I'll search a bit more)

For your own digging: the place to find Python projects is first and foremost PyPI.

(@cues7a: Twisted, while indeed being an event framework, is too low-level to be comparable to Esper.)

Edit: It seems you can control Sybase's CEP products with Python

Answer 2

The ruleCore CEP Server uses dynamically loaded Python modules which can be used to extend it. The internal architecture is build around a central event bus which uses a pub/sub approach. So each module can subscribe to internal events and publish events based on logic inside the module.

Answer 3

The Kaskada project isn't exactly a CEP tool, but it has some similarities to CEP systems and provides a Python client designed for notebook environments like Jupyter.

For example, you can query for things like "how many pageviews have occurred since the last time a user made a purchase":

PageView | count(window=since(Purchase))

Kaskada's query language is based on the idea of a timeline, and many of the operations provided by CEP languages (followed-by, repeated, etc) have analogs in Kaskada's query language.

Disclaimer: I am a contributor to the Kaskada project.

Answer 4

What about Twisted Matrix?