Browse Prior Art Database

Analytics-Powered Pattern Recognition Enabling Learner-Agents to Process Real-World Data to Automatically Implement Mock Objects

IP.com Disclosure Number: IPCOM000241863D
Publication Date: 2015-Jun-05
Document File: 5 page(s) / 132K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a system for analytics-powered pattern recognition that enables learner-agents to process real-world data to automatically implement mock objects. The system generates learner objects such that when learning is enabled, the objects intercept calls to real objects and record inputs, outputs, object state (e.g., members, etc.), and object interactions (e.g., Hypertext Transfer Protocol (HTTP) calls, system input/output (I/O), etc.).

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 01 of 5

Analytics-Powered Pattern Recognition Enabling Learner-Agents to Process Real-World Data to Automatically Implement Mock Objects

Writing mock objects to simulate behavior for systems that are dependent upon functional units that encompass complex logic is expensive and/or prohibitive.

An idealistic approach suggests that every object is mocked and all possible paths are fully tested. This works upon a new, initial design, but quickly fails when trying to develop improvements in existing systems (without full Mock objects) or when integrating code from such existing systems.

Such complex systems exist, and measuring the general use of such systems should allow an intelligent application to observe, learn about, predict, and then simulate the behavior of those systems. This would then allow a developer to start with a simulated (mock) object that was automatically generated, and then extend that if/as needed, rather than beginning anew, from nothing, and implementing every feature that system provides.

While capabilities exist to automatically generate empty Mock classes, no known products actually implement simulated functionality.

The novel contribution is a system for analytics-powered pattern recognition that enables learner-agents to process real-world data to automatically implement mock objects.

The system generates learner objects such that when learning is enabled, the objects intercept calls to real objects and record inputs, outputs, object state (e.g., members, etc.), and object interactions (e.g., Hypertext Transfer Protocol (HTTP) calls, system input/output (I/O), etc.). With learning enabled, the system then exercises real objects through any means such that a varied real-world flow through those object methods is realized and recorded by the learner objects . This can involve real-world use, targeted tests of a system, existing automation on a system, etc. The system submits gathered data to an analytics engine to interpret the harvested data and make meaningful connections between input , output, and object state (including inferred states of external objects based on external interactions). Next, the system feeds back state definitions and input/output patterns to coded mock objects, allowing realistic mock interpretation as well as powerful...