Data Models
Automated Software Quality Evaluation Framework Component
 
Description

Data Model group covers all the EMF data models used by Hyades project. We use EMF to generate 99.9% of the data model code (0.1% hand-coded extensions).

 
Development Resource

Model documentation generated from the HEAD stream version of the model:

Rational Rose models for EMF code generation
Packages, Classes and Attributes
Design Document

CVS Web links for data model plugins:

org.eclipse.hyades.models.cbe/
org.eclipse.hyades.models.common/
org.eclipse.hyades.models.hierarchy/
org.eclipse.hyades.models.sdb/
org.eclipse.hyades.models.statistical/
org.eclipse.hyades.models.test/
org.eclipse.hyades.models.trace/

 
Development Plan

Data Model group is organized in the following sub-groups:

  • Common (covers common package - UML2TP and Behavior model)
  • Logging (covers cbe package - Common Base Event model)
  • Statistical (covers statistical package - Statistical model)
  • Test (covers test package - Test related model, extenstions to common)
  • Trace (covers trace and hierarchy package - Trace model and hierarchy/resources related artifacts)


  Common

  • Content coming soon !

  Logging

  • Content coming soon !

  Statistical

  • Content coming soon !

  Test

  • Here are the next steps that we plan to address in the Test Model sub-group:
    1. Continue to identify key attributes and incorporate feedback from UI and Execution working groups
    2. Consider the implementation of 1 or more non-mandatory façades to facilitate Test Model interop.  Begin defining a façade for suite behaviors.
    3. Organize the Data subgroup to begin defining the test data package with the trace team (convene within 1-2 weeks)
    4. Continue specifying the common configuration package, addressing where we match up with the Trace Model team.

  Trace

  • Trace sub-group target deliverables for version 1.0.0:
    1. Improve, complete, and correct documentation for every diagram, class, enum, attribute, and association.
    2. Resolve the issues represented by red boxes in the trace model.
      a. Design an aggregation/annotation system for storing tool-computed, "rolled up" data like durations and averages.
      b. Design a representation for the data that is more compact when less information is being collected.
    3. Design how to represent a heap dump - that is, a snapshot of objects in the heap at a given moment, and their references to other objects at that moment.
    4. Design the representation of line level trace information.