E&S CVS Commit: PEAK - Refactoring to support uniform internal data structure for features. All

pje at eby-sarna.com pje at eby-sarna.com
Thu Feb 6 14:44:00 EST 2003


Module Name:	PEAK
Committed By:	pje
Date:		Thu Feb  6 19:43:19 UTC 2003

Modified Files:
	PEAK/src/peak/model: structural.py
	PEAK/src/peak/storage: xmi.py

Log Message:
Refactoring to support uniform internal data structure for features.  All
features with a maximum multiplicity of 1 are now stored simply as the
object representing their value.  (Previously, 'model.Reference' features
were represented as a list containing zero or one items.)  This should
make it easier for most people implementing data managers to get the
object state correct, but it was necessary to make some changes to the
current XMI DM to support it.

This change is part of a general trend towards encapsulating all of
features' behavior in one or two base classes, with the current Reference,
Field, Sequence, etc. classes becoming just shorthand for specifying the
appropriate metadata.  (This will ultimately make it easier to do code
generation that's based on pure metadata from a MOF or UML model.)


To view diffs of this commit, you can use the following URL(s):
http://cvs.eby-sarna.com/PEAK/src/peak/model/structural.py.diff?r1=1.11&r2=1.12
http://cvs.eby-sarna.com/PEAK/src/peak/storage/xmi.py.diff?r1=1.12&r2=1.13

To generate a diff of this commit:
cvs rdiff -r1.11 -r1.12 PEAK/src/peak/model/structural.py
cvs rdiff -r1.12 -r1.13 PEAK/src/peak/storage/xmi.py

Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.




More information about the source-changes mailing list