1
edit
Changes
→Project News
''This is where your regular updates will go. In these you should discuss the status or your work, your interactions with other members of the community (e.g., Seneca and Mozilla), problems you have encountered, etc. Put detailed technical information into the Project Details page (i.e., update it as you go), and save this section for news about participation in the project.''
=== Dec. 09, 2006 ===
What has been done since last week?
* Got a test program and uploaded it to svn://cdot.senecac.on.ca/deltatest. The pristine working version is revision 4. The latest committed copy is revision 8. The regressive code was committed somewhere in between.
* Started testing the delta debugging framework.
The results of the testing?
'''Finding the minimal revision set/last known good revision'''
Works. The delta debugger correctly reverts to a previous revision, builds the source code, and runs the test case. The test case returns the proper results on whether or not it passes or fails. The delta debugger correctly stops at revision 4 - the last known good version.
'''Finding the minimal failure-inducing set of directories'''
Indeterminate. There is only 1 directory in the source repository so that directory should be returned as the minimal failure inducing set of directories. Does it return it? yes and no.
The delta debugger correctly applies all of the changes within that directory. And I think it correctly builds the source tree and runs the test case. However, the return code of the test case is not as expected. I expect the test case to report that the test fails, however, it reports that it passes. Thus, the delta debugger returns no directories as failure inducing.
However, if I force the test case to return the expected result, then the delta debugger correctly returns the directory as the failure-inducing one.
I suspect (or at least hope) that the indeterminate results of finding the failure inducing set of directories is because of a possibly unreliable or inconsistent test case. However, I can not be sure until I rule out the test case as the problem.
'''Finding the minimal failure-inducing set of files'''
Indeterminate. There are multiple source files in the repository. Does it return the correct failure-inducing source file? I don't know. I have the same suspicions for this as for the directory changeset.
Based on the testing, it seems to be able to cycle through every combination of changes in the changeset, apply the combination of changes, build the source code, and run the test case. The test case just seems to not report the correct test results.
=== Dec. 03, 2006 ===
