3
edits
Changes
m
'''Project Facts:'''
# Bonsai is a tool that can produce a list of differences between versions of a source file. (Bonsai's functionality has not been examined closely yet but will have to as it may be a key component to the framework)
'''Possible Vision of the Delta Debugging Framework''':
(subject to change based on stakeholder consultation/feedback, feasibility study)
# Since the last time a developer executed a test case that passed, the developer modified some source files. The source files may be of the same type or mixed type, same directory or different directory. It shouldn't matter. The framework should be source type and location agnostic. Upon executing the test case again, the result is now a failure. The developer panics. It's only days before the deadline to submit bug patches before the source tree is supposed to be closed for release and the bug is a blocker. The developer doesn't want to be shamed for delaying the release, and the source code is too complex to find the bug in time, so what should they do? Use the delta debugging framework! that's what. How? you may ask. Well keep reading to find out. <small>* scenario may vary.</small>
# The delta debugging framework may require the developer to input one piece of information. The test case/function that used to pass but now fails. It will be used to determine whether the source files with progressive changes passes/fails the test.
# Once the developer has inputted this piece of information, it will use Bonsai to query the source tree and compile a list of all the changes to the source files since a certain amount of time.
# (If there was a method of determining change dependencies so as to eliminate the possibility of inconsistencies, it would be done in this step. One possible way of reducing the possibility of inconsistencies is to logically group changes by location or check in time.)
# This step would be where the delta debugging algorithm would come into play. The algorithm should basically:
## Recursively, incrementally remove changes from the source code with the regression.
## Recompile the source tree.
## Execute the test case. There may be 3 outcomes:
### The test case passes. We know that the failure-inducing change(s) are in the change(s) that were removed.
### The test case fails. We know that the failure-inducing change(s) are not exclusively in the change(s) that were removed. I say not exclusively because of the concept of Interference (described above).
### The test case is indeterminate. There were some inconsistencies.
Assuming you have [http://subversion.tigrisMost of the classes in blue exist in the source repository.org/ SVN], the projectThe classes in pale yellow are classes that won's source can t be obtained via SVN using completed in the following commandfirst release. [[Image:Dd_partialclassdiagram2.PNG]]
svn checkout svn://cdot.senecac.on.ca/deltadbg
<table style''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.'' =="width: 100%;" class="standard-table" borderDec. 22, 2006 ="1" cellpadding="2" cellspacing=" I haven't posted an update in a while. So what's been done? I finally had some time to do a second round of testing & debugging of the delta debugging framework. And guess what? It ''seems'' to work now. The problem? Combination of logical errors when applying and unapplying changes in the framework and a bad test case. Go figure. However, before I get ahead of myself and officially tag and release the delta debugging framework as version 0">.1, I would like to test it out on another test program. Hopefully, this can be done this weekend. And if all goes well, version 0.1 will be officially released before the end of the year. <tr><td colspan="4"><strong>Legend</strong></td></tr> <tr> <td bgcolor="#00FF00" style="width: 25%;">Task completed</td> <td bgcolor="#FFFF00" styleDec. 13, 2006 ="width: 25%;">Task started but not complete</td> <td bgcolor="#FFA500" style="width Created [[Delta_debugging_testcases|Delta Debugging Testcases]] page to discuss the nature of the test cases created to test the algorithm. Included in the page are 2 testcases created so far, the '''HelloWorld''' binary test and the '''Sudoku''' test. Both tests can be found in the '''deltatest''' svn repository. The repository can be checked out using this command: 25%;">Task being discussed</tdpre> <td bgcolor="#FF0000" style="widthsvn checkout svn: 25%;">Task not started</td> </tr>cdot.senecac.on.ca/deltatest </tablepre><br /><table style="widthExactly 12 days before Christmas, the delta debugging framework has been released under the [http: 100%;" class="standard-table" border="1" cellpadding="0" cellspacing="0"> <tr> <th>Task</th> <th>Description</th> <th>Assigned to<www.opensource.org/licenses/th>gpl-license.php GPL Version 2] License. <th>Status</th> <Unfortunately, we haven't had the time to test the delta debugger much since Dec. 09, 2006 because of exams, other school work; Planning to spend some time this weekend to test the delta debugger and figure out why it currently seems to not be able find the minimal set of failure inducing directories/tr>files (whether its because of unreliable test case or logical error in the program).
<tr>
<td colspan="4"><strong>Change set / Change</strong>
</td>
</tr>
<tr> <td>Retrieval of Change / Change set</td> <td>The Granularity concept. A single revision may consist of hundreds or thousands of lines of code that were changed, yet only a couple lines of the change may be responsible for the regression. Thus, There must be a method to break the change into smaller manageable chunks. The different types of chunks we may breaking up a changeset are: Revision, Directories, Files, Code Blocks, and Lines.</td> <td>[[User:RichardChu|Richard Chu]]</td> <td> <table style="width: 100%;" class="standard-table" border="0" cellspacingDec. 11, 2006 ="0" cellpadding="0"> <tr><td bgcolor="#00FF00">Currently can retrieve change sets of type Revision, Directory, and File. NOT going to complete retrieval of Code Block, Line of Code change set.</td></tr> <tr><td bgcolor="#FFFF00">Requires more thorough test suite (ChangesetTest.pl needs more test cases)</td></tr> </table> </td> </tr>
<tr> <td>Application of Change / Change set</td> <td>OKUploaded testcase for '''HelloWorld''' binary at '''deltatest svn'''. Change sets can be retrievedThe test simulates the error that may occur when compilation fails due to syntax error. Now what? You must be able to apply a change or change set or subset of a change set to The exalted HelloWorld program is located on the HelloWorld directory, while the source treetest definition is at HelloTestCase1. Your mission is pm. The algorithm detects failed test and reverts the affected file to figure out how to do thatthe version where the test passes.</td> <td>[[User:RichardChu|Richard Chu]]</td> <td>
<table style="widthNote for the future: 100%;" class="standard-table" border="0" cellspacing="0" cellpadding="0"> <tr><td bgcolor="#00FF00">Can apply a changeset (specified by array of indices passed in) for a Revision, Directory, and File Changeset. NOT going to complete application of Code Block or Line changeset.</td></tr> <tr><td bgcolor="#FFFF00">Requires test cases (ChangesetTest.pl is outdated.)</td></tr> </table>improve user feedback functions!
</td>
</tr>
<tr> <td>Unapplication of Change / Change set</td> <td>Changesets obviously must be able to be applied. But changesets must also be able to be unapplied. Your mission is to figure out how to do that.</td> <td>[[User:RichardChu|Richard Chu]]</td> <td bgcolor="#FFFF00"> <table style="width: 100%;" class="standard-table" border="0" cellspacing="0" cellpaddingDec. 10, 2006 ="0"> <tr><td bgcolor="#00FF00">Can unapply a changeset (specified by array of indices passed in) for a Revision, Directory, and File Changeset. NOT going to complete unapplication of Code Block or Line changeset.</td></tr> <tr><td bgcolor="#FFFF00">UNTESTED.</td></tr> </table>
<Where is the CVS/td>Bonsai work heading? Here is a breakdown of the past 3-4 weeks: <* Initially was going for a straight wrapper around CVS ala the style Richard used for SVN.* Tried to find some functionality within Bonsai that could make it easier.* Talked to Reed Loden, he set up a repository for us to try with. Thanks Reed!* Thought that there may be some additional (read: unpublished) tools that could be worked with. Got in contact with some of the "Project Participants" listed on [http://www.mozilla.org/projects/bonsai/]. Was told the person in particular wasn't a contributor (just submitted a bug report). They in turn pointed me to [irc://irc.mozilla.org/#mozwebtools #mozwebtools].* Lurked on [irc://irc.mozilla.org/tr>#mozwebtools #mozwebtools] for a few weeks. Talked to 'justdave' about Bonsai. Reed Loden chimed up and informed me that Bonsai can output to XML using ?xml=1 on the query (score! thanks again).* Researched some PERL parsing utilities. Trying out XML::LibXML for DOM-style parsing.* Hopefully wrap something up by Wednesday. Failing that, might just go with simple CVS wrapper of some sort.
<tr>
<td colspan="4"><strong>GNU Make</strong> ([http://www.gnu.org/software/make/ http://www.gnu.org/software/make/])
</td>
</tr>
<tr> <td>Wrapper around the GNU make utility</td> <td>Mozilla uses the GNU make utility to build their source tree. your mission is to make a wrapper around the GNU make utility so that the make command can be programmatically called to build the source tree.</td> <td>[[User:RichardChu|Richard Chu]]</td> <td> <table style="width: 100%;" class="standard-table" border="0" cellspacingDec. 09, 2006 ="0" cellpadding="0"> <tr><td bgcolor="#00FF00">Wrapper created: ''makewrapper.pl''. Can execute the make command with options specified by the user.</td></tr> <tr><td bgcolor="#FFFF00">Requires more thorough test case (''maketest.pl'' needs more test cases).</td></tr> </table> </td> </tr>
<tr>What has been done since last week? <td colspan="4"><strong>Subversion (SVN) Repository</strong> ([http* Got a test program and uploaded it to svn://subversioncdot.tigrissenecac.org/ http://subversionon.tigris.orgca/], [http://svnbookdeltatest.red-bean.com/nightly/en/indexThe pristine working version is revision 4.html http://svnbookThe latest committed copy is revision 8.red-beanThe regressive code was committed somewhere in between.com/nightly/en/index* Started testing the delta debugging framework.html]) </td> </tr>
<tr> <td>Wrapper around The results of the necessary SVN commands</td> <td>For the automated debugging to work, we may need to automatically modify the working copy by reverting to a different revision or updating certain directories and files. It may also need to know the differences between revisions and changesets.</td> <td>[[User:RichardChu|Richard Chu]]</td> <td> <table style="width: 100%;" class="standard-table" border="0" cellspacing="0" cellpadding="0"> <tr><td bgcolor="#00FF00">Wrapper created: ''svn.pl''. Currently has subroutines for ''commit'', ''update'', ''diff'', and ''checkout'' commands. May need to wrap other SVN commands.</td></tr> <tr><td bgcolor="#FFFF00">Requires more thorough test case (''svntest.pl'' needs more test cases).</td></tr> </table> </td> </tr>testing?
<tr> <td>Query SVN repository for differences between two revisions</td> <td>Your mission is to find out the relevant commands that can return the differences between two revisions, '''Finding the meta-data that is kept with each minimal revision, how differences between two revisions are stored and formatted, and how this data can be parsed into a usable form for our project (wrapper?).<set/td> <td>[[User:RichardChu|Richard Chu]]</td> <td bgcolor="#00FF00">Done.</td> </tr>last known good revision'''
<tr> <td colspan="4"><strong>CVS/Mozilla Bonsai</strong> ([http://wwwIndeterminate.mozillaThere is only 1 directory in the source repository so that directory should be returned as the minimal failure inducing set of directories.org/bonsai.html http://www.mozilla.org/bonsai.html], [http://cvsbook.red-bean.com/OSDevWithCVS_3E.pdf CVS Book])<br />In my mind, Bonsai may be too bloated for our needsDoes it return it? yes and no. </td> </tr>
<tr> <td>Wrapper around The delta debugger correctly applies all of the necessary CVS commands</td> <td>For changes within that directory. And I think it correctly builds the automated debugging to worksource tree and runs the test case. However, we may need to automatically modify the working copy by reverting to a different revision or updating certain directories and files. It may also need to know return code of the differences between revisions and changesets.</td> <td>[[User:RichardChu|Richard Chu]]</td> <td bgcolor="#FF0000">Not startedtest case is not as expected. Will this be started before I expect the end of test case to report that the semester? i doubt test fails, however, itreports that it passes. After Thus, the end of the semester? most likelydelta debugger returns no directories as failure inducing.</td> </tr>
<tr> <td>Query CVS repository for differences between two revisions</td> <td>Your mission is However, if I force the test case to find out the relevant commands that can return the differences between two revisions, the meta-data that is kept with each revision, how differences between two revisions are stored and formattedexpected result, and how this data can be parsed into a usable form for our project (wrapper?).</td> <td>[[User:RichardChu|Richard Chu]]</td> <td bgcolor="#FF0000">Not started. Will this be started before then the end of delta debugger correctly returns the semester? i doubt it. After directory as the end of the semester? most likelyfailure-inducing one.</td> </tr>
<tr> <td colspan="4"><strong>Test CaseI suspect (sor at least hope)</strong> ([http://wwwthat the indeterminate results of finding the failure inducing set of directories is because of a possibly unreliable or inconsistent test case.mozillaHowever, I can not be sure until I rule out the test case as the problem.org/tinderbox.html Tindexbox])<br /> </td> </tr>
<tr> <td>Creation / Extraction of Test Case(s)</td> <td>We need test cases that can return whether or not '''Finding the test passes or fails. Tinderbox has a couple minimal failure-inducing set of tests that are executed after the source is built. Extract those tests from the Tinderbox source code so that we can use them in this project. We also need a test case that can pass/fail consistently so that we can test the delta debugger.</td> <td>[[User:Ankuswan|Aditya Nanda Kuswanto]]</td> <td bgcolor="#FFFF00">Work in progress. Found the tests! Now need to figure out how to run them and how they work.</td> </tr>files'''
<tr> <td>Test Framework</td> <td>We ideally need a way to allow users to specify the test(s) to be run easily without them having to modify the delta debugging moduleIndeterminate. </td> <td>TBD.</td> <td bgcolor="#FFA500">There are differences multiple source files in opinion on how the repository. Does it return the correct failure-inducing source file? I don't know. I have the same suspicions for this should be doneas for the directory changeset.</td> </tr>
<tr>
<td colspan="4"><strong>Implementation of Delta Debugging Algorithm</strong> ([http://www.infosun.fmi.uni-passau.de/st/papers/tr-99-01/ Yesterday, my program worked. Today, it does not. Why?])<br />
</td>
</tr>
<tr>
<td>The Algorithm</td>
<td>The delta debugging algorithm. Drives the framework to retrieve change sets, apply changes, build source tree, run test case(s) to find the minimal set of failure inducing changes. The intersection of all other parts of the framework to make them work together. Ideally, should be abstract enough for easy extensibility with little impact.</td>
<td>[[User:dwwoodsi|Dean Woodside]]</td>
<td bgcolor="#FFFF00">Work in progress.</td>
</tr>
</table>
== Project References ==
([[#top|↑ top]])=== Dec. 03, 2006 ===
[httpCommitted some updated to the SVN repository.* The test framework. There are a couple of files to the framework://programmingTest.newsforgepl, TestCase.com/articlepl, TestSuite.pl, TestResult.pl, TestRunner.pl. It is loosely based off of the design of the JUnit framework. Why such an elaborate design just for the need of users to define the test case that can determine whether or not a piece of functionality works or not to be run?sid=05/06/30/1549248&from=rss NewsForgeFor a few reasons that I may be adamant about: An Introduction to Delta Debugging]Delta # To use the delta debugging simplifies debugging process in a program by automating framework, the process and continually splitting user should not have to touch the program DeltaDebugger.pl file to smaller chunks called deltas. This technique is useful in three circumstances:*Error occurs due define the tests and how to user inputs (erun them.gUsing the testing framework, this can be done by subclassing the TestCase. keypress, file I/Opl class and overriding the run()subroutine. Delta debugging is used # For the delta debugger to eliminate user actions irrelevant work, it needs to know whether the nature of test case passes or fails. Using the error and pinpoint test framework, I hope to control the cause possible return codes of the errortests to either pass or fail only.*Error occurs due to recent changes to testtest.pl that tests the code. In this situation, deltas are retrieved from functionality of the net differences from both codestest framework.*Multithreading environmentupdates to DeltaDebugger. Delta debugging can track down the exact order pl to make use of operations originating from multiple threads that caused the errortest framework.
== Points of Confusion ==
(Crunch time. One week left. The high priority tasks that still need to be done:# Acquisition of a program we could use to test the delta debugging framework. See [[#topHow_to_Get_Involved|↑ topHow To Get Involved]])for more info.# Test, debug the delta debugging framework.
[[Bonsai issue]] -- '''unresolved'''
When Committed some updates to the SVN repository.* Updated the delta debugging algorithm module. I get confused, didn't realize this yesterday but the algorithm to find the minimal set of failure inducing files (and code block and line of code changes if those changeset types ever gets dones) is the same (with minor modifications) as the algorithm that can find the minimal set of failure inducing directories. Thus I draw diagramsgeneralized that algorithm to remove the directory changeset specific code so that it will work with all other types of changesets.* Removed the debugging/test related code from the source files.CVS Repository Setup (thanks to [[user:reed|Reed Loden!]]): '''hera.senecac.on.ca/deltatest'''*[http://hera.senecac.on.ca:43080/viewvc.cgi/?root=deltatest ViewVC Web Repository Browser]*If you want commit access for whatever reason, email one of the project members
[[Image:Dd_partialclassdiagram=== Nov.PNG]]25, 2006 ===
'''The ClearCommitted some updates to the SVN repository.* Modified the Changeset hierarchy of classes. Added a getChange() subroutine that takes an index and retrieves the change from the changeset. Also modified the getChangeset() subroutine to optionally take an array of directories/files to limit the search scope to within the directories/files passed in. These changes are possibly dangerously untested.* Committed the DeltaDebugger.pl file. This file houses the actual delta debugging algorithm. It requires three user-defined pieces of information: Seemingly Straight Forward'''a Build object, an RCS object, and the automated test cases. Currently, it can theoretically find the failure inducing revision, and the minimal failure inducing set of directories. * Committed a DeltaDebuggerTest.pl file. It just tests the correctivity of the theoretical.
The RCS tree is straight forward. It will encapsulate the data and operations related to the revision control system. SVN wraps the operations of the SVN revision control system, CVS will wrap the operations of the CVS revision control system, etc.
The Build tree is straight forwardIn the works:* Continue working on the delta debugging algorithm. It wraps Need to be able to find the build tool used minimal failure inducing set of files.* Test framework. Allow users to build plug in test cases/suites without touching the source treeDeltaDebugger.pl module.
'''The Blurry: Current Points of Confusion'''
RCS's can remember The deadline for a version 0.01 release is looming. 1-2 weeks left to get this done. What needs to be done to accomplish this?* Finish everything that is in the works real soon.* Need a test program that we could use and upload to our test SVN repository to test the delta debugging framework. Ideally, the test program will meet the changes (deltas) following requirements:# Has source files that span multiple directories wide and deep yet be small enough that occurred the delta debugging can be done in previous versions a short amount of time so that all aspects of the delta debugger can be tested.# Has a file, regression. Or can easily be modified so that some functionality will stop working.# Has an automated test case that tests the regressive functionality.* Put theory into practice. So far the history delta debugging algorithm has not been tested on a real program. The correctness of changes that occur between revisions, etcthe algorithm has only been confirmed in theory. We need to test the algorithm in a production environment real soon.
A Changeset and its subclasses will encapsulate the idea of a set of changes. A set of changes could be broken down into various categories such as a specific revision, a list of directories, a list of files, a list of blocks of code, and finally a line of code.
A Change and its subclasses encapsulate the idea of a single change=== Nov. A change can be a change made within a directory19, change made within a file, change made to block of code, or a change to a line.2006 ===
A ChangesetFactory is supposed to return a change set based on the type of change set requested. To get the requested change set, one needs to know The earlier crash case we had (see the type of revision control system (SVN, CVS, other, etc.update directly below) and/or the data required to connect to it. So was a non-regressive bug--there obviously need a link between RCS and ChangesetFactory/Changeset. The question is how? What is the proper/best way to link them together? One way is to pass in an RCS object to the ChangesetFactory which would then pass that object to the appropriate Changeset subclass. I don't like was no former build that solution but worked with it's the simplest.
Also, the method Going to get a change set for SVN may be different from CVSuse [https://bugzilla.mozilla.org/show_bug. So there may be a Changeset hierarchy for SVN and another one for CVScgi?id=325377 Bug #325377] instead. I don Having difficulty identifying when it was first introduced--the information in the bug report doesn't like seem to be quite accurate. Using the idea of that nightly builds as archived at all[http://archive.mozilla. There must be another wayorg/pub/mozilla/nightly/ http://archive.mozilla.org/pub/mozilla/nightly/] to narrow it down.
'''The Blind: Future Points of Confusions'''Fortunately this crash is easily automated and does not require user interaction.
* Applying a change in a changeset. Should the Changeset subclasses be able to do that? Are they the information expert? They know about the changes. Should they know how to apply them? How would we go abouts applying a subset of changes in a changeset? For example, there may have been changes in 10 different directories, how would we apply the changes from say 4 of the 10 directories and not the others?
* Connecting all 3 hierarchies together. Need to be able to connect to SVN, need to be able to get and apply changes, need to be able to build the source tree.
* The actual delta debugging algorithm.
But that's all for the future=== Nov.18, 2006 ===
''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.''
([[#top|↑ top]])
[http://en.wikipedia.org/wiki/Delta_Debugging Delta debugging] is an automated approach to debugging that isolates failures systematically. Given a failing test that can be mechanically verified (including a browser crash), [http://en.wikipedia.org/wiki/Delta_Debugging delta debugging] is a way of automatically isolating the change that introduced the failure. Having For developers, the occurrence of a scenario similar to this happens all too often: a developer codes a piece of functionality that works. Then, over a period of time, multiple changes to the program source files are made and that piece of functionality stops working. The cause of the regression could be any of the changes that were made since the time the functionality was last known to work. To isolate the cause of the regression, the developer begins the debugging process. Generally, debugging is a manual process where the developer must walk through the code while trying to keep track of variables and function calls. Sure, there are debuggers that can help you keep track of variables, the call stack, watch certain blocks of code, and execute the code step by step, however debugging is still mainly a manual process. Written in perl, given# that the source code is located in an SVN repository (support for CVS in the future)# a test case that can automatically verify whether or not a piece of functionality of a program works or not# a way to automatically build the program from the source code (if needed)the delta debugging framework aims to automatically isolate the failure-inducing changes to the source code that caused a regression. == Project License == Written in place perl, given a test case that can automatically verify whether or<br />not a piece of functionality of a program works or not, the delta debugging<br />framework aims to pull builds from CVSautomatically isolate the failure-inducing changes to the<br />source code that caused a regression.<br /><br />Copyright (C) 2006 Richard Chu, Aditya Nanda Kuswanto, bisect Dean William Woodside<br /><br />This program is free software; you can redistribute it and/or modify it under<br />the terms of the GNU General Public License as published by date and change set the Free Software<br />Foundation; either version 2 of the License, or (using at your option) any later<br />version.<br /><br />This program is distributed in the hope that it will be useful, but WITHOUT<br />ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS<br />FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.<br /><br />You should have received a copy of the GNU General Public License along with<br />this program; if not, write to the Free Software Foundation, Inc., 59 Temple<br />Place, Suite 330, Boston, MA 02111-1307 USA<br /><br />[http://enwww.wikipediaopensource.org/wikilicenses/Bonsai_CVS_code_management_system bonsai ] data gpl-- remember, CVS doesn't have changesets!license.php The GNU General Public License (GPL)Version 2, and report results would let computers make developers more productiveJune 1991]<br /><br />Contact Information about the authors of the delta debugging framework can be found<br />on our individual profile pages.
== Project Leader(s) ==
''Name(s) of people casually working on the project, or who have contributed significant help. Include links to personal pages within wiki. <br />NOTE: only Project Leader(s) should add names here. You '''can’t''' add your own name to the Contributor list.''
[[User:Reed|Reed Loden]] - Setup the CVS repository for us with the web front-end. Provided direction in querying Bonsai (a means to extract the output in XML).<br>
[[User:Elichak|Liz Chak]] - Documentation of the subroutines in the svn.pl and makewrapper.pl source files.
== Project Source Repository ==
([[#top|↑ top]])
Assuming you have [http://subversion.tigris.org/ SVN], the project's source can be obtained via SVN using the following command:
svn checkout svn://cdot.senecac.on.ca/deltadbg
The source can also be obtained at the following links:
* [http://matrix.senecac.on.ca/~rchu2/ddf/ddf.zip Delta Debugging Framework.zip]
* [http://matrix.senecac.on.ca/~rchu2/ddf/ddf.tar.bz2 Delta Debugging Framework.tar.bz2]
The test cases can be obtained via SVN using the following command
svn checkout svn://cdot.senecac.on.ca/deltatest
The source can also be obtained at the following links:
* [http://matrix.senecac.on.ca/~rchu2/ddf/deltatest.zip DeltaTest.zip]
* [http://matrix.senecac.on.ca/~rchu2/ddf/deltatest.tar.bz2 DeltaTest.tar.bz2]
== Project Details ==
Now that we are aware of the different concepts that we must take into account with regards to delta debugging, the next section will outline some facts and assumptions that are being made, and attempt to define the vision and process of the delta debugging framework.
== Project Facts and Assumptions Principles ==
([[#top|↑ top]])
# The source tree for the Mozilla project is HUGE. With many different source file types (C++, JS, XUL, etc.) in many different directories.
# The developer has a test case that can be used indicate whether the test passes/fails/is indeterminate.
# The developer will NOT know the date/version of the last known good version.
== Project Flowchart ==
# Once this is done, the framework will deliver the cause of the failure in a report for the user and the operation is finished.
== Project Source Repository Test Cases == The test cases used in this project are located in the [[Delta_debugging_testcases|Delta Debugging Testcases page]]. == Project Roadmap == ([[#top|↑ top]]) [[Delta Debugging Framework Roadmap|Delta Debugging Framework Roadmap]] The page outlines our current vision of the delta debugging framework and a roadmap of the work that needs to be completed to accomplish our vision. This roadmap may be subject to change and/or expanded as our vision expands through feedback and requests from others, our own ideas, etc. == Partial Class Diagram ==
([[#top|↑ top]])
== Project Task List News ==
([[#top|↑ top]])
A roadmap of our vision of the direction of the project will be heading in the future will be created and posted soon.
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'''
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.
=== Nov. 26, 2006 ===
'''Milestone:'''
* Even though the test framework is incomplete, I think we can go ahead and begin the initial testing of the delta debugger on a real regressive program as I think we are ready. Coincidentally, exactly 2 months after the first project news posting on Sept. 26, 2006.
I haven't posted an update in a while so here goes. What's been done since then?
*<strike>Found a suitable crash case thanks to the people of [irc://irc.mozilla.org#qa #qa] (in particular, asqueella and Aleksej). For full details on the bug, see [https://bugzilla.mozilla.org/show_bug.cgi?id== Project News ==354300 Bug #354300].</strike>
*Talked to Reed Loden on IRC. He will be setting up a CVS repository for us something this coming week ([[#top|↑ top]]Tuesday at earliest).
=== Nov. 17, 2006 ===
* Delta Debugging Algorithm partially complete. Unthoroughly tested though can theoretically find the directories that contain the failure inducing changes.
* Test cases and samples we may be able to use to test the algorithm.
Uploaded files into '''scen1''' directory, containing test module for '''binaryTest'''. The test is ready to be used in the algorithm. The directory contains:
* '''binaryTest.pl''' - test to detect the existence of a file.
* '''helloWorld.pl''' - enough said!
* '''binaryTestCaller.pl''' - runs '''helloWorld.pl''', pipe the result to '''hello.log''', and have '''binaryTest.pl''' attempt to detect it.
This is the working version of the code, labeled '''revision 12'''. Now I have to find a way to wreck it........
I read through your documentation here, and it is looking good. I also spoke to Shaver by phone this morning, and we chatted briefly about this project. He suggests that you start your work by looking for a suitable '''Crash Case''', one that happens reliably. Then you need at what would be necessary in order to bisect the change set (e.g., [http://www.mozilla.org/bonsai.html bonsai] data) in order to get closer to the change that introduced the bug. Shaver suggested that robc (Rob Campbell) might be a good person to help you brainstorm on this.
== How to Get Involved ==
We need a test program that we could use and upload to our test SVN repository to test the delta debugging framework. Ideally, the test program will meet the following requirements:
# Has source files that span multiple directories wide and deep yet be small enough that the delta debugging can be done in a short amount of time so that all aspects of the delta debugger can be tested.
# Has a regression. Or can easily be modified so that some functionality will stop working.
# Has an automated test case that tests the regressive functionality.
If you don't have a program that meets the first requirement, we could also use test programs that have multiple source files. The key being that the program has more than one source file. Programs that are contained in only one source files are useless to us.
If you have a program that meets these requirements, and you want to contribute to this project, then holla.
<hr />
If you are looking for an easy way in which to contribute to this project, you can jump in by writing one or more tests for the test suite. This does not require that you learn about the delta debugging inner-workings or structure.
Basic Advice:
* You '''must''' be able to automate the test--no human intervention is allowed.
* Possible test types include:
*: '''Crashing'''
*:: Can you crash the program with a minimal collection of circumstances (steps) that are easily reproducable? (In other words, can you write a script so that this happens in a controlled manner.)
*: '''Performance-related'''
*:: Is there a threshold for unacceptable consumption of time and/or space that is reason for concern?
*: '''Program hanging'''
*:: Does the program hang? Will it occur in a certain functionality of the software that is possible to isolate (reproduce) through scripted means?
*: '''Unexpected return codes'''
*:: What is a normal return code for the program? What is considered unexpected? Script a series of actions and pass the return code up to the test framework.
* Each test will fit into the test framework (which, at this point, still has to be designed). The tests must follow a few rules (again, undecided at this point).
Please check back in a few days. Expect some templates and samples up shortly to help get you going. <u>The currently listed test types are subject to change.</u>
==Future of the Project==
Here are some of the ideas related to the continuation of this project. Included are some personal ideas of the team members, tasks to reach the overall objective (a working, robust, Delta Debugging Framework for Mozilla), and additional features/functionality that would enhance the framework. This is subject to change, and a project roadmap will be written in the near future.
===CVS Support via Bonsai===
For the exploration into Bonsai and to see where it is/was heading, please view the [[delta debugging framework bonsai direction|Bonsai Direction]]. It is likely that a workable solution could be produced utilizing some of the details found in the link. This functionality would be particularly useful to Mozilla as this [Bonsai] is the technology they currently use.
===Enhancement of the Algorithm===
Richard's great algorithm can be further enhanced using a binary search-like approach that splits the revision from the current, all the way back to when the regression was first noticed (or, alternatively, when the crash case last known to have worked). Currently it works in a sequential manner, testing all previous revisions in order.
:'''More Granularity'''
:For this course, Richard's algorithm supported down to the file-level of change. In the future, it could go as far as evaluating changes in lines of code.
===Fleshed Out Test Suite Design===
The test suite test types should be further fleshed out and individual tests gathered (no participation from the class was possible due to time constraints; the test suite design wasn't fully explored and documented). Test suites could be put together for each major Mozilla.org project (Firefox, Thunderbird, Sunbird, Bugzilla, etc.).
===More Crash Cases===
More crash cases need to be found for the success in testing the project.
===Unit Tests===
A debugging framework, more so than other projects, should have its code quality tested and scrutinized heavily.
===Code Review===
Perhaps some manual audits could be performed by hand from outside contributors in the future.