[Building Sakai] Sakai-2.6.0-beta02 - issue or explaination of function processConfirmContinue() (in SiteAddParticipantHandler.java)

[Zhen Qian]

Kimberly:

I found this is actually a bug generated during the module rewrite of  
"Add Participant" from main SiteAction.java file.

In fact, the list of eIdRoles is not needed in the function anymore.  
We can directly use the member attribute list userRoleEntries to  
compare with the return list of user eids which have been  
successfully added into the site.

If there is any difference between those two lists, alert message  
will be shown with those failing eids inside.

I've documented the issue inside "SAK-15821: bug in the  
processConfirmContinue function of Add Participant module" and fixed  
in r58417.

Thanks,

- Zhen

On Mar 9, 2009, at 9:50 AM, Kimberly Ma wrote:

> Hi,
> This is Kim From Rutgers University, Office from Instructional and  
> Research Technology.(kimhuang at rci.rutgers.edu)
>
> Currently I am working on integrating Rutgers changes to Sakai  
> 2.6.0.  The source code which I used is 2.6.0 -beta02.  For the  
> feature of " (SAK-15604) Allow site owner to add user name for  
> guest account upon adding user to the site as "adding participant"  
> ", please help me clarify the following:
>
> in the functon processConfirmcontinue(), - file of  
> SiteAddParticipantHandler.java,
>
> When the participant is official, the code will add his eId to the  
> hash table.
>
> if (isOfficialAccount(eId)) {
>                 // if this is a officialAccount
>                 // update the hash table
>                 eIdRoles.put(eId, role);
>
> but if the participant is not official, the code will check whether  
> he is already in the UserDirectoryService, if it is not, add him to  
> the UserDirectoryService
>
> try {
>                     UserDirectoryService.getUserByEid(eId);
>                 } catch (UserNotDefinedException e) {
>                     // if there is no such user yet, add the user
>                     try {
>                         UserEdit uEdit = UserDirectoryService
>                                 .addUser(null, eId);
> ...
> ...
>         UserDirectoryService.commitEdit(uEdit);
> ....
>
> Here is the problem :
> In 2.6 code of sakai,
> When the participant is not official, he won't never be added to  
> the hash table regardless he is  the new user to  
> UserDirectoryService or not.
>
> This part was handled differently in the 2.5, which is  the  
> nonofficialparticipant will also be added to the harsh table  
> eventually.
> by:
>
>
>                 if (state.getAttribute(STATE_MESSAGE) == null) {
>                     eIdRoles.put(eId, role);  // in sakai 2.5
>
> Is it a bug? or has it been taken care of by the other approach ?
>
> Thanks,
>
> -Kim
>
>
>
>
>
>
> On Sun, Mar 1, 2009 at 12:29 PM, Pete Peterson  
> <plpeterson at ucdavis.edu> wrote:
> Greetings QA'ers, Developers and all ,
>
>
> The 2.6.0-beta03 tag has been cut and is ready to be installed on  
> the QA instances. This release is critical as we will shortly (the  
> next 1-2 weeks) be looking at freezing the code in preparation for  
> cutting the release candidates.
>
>
> There are many things left to be done and all of them are critical  
> and important to the success of this release:
>
>
> QA TESTING – URGENT, please update the QA pages and jiras with your  
> results!
>
> ·         Make sure the test server (QA servers) you are using is  
> running the most current release.
>
> ·         Here is the list of QA assignment opportunities and  
> testing results page (http://confluence.sakaiproject.org/confluence/ 
> display/QA/2.6.0+Changes);
>
> ·         As you test and gather result please report anomalies and  
> bugs in Jira and also on the confluence page.
>
> ·         There is still a large number of issues (90) with the  
> status (Sakai 2.6.x Fixes to Merge – Unverified). These issues  
> cannot be merged into the upcoming 2.6 release, unless they are  
> QA’ed and verified fixed.
>
> ·          Once the fixes are verified the issue should be set to  
> “Closed”.  This will move the issue to the filter the Branch  
> Managers are monitoring (Sakai 2.6.x Fixes to Merge – Verified) and  
> they will merging the fix into the branch.
>
> ·         Sakai 2.6.x Maintenance Branch Fixes (since 2.6.0- 
> beta01):http://bugs.sakaiproject.org/jira/secure/ 
> IssueNavigator.jspa?mode=hide&requestId=11833.
>
> ·         Report any new blockers via jira ASAP.
>
>
> FIXES (developers)
>
> ·         There is a short window of opportunity to include any  
> fixes in the 2.6.0 release. We will need to curtail merges soon and  
> start to focus on a stable line in the sand for the 2.6.0 release.
>
> ·         Please let me know if there are bugs, especially security  
> bugs, that you are working on currently and hope to have fixed and  
> tested soon, and we will try to accommodate your schedule.
>
>
> ALL
>
> ·         If you are running 2.6.0(?) and have specific issues,  
> bugs or comments please let me know ASAP, so we can try to address  
> them prior to the release.
>
>
> SCHEDULE
>
> ·         Updated release date is March 20th (tentative). This is  
> an aggressive release timeline, but we want to allow institution  
> time to install, test and evaluate this release prior to fall 2009.
>
>
> Thanks in advance for all the help, effort and support,
>
>
> Pete aka P^2
>
>
>
>
> This automatic notification message was sent by Sakai Collab  
> (https://collab.sakaiproject.org//portal) from the DG:  
> Development / Building Sakai site.
> You can modify how you receive notifications at My Workspace >  
> Preferences.
>
> _______________________________________________
> sakai-dev mailing list
> sakai-dev at collab.sakaiproject.org
> http://collab.sakaiproject.org/mailman/listinfo/sakai-dev
>
> TO UNSUBSCRIBE: send email to sakai-dev- 
> unsubscribe at collab.sakaiproject.org with a subject of "unsubscribe"

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://collab.sakaiproject.org/pipermail/sakai-dev/attachments/20090309/297577e8/attachment-0001.html