[Building Sakai] 2.6 QA Push Needed

[Zhen Qian]

Maybe David Horwitz would have the ultimate answers about the FreeMarker
and the check against auto.ddl. 

However, meanwhile, I changed the code a little bit so that
newNonOfficialAccount value is a simple String value of "true" or "false".
According to the definition of
EmailTemplateService.getRenderedTemplateForUser, the replacementValues
HashTable can only accept String as the hashed Object type.

The check against auto.ddl is also removed from init function.

Checked in as r58526.

Adrian, could you please try again and see if it fixes your problem, or
maybe breaks the nightly2 build :)

Thanks,

- Zhen

On Wed, 11 Mar 2009 14:38:04 -0400, Noah Botimer <botimer at umich.edu> wrote:
> It was not happening on my machine, but I looked at the one in  
> question here. I was puzzled as to why the evaluation of the template  
> is failing. It was tested with a clean Tomcat, local M2 repo, trunk  
> working copy, and database. Here's what I do know:
> 
> Line 113 in the ETSUserNotificationProviderImpl that evaluates the  
> notifyAddedParticipant.xml template (or something else that has the  
> same structure/conditional). I'm a bit confused about the FreeMarker- 
> in-XML method, but it is definitely with the conditional on line 6.  
> There is a string equality check between the variable  
> newNonOfficialAccount and the literal "true". For some reason, this  
> if-statement is being treated as an error, in that it does not  
> evaluate to a boolean, while boolean is expected. This doesn't make  
> much sense to me, but perhaps someone who knows more about the  
> FreeMarker technique used can help.
> 
> This is all wrapped up in a check at initialization time on the  
> notification templates. There is a check against auto.ddl, too. It's  
> unclear to me what this is all about or which parts should run every  
> time vs. when auto.ddl is on. Or when and why an error parsing an  
> email template should abort startup. Can someone clarify?
> 
> A fix for the symptom may be as simple as adding a true or false  
> literal to the replacementValues map so the parsing succeeds, but I  
> think this bit of code could use some commentary as to what the  
> business rules/assumptions/behaviors are. This could be a wicked  
> surprise for someone who botches a template tweak in production.
> 
> Thanks,
> -Noah
> 
> On Mar 11, 2009, at 11:34 AM, Zhen Qian wrote:
> 
>>
>> For my local build, I tried clean tomcat, empty ~/.m2 folder, and new
>> database.
>>
>> But those steps doesn't help in the particular case I mentioned  
>> before.
>>
>> Thanks,
>>
>> - Zhen
>>
>>
>> On Wed, 11 Mar 2009 15:22:51 +0000, Adrian Fish  
>> <a.fish at lancaster.ac.uk>
>> wrote:
>>> Maybe a clean out my local repo is in order. Have you tried that?
>>>
>>> Cheers,
>>>
>>> Adrian.
>>>
>>> Zhen Qian wrote:
>>>> Adrian and Chris:
>>>>
>>>> This is related to SAK-12995, and I made those commits several  
>>>> days ago.
>>>>
>>>> We found one instance of similar problem here this week. And we  
>>>> don't
>>>> have
>>>> a solution yet.
>>>>
>>>> You might want to try with a new dev environment and see if that
>> helps...
>>>>
>>>> Thanks,
>>>>
>>>> - Zhen
>>>>
>>>> On Wed, 11 Mar 2009 11:04:40 -0400, "Maurer, Christopher Wayne"
>>>> <chmaurer at iupui.edu> wrote:
>>>>
>>>>> All the nightly builds are up, so it either happened after 8am  
>>>>> EDT or
>>>>> there¹s something up with your local build.
>>>>>
>>>>> Chris
>>>>>
>>>>>
>>>>> On 3/11/09 10:58 AM, "Adrian Fish" <a.fish at lancaster.ac.uk> wrote:
>>>>>
>>>>>
>>>>>> Hi All,
>>>>>>
>>>>>> Just started looking into fixing the Blog blockers and it seems  
>>>>>> trunk
>>>>>> is
>>>>>> broken. Something to do with the EmailTemplateService and  
>>>>>> freemarker.
>>>>>> Is
>>>>>> anybody else having problems?
>>>>>>
>>>>>> Cheers,
>>>>>>
>>>>>> Adrian.
>>>>>>
>>>>>> Michael Korcuska wrote:
>>>>>>
>>>>>>>> Greetings,
>>>>>>>>
>>>>>>>> We are nearing the hoped-for release date for 2.6. and we  
>>>>>>>> need more
>>>>>>>> help. As it is, we aren't going to make the target date of  
>>>>>>>> March 20
>>>>>>>> but we'd like to get as close as possible. Even if your  
>>>>>>>> organization
>>>>>>>> is planning to stick with 2.5 for now it would be very much
>>>>>>>> appreciated if you could pitch in.
>>>>>>>>
>>>>>>>> We need to get the QA servers updated as quickly as possible  
>>>>>>>> to the
>>>>>>>> 2.6.0-beta03 tag, we need the remaining blocker and critical  
>>>>>>>> bugs
>>>>>>>> closed (ideally fixed!) and we need some hands on testing.
>>>>>>>>
>>>>>>>> Here are some relevant links:
>>>>>>>>
>>>>>>>> List of 2.6 Changes:
>>>>>>>>
>>>>>>>>
>>>> http://confluence.sakaiproject.org/confluence/display/QA/2.6.0 
>>>> +Changes
>>>>
>>>>>>>> 2.6 Testing Signup Page:
>>>>>>>>
>>>>>>>>
>>>> http://confluence.sakaiproject.org/confluence/display/QA/2.6.0 
>>>> +Release
>>>>
>>>>>>>> 2.6 Scheduled, Unresolved Issues (sorted by priority):
>>>>>>>>
>>>>>>>>
>>>>
>> http://jira.sakaiproject.org/jira/secure/IssueNavigator.jspa? 
>> mode=hide&reques
>>>>
>>>>>>> tId=11418
>>>>>>>
>>>>>>>> There are currently 25 blocker/critical issues. A manageable  
>>>>>>>> number,
>>>>>>>> but we do need attention on these ASAP.
>>>>>>>>
>>>>>>>> Please contact Pete Peterson (plpeterson at ucdavis.edu) if you  
>>>>>>>> would
>>>>>>>> like to help and are unsure where to begin.
>>>>>>>>
>>>>>>>> Thank you!
>>>>>>>>
>>>>>>>> Michael
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>> --
>>>>>> ==================================
>>>>>> Adrian Fish
>>>>>> Software Engineer
>>>>>> Centre for e-Science
>>>>>> Bowland Tower South C Floor
>>>>>> Lancaster University
>>>>>> Lancaster
>>>>>> LA1 4YW
>>>>>>
>>>>>> email: a.fish at lancaster.ac.uk
>>>>>> web:   http://agora.lancs.ac.uk
>>>>>>
>>>>>>
>>>>>>
>>>>
>>>>
>>>
>>>
>> _______________________________________________
>> 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"
>