On Tue, Nov 18, 2008 at 10:50 AM, ant elder <[EMAIL PROTECTED]> wrote:

>
>
> On Tue, Nov 18, 2008 at 10:44 AM, Simon Laws <[EMAIL PROTECTED]>wrote:
>
>>
>>
>> On Tue, Nov 18, 2008 at 10:36 AM, ant elder <[EMAIL PROTECTED]> wrote:
>>
>>>
>>>
>>> On Tue, Nov 18, 2008 at 10:00 AM, Simon Laws <[EMAIL PROTECTED]>wrote:
>>>
>>>>
>>>>
>>>> On Tue, Nov 18, 2008 at 9:39 AM, Simon Laws <[EMAIL PROTECTED]>wrote:
>>>>
>>>>> From the parent thread we debated whether to start from the equinox
>>>>> branch or from trunk. I'm keen to understand the code changes that 
>>>>> implement
>>>>> the features Ramyond has documented [1]. So as a compromise to just
>>>>> replacing trunk wholesale with the equinox branch I'm going to look at it 
>>>>> a
>>>>> module at a time. The objective being to note changes and ask questions
>>>>> about the code changes here. I'll make a start and see how it goes. If 
>>>>> it's
>>>>> too difficult the fallback is to just go with the branch. Regardless we 
>>>>> can
>>>>> use this exercise to review the changes so that we understand the current
>>>>> features in the branch and, more importantly, how they are realized.
>>>>>
>>>>> Regards
>>>>>
>>>>> Simon
>>>>>
>>>>> [1]
>>>>> http://cwiki.apache.org/confluence/display/TUSCANYWIKI/OSGi+Enablement+for+Tuscany+Runtime
>>>>
>>>>
>>>> So first things first. The modules that are missing from the equinox
>>>> branch but are still in trunk. Clearly these are not required for the
>>>> branches view of OSGi support but they seem to fall into two categories.
>>>> Those that are junk which I'll just remove. Those that I'm not sure about 
>>>> so
>>>> I'll move into a /contrib/modules/folder and we can delete them at our
>>>> leisure so that we understand if they are obsolete or not.
>>>>
>>>> modules/
>>>>   api
>>>>   assembly-java-dsj
>>>>   binding-feed
>>>>   binding-gdata2
>>>>   binding-gdata2-runtime
>>>>   core-spring
>>>>   data-engine-helper
>>>>   host-embedded
>>>>   implementation-bpel-jbpm
>>>>   monitor-logging
>>>>
>>>> modules/
>>>>   binding-dwr
>>>>   binding-jms-policy
>>>>   binding-notification
>>>>   binding-sca-jms
>>>>   contribution-groovy
>>>>   contribution-impl
>>>>   contribution-jee
>>>>   contribution-updater
>>>>   contribution-updater-impl
>>>>   databinding-job
>>>>   databinding-xstream
>>>>   extensibility-osgi
>>>>   extensibility-helper
>>>>   host-ejb
>>>>   host-openejb
>>>>   host-osgi
>>>>   implementation-das
>>>>   implementation-data-api
>>>>   implementation-data-xml
>>>>   implementation-jee
>>>>   implementation-notification
>>>>   implementatoin-openjpa
>>>>   implementation-web
>>>>   implementation-web-runtime
>>>>   node-dynamic
>>>>   node-launcher-osgi
>>>>   osgi-runtime
>>>>   policy-reliability
>>>>   runtime
>>>>   runtime2
>>>>   runtime-standalone
>>>>   runtime-tomcat
>>>>   runtime-war
>>>>   scdl4j
>>>>   thirdparty-library
>>>>
>>>> Simon
>>>>
>>>
>>> I thought the plan was to bring up just the minimal modules to run the
>>> calculator - the modules identified in [1] plus the test/sample from [2] so
>>> we know when it works. So move everything accept those to contrib/.
>>>
>>>    ...ant
>>>
>>> [1] http://apache.markmail.org/message/ofdwyjikioosc7bn
>>> [2] http://apache.markmail.org/message/qocoqvieeux2ct66
>>>
>>>
>> Yes, agreed. But in the branch some modules have been explicitly removed
>> so I wanted to call those out explicitly before going on the next stage of
>> moving all but the 36 or so modules that we need to get the calculator
>> working.
>>
>> I should have said also that the first block of modules I think are Junk
>> (although I shouldn't have put host-embedded in there just yet) the second
>> block are the ones I'm not sure about so should go in /contrib.
>>
>> Do you have the list of other modules that need moving and how do you want
>> to do it?
>>
>> Simon
>>
>
> Just move all except the minimal modules to contrib/ as thats already been
> proposed on the ML and no one has objected. Anything else, labelling certain
> ones "junk" or sorting through why things have/haven't been included in the
> sca-equinox branch i think would be better with some discussion.
>
>    ...ant
>
ok, fair point, let me undo the delete and let this thread record the
difference for the time being.  We can do the move en-mass a little later.
Let's move on and look at the diffs.

Simon

Reply via email to