You are viewing a plain text version of this content. The canonical link for it is here.
Posted to oak-dev@jackrabbit.apache.org by Angela Schreiber <an...@adobe.com> on 2014/08/12 16:54:07 UTC
Cleanup NodeStore and MK implementations (was: Re: AW: Confusions
about API's and Flavours in the documentation ...)
hi all
i recently had another look at the oak-core module and was thinking
if it wouldn't be better if we would move the NodeStore implementations
into separate modules.
to begin with i just tried 2 separate modules:
- oak-ns-document:
> everything below oak.plugins.document
- oak-ns-segment:
> everything below oak.plugins.segment
> segment specific parts of oak.plugins.backup
i found the following issues:
- org.apache.jackrabbit.oak.plugins.cache is not part of the exported
packages
- oak.plugins.backup contains both public API and implementations without
separation
- the following test-classes have a hard dependency on one or more ns
implementations:
> KernelNodeStoreCacheTest
> ClusterPermissionsTest
> NodeStoreFixture
to fix those we could need to be able to run the tests with the
individual nodestore
modules and move those tests that are just intended to work with a
particular impl.
such a move would not only prevent us from introducing unintended
package dependencies but would also reduce the number of dependencies
present with oak-core.
wdyt?
kind regards
angela
On 12/08/14 16:20, "Michael Dürig" <md...@apache.org> wrote:
>
>
>On 12.8.14 4:08 , Angela Schreiber wrote:
>> hi claus
>>
>>>> And yes, it's confusing.
>>> .. so I'm not alone here :-)
>>
>> no... nobody gets it... in particular as one term is also the
>> marketing term for the other... just makes the mess a complete mess.
>>
>> as far as i am concerned i don't know any good reason for keeping two
>> APIs for the same thing. IMO we should move the microkernel API to
>> to the sandbox... the oak code base evolved and it doesn't make sense
>> to keep things just for nostalgia if they add so much confusion.
>
>+1 see https://issues.apache.org/jira/browse/OAK-1327
>
>Michael
>
>>
>> kind regards
>> angela
>>
AW: Cleanup NodeStore and MK implementations (was: Re: AW:
Confusions about API's and Flavours in the documentation ...)
Posted by KÖLL Claus <C....@TIROL.GV.AT>.
hi angela,
+1 for anything that clears that mess !!
separating the two storage concepts is ok for me and also to move the MK code in a sandbox project ...
on the other hand we should then update the documentation so that the people are not so confused about the wording/architecture ..
greets
claus
Re: AW: Cleanup NodeStore and MK implementations (was: Re: AW:
Confusions about API's and Flavours in the documentation ...)
Posted by Chetan Mehrotra <ch...@gmail.com>.
On Mon, Aug 18, 2014 at 1:02 PM, Michael Dürig <md...@apache.org> wrote:
> This will affect the ongoing work in theses areas, which is all merged back
> into the 1.0 branch. We should also consider Marcel and Alex, who will
> probably be most affected by the additional merge effort.
I also think that this refactoring should be delayed (if that is to be
done) for some time till we get more stability in the two NodeStore
implementations.
Chetan Mehrotra
Re: AW: Cleanup NodeStore and MK implementations (was: Re: AW: Confusions
about API's and Flavours in the documentation ...)
Posted by Michael Dürig <md...@apache.org>.
On 18.8.14 8:41 , KÖLL Claus wrote:
> The longer we wait now the harder it will be to create clean and easier to understand documentation
I agree we should do the move, however doubt that right now is a good
point in time. This will affect the ongoing work in theses areas, which
is all merged back into the 1.0 branch. We should also consider Marcel
and Alex, who will probably be most affected by the additional merge effort.
Michael
AW: Cleanup NodeStore and MK implementations (was: Re: AW:
Confusions about API's and Flavours in the documentation ...)
Posted by KÖLL Claus <C....@TIROL.GV.AT>.
Hi @ all
I see it like angela . I think if we want to clear the code and the documentation it's time to do it.
I have also found some System-Properties in the DocumentNodeStore starting with "oak.documentMK...."
The longer we wait now the harder it will be to create clean and easier to understand documentation
greets
claus
Re: Cleanup NodeStore and MK implementations (was: Re: AW: Confusions
about API's and Flavours in the documentation ...)
Posted by Angela Schreiber <an...@adobe.com>.
hi michael
maybe that's just the price we have to pay for not designing
the scope of oak-core right from the beginning.
if we don't tear it apart, i fear that problems like the one i
found with oak.plugins.backup will spread throughout oak-core.
kind regards
angela
On 13/08/14 09:00, "Michael Dürig" <md...@apache.org> wrote:
>
>Hi,
>
>I'm generally in favour of such a move. However, as currently there is
>still a lot of fixes from the affected packaged being merged to the 1.0
>branch we should be very careful not to unnecessarily complicate this
>process.
>
>Michael
>
>
>
>On 12.8.14 4:54 , Angela Schreiber wrote:
>> hi all
>>
>> i recently had another look at the oak-core module and was thinking
>> if it wouldn't be better if we would move the NodeStore implementations
>> into separate modules.
>>
>> to begin with i just tried 2 separate modules:
>>
>> - oak-ns-document:
>> > everything below oak.plugins.document
>>
>> - oak-ns-segment:
>> > everything below oak.plugins.segment
>> > segment specific parts of oak.plugins.backup
>>
>> i found the following issues:
>>
>> - org.apache.jackrabbit.oak.plugins.cache is not part of the exported
>> packages
>> - oak.plugins.backup contains both public API and implementations
>>without
>> separation
>> - the following test-classes have a hard dependency on one or more ns
>> implementations:
>> > KernelNodeStoreCacheTest
>> > ClusterPermissionsTest
>> > NodeStoreFixture
>> to fix those we could need to be able to run the tests with the
>> individual nodestore
>> modules and move those tests that are just intended to work with a
>> particular impl.
>>
>> such a move would not only prevent us from introducing unintended
>> package dependencies but would also reduce the number of dependencies
>> present with oak-core.
>>
>> wdyt?
>>
>> kind regards
>> angela
>>
>>
>> On 12/08/14 16:20, "Michael Dürig" <md...@apache.org> wrote:
>>
>>>
>>>
>>> On 12.8.14 4:08 , Angela Schreiber wrote:
>>>> hi claus
>>>>
>>>>>> And yes, it's confusing.
>>>>> .. so I'm not alone here :-)
>>>>
>>>> no... nobody gets it... in particular as one term is also the
>>>> marketing term for the other... just makes the mess a complete mess.
>>>>
>>>> as far as i am concerned i don't know any good reason for keeping two
>>>> APIs for the same thing. IMO we should move the microkernel API to
>>>> to the sandbox... the oak code base evolved and it doesn't make sense
>>>> to keep things just for nostalgia if they add so much confusion.
>>>
>>> +1 see https://issues.apache.org/jira/browse/OAK-1327
>>>
>>> Michael
>>>
>>>>
>>>> kind regards
>>>> angela
>>>>
>>
Re: Cleanup NodeStore and MK implementations (was: Re: AW: Confusions
about API's and Flavours in the documentation ...)
Posted by Michael Dürig <md...@apache.org>.
Hi,
I'm generally in favour of such a move. However, as currently there is
still a lot of fixes from the affected packaged being merged to the 1.0
branch we should be very careful not to unnecessarily complicate this
process.
Michael
On 12.8.14 4:54 , Angela Schreiber wrote:
> hi all
>
> i recently had another look at the oak-core module and was thinking
> if it wouldn't be better if we would move the NodeStore implementations
> into separate modules.
>
> to begin with i just tried 2 separate modules:
>
> - oak-ns-document:
> > everything below oak.plugins.document
>
> - oak-ns-segment:
> > everything below oak.plugins.segment
> > segment specific parts of oak.plugins.backup
>
> i found the following issues:
>
> - org.apache.jackrabbit.oak.plugins.cache is not part of the exported
> packages
> - oak.plugins.backup contains both public API and implementations without
> separation
> - the following test-classes have a hard dependency on one or more ns
> implementations:
> > KernelNodeStoreCacheTest
> > ClusterPermissionsTest
> > NodeStoreFixture
> to fix those we could need to be able to run the tests with the
> individual nodestore
> modules and move those tests that are just intended to work with a
> particular impl.
>
> such a move would not only prevent us from introducing unintended
> package dependencies but would also reduce the number of dependencies
> present with oak-core.
>
> wdyt?
>
> kind regards
> angela
>
>
> On 12/08/14 16:20, "Michael Dürig" <md...@apache.org> wrote:
>
>>
>>
>> On 12.8.14 4:08 , Angela Schreiber wrote:
>>> hi claus
>>>
>>>>> And yes, it's confusing.
>>>> .. so I'm not alone here :-)
>>>
>>> no... nobody gets it... in particular as one term is also the
>>> marketing term for the other... just makes the mess a complete mess.
>>>
>>> as far as i am concerned i don't know any good reason for keeping two
>>> APIs for the same thing. IMO we should move the microkernel API to
>>> to the sandbox... the oak code base evolved and it doesn't make sense
>>> to keep things just for nostalgia if they add so much confusion.
>>
>> +1 see https://issues.apache.org/jira/browse/OAK-1327
>>
>> Michael
>>
>>>
>>> kind regards
>>> angela
>>>
>