You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@openjpa.apache.org by Kevin Sutter <kw...@gmail.com> on 2011/10/10 15:18:16 UTC
[DISCUSS] JDO usage end-of-life?
Hi,
Sorry to cross post to both forums, but I wanted to ensure that I hit
everybody that might have an opinion on this JDO topic...
Is the JDO personality of OpenJPA still being utilized? Marc's recent post
about possibly pulling in javax.jdo.* packages during the enhancement
processing [1] reminded me that we still have old remnants of JDO (and Kodo)
in the OpenJPA code base. OpenJPA has never claimed support for JDO (nor
Kodo). Way back when, BEA provided a JDO implementation as part of their
offering that sat on top of OpenJPA. As far as I know, BEA (and Oracle)
only support the 1.1.x service stream of OpenJPA. So, if we did this in the
2.x stream, there should be no effects to that set of users.
Would there be a concern with the current users of OpenJPA to clean up the
code base and remove these JDO/Kodo references? From a JPA/OpenJPA
perspective, you should see no differences in functionality.
Like I said, Marc's posting prompted me to revisit this topic. I'm just
exploring the option with no immediate plans of actually doing the work...
Thanks,
Kevin
[1]
http://openjpa.208410.n2.nabble.com/weird-JDO-Exception-when-using-OpenJPA-2-Enhancer-tc6870122.html
Re: [DISCUSS] JDO usage end-of-life?
Posted by Kevin Sutter <kw...@gmail.com>.
Hi Pinaki,
I haven't spent a great deal of time looking into this. But, you are
probably right -- the JDO footprint is probably very small. I was just
fishing to see how much interest there was in JDO -- just in case we would
find some area of the code that is JDO-centric and it's either confusing,
incorrect, or affecting performance.
Thanks,
Kevin
On Thu, Oct 13, 2011 at 8:20 AM, Pinaki Poddar <pp...@apache.org> wrote:
> Hi Kevin,
> > OpenJPA project is for providing a JPA implementation, not a JDO
> implementation.
> OpenJPA is a misnomer :)
>
> Seriously speaking, what exactly do you suggest to remove to eol/get rid
> of JDO? My read is that JDO-awareness has a very small footprint in terms
> of
> codelines.
>
> -----
> Pinaki Poddar
> Chair, Apache OpenJPA Project
> --
> View this message in context:
> http://openjpa.208410.n2.nabble.com/DISCUSS-JDO-usage-end-of-life-tp6876837p6888872.html
> Sent from the OpenJPA Developers mailing list archive at Nabble.com.
>
Re: [DISCUSS] JDO usage end-of-life?
Posted by Pinaki Poddar <pp...@apache.org>.
Hi Kevin,
> OpenJPA project is for providing a JPA implementation, not a JDO
implementation.
OpenJPA is a misnomer :)
Seriously speaking, what exactly do you suggest to remove to eol/get rid
of JDO? My read is that JDO-awareness has a very small footprint in terms of
codelines.
-----
Pinaki Poddar
Chair, Apache OpenJPA Project
--
View this message in context: http://openjpa.208410.n2.nabble.com/DISCUSS-JDO-usage-end-of-life-tp6876837p6888872.html
Sent from the OpenJPA Developers mailing list archive at Nabble.com.
Re: [DISCUSS] JDO usage end-of-life?
Posted by Kevin Sutter <kw...@gmail.com>.
Hi Pinaki,
I can see validity in all of your stated reasons. But, the OpenJPA project
is for providing a JPA implementation, not a JDO implementation. If
somebody wants to spin off and support their own "OpenJDO" implementation
that is based on OpenJPA or its kernel, then all the more power to them. I
am just questioning why we should continue to work around or with JDO
idiosyncrasies with our JPA implementation.
Thanks,
Kevin
On Wed, Oct 12, 2011 at 11:14 AM, Pinaki Poddar <pp...@apache.org> wrote:
> I do not see need to remove JDO support from OpenJPA for following reasons
>
> a) there are current JDO usage/users -- may not be as predominant as JPA
> community, but they are there.
> Google AppEngine environment, for example, supports JDO for persistent
> services. I have also seen active JDO users in this forum.
>
> b) OpenJPA architecture had always maintained the goal of being
> facade-agnostic (i.e. JPA, JDO or something else) and datastore-agnostic
> (i.e. whether JDBC or NoSQL or OODB). That goal was not easy to achieve but
> had been executed with a degree of discipline that, imo, is one of its
> unique and key strength. Letting go that primary strength would not be a
> positive step.
>
> c) I have come across migration scenarios where an existing JDO schema is
> being ported to JPA schema. Those use cases may be even more frequent as
> JPA
> gains more traction. Such migration scenario are supported in current
> model.
>
> d) The reported error that prompted this discussion is not a critical
> failure. Possibly could be fixed with moderate effort. It does not seem to
> indicate that we should throw JDO under the bus.
>
>
> -----
> Pinaki Poddar
> Chair, Apache OpenJPA Project
> --
> View this message in context:
> http://openjpa.208410.n2.nabble.com/DISCUSS-JDO-usage-end-of-life-tp6876837p6885650.html
> Sent from the OpenJPA Developers mailing list archive at Nabble.com.
>
Re: [DISCUSS] JDO usage end-of-life?
Posted by Kevin Sutter <kw...@gmail.com>.
On Wed, Oct 12, 2011 at 11:14 AM, Pinaki Poddar <pp...@apache.org> wrote:
>
> d) The reported error that prompted this discussion is not a critical
> failure. Possibly could be fixed with moderate effort. It does not seem to
> indicate that we should throw JDO under the bus.
>
That reported error just jogged my memory. That error has nothing directly
to do with my proposal to jettison JDO.
>
> -----
> Pinaki Poddar
> Chair, Apache OpenJPA Project
> --
> View this message in context:
> http://openjpa.208410.n2.nabble.com/DISCUSS-JDO-usage-end-of-life-tp6876837p6885650.html
> Sent from the OpenJPA Developers mailing list archive at Nabble.com.
>
Re: [DISCUSS] JDO usage end-of-life?
Posted by Pinaki Poddar <pp...@apache.org>.
I do not see need to remove JDO support from OpenJPA for following reasons
a) there are current JDO usage/users -- may not be as predominant as JPA
community, but they are there.
Google AppEngine environment, for example, supports JDO for persistent
services. I have also seen active JDO users in this forum.
b) OpenJPA architecture had always maintained the goal of being
facade-agnostic (i.e. JPA, JDO or something else) and datastore-agnostic
(i.e. whether JDBC or NoSQL or OODB). That goal was not easy to achieve but
had been executed with a degree of discipline that, imo, is one of its
unique and key strength. Letting go that primary strength would not be a
positive step.
c) I have come across migration scenarios where an existing JDO schema is
being ported to JPA schema. Those use cases may be even more frequent as JPA
gains more traction. Such migration scenario are supported in current model.
d) The reported error that prompted this discussion is not a critical
failure. Possibly could be fixed with moderate effort. It does not seem to
indicate that we should throw JDO under the bus.
-----
Pinaki Poddar
Chair, Apache OpenJPA Project
--
View this message in context: http://openjpa.208410.n2.nabble.com/DISCUSS-JDO-usage-end-of-life-tp6876837p6885650.html
Sent from the OpenJPA Developers mailing list archive at Nabble.com.
Re: [DISCUSS] JDO usage end-of-life?
Posted by Craig L Russell <cr...@oracle.com>.
Hi strub,
On Oct 11, 2011, at 10:26 AM, Mark Struberg wrote:
> There is a JDO-3.0 spec out there since almost a year now. I'm not
> sure if we can/like to catch up.
There have been non-trivial but minor changes since JDO 2.0. IMHO
"Catching up" would be a small task compared to re-activating the JDO
support in OpenJPA.
>
> Isn't there a pure JDO impl at db.apache.org which is up2date anyway?
No, the reference implementation for JDO is/has been DataNucleus. JDO
1.0 had a reference implementation that didn't support relational
databases, just a "toy" key/value store.
Best,
Craig
>
>
> LieGrue,
> strub
>
>
>
> ----- Original Message -----
>> From: Michael Dick <mi...@gmail.com>
>> To: users@openjpa.apache.org; dev@openjpa.apache.org
>> Cc:
>> Sent: Tuesday, October 11, 2011 4:21 AM
>> Subject: Re: [DISCUSS] JDO usage end-of-life?
>>
>> T here is at least some interest from a subset of our users.
>> Matthew Adams
>> and issue: OPENJPA-1744
>> <https://issues.apache.org/jira/browse/OPENJPA-1744>to add support
>> for
>> JDO last July. I closed the issue, but Matthew responded
>> and the issue was reopened.
>>
>> There hasn't been a lot of activity on the JIRA since then. There
>> are some
>> users watching it, but no one has voted for it. If there's an
>> outpouring of
>> support from the users list, and a committer (or aspiriing
>> committer) is
>> interested in championing the effort, I'd be all for adding a JDO
>> persona.
>> Absent a champion who is ready to dive into the code, I think that
>> we should
>> clean up the references to jdo.
>>
>> Even if OpenJPA removes the references to JDO, I'm sure a separate
>> module
>> could be written that sits on top of our binaries. I suspect that's
>> what BEA
>> / Oracle did.
>>
>> -mike
>>
>>
>>
>> On Mon, Oct 10, 2011 at 8:18 AM, Kevin Sutter <kw...@gmail.com>
>> wrote:
>>
>>> Hi,
>>> Sorry to cross post to both forums, but I wanted to ensure that I
>>> hit
>>> everybody that might have an opinion on this JDO topic...
>>>
>>> Is the JDO personality of OpenJPA still being utilized? Marc's
>>> recent
>> post
>>> about possibly pulling in javax.jdo.* packages during the
>>> enhancement
>>> processing [1] reminded me that we still have old remnants of JDO
>>> (and
>>> Kodo)
>>> in the OpenJPA code base. OpenJPA has never claimed support for
>>> JDO (nor
>>> Kodo). Way back when, BEA provided a JDO implementation as part
>>> of their
>>> offering that sat on top of OpenJPA. As far as I know, BEA (and
>>> Oracle)
>>> only support the 1.1.x service stream of OpenJPA. So, if we did
>>> this in
>>> the
>>> 2.x stream, there should be no effects to that set of users.
>>>
>>> Would there be a concern with the current users of OpenJPA to
>>> clean up the
>>> code base and remove these JDO/Kodo references? From a JPA/OpenJPA
>>> perspective, you should see no differences in functionality.
>>>
>>> Like I said, Marc's posting prompted me to revisit this topic. I'm
>> just
>>> exploring the option with no immediate plans of actually doing the
>>> work...
>>>
>>> Thanks,
>>> Kevin
>>>
>>> [1]
>>>
>>>
>> http://openjpa.208410.n2.nabble.com/weird-JDO-Exception-when-using-OpenJPA-2-Enhancer-tc6870122.html
>>>
>>
Craig L Russell
Architect, Oracle
http://db.apache.org/jdo
408 276-5638 mailto:Craig.Russell@oracle.com
P.S. A good JDO? O, Gasp!
Re: [DISCUSS] JDO usage end-of-life?
Posted by Craig L Russell <cr...@oracle.com>.
Hi strub,
On Oct 11, 2011, at 10:26 AM, Mark Struberg wrote:
> There is a JDO-3.0 spec out there since almost a year now. I'm not
> sure if we can/like to catch up.
There have been non-trivial but minor changes since JDO 2.0. IMHO
"Catching up" would be a small task compared to re-activating the JDO
support in OpenJPA.
>
> Isn't there a pure JDO impl at db.apache.org which is up2date anyway?
No, the reference implementation for JDO is/has been DataNucleus. JDO
1.0 had a reference implementation that didn't support relational
databases, just a "toy" key/value store.
Best,
Craig
>
>
> LieGrue,
> strub
>
>
>
> ----- Original Message -----
>> From: Michael Dick <mi...@gmail.com>
>> To: users@openjpa.apache.org; dev@openjpa.apache.org
>> Cc:
>> Sent: Tuesday, October 11, 2011 4:21 AM
>> Subject: Re: [DISCUSS] JDO usage end-of-life?
>>
>> T here is at least some interest from a subset of our users.
>> Matthew Adams
>> and issue: OPENJPA-1744
>> <https://issues.apache.org/jira/browse/OPENJPA-1744>to add support
>> for
>> JDO last July. I closed the issue, but Matthew responded
>> and the issue was reopened.
>>
>> There hasn't been a lot of activity on the JIRA since then. There
>> are some
>> users watching it, but no one has voted for it. If there's an
>> outpouring of
>> support from the users list, and a committer (or aspiriing
>> committer) is
>> interested in championing the effort, I'd be all for adding a JDO
>> persona.
>> Absent a champion who is ready to dive into the code, I think that
>> we should
>> clean up the references to jdo.
>>
>> Even if OpenJPA removes the references to JDO, I'm sure a separate
>> module
>> could be written that sits on top of our binaries. I suspect that's
>> what BEA
>> / Oracle did.
>>
>> -mike
>>
>>
>>
>> On Mon, Oct 10, 2011 at 8:18 AM, Kevin Sutter <kw...@gmail.com>
>> wrote:
>>
>>> Hi,
>>> Sorry to cross post to both forums, but I wanted to ensure that I
>>> hit
>>> everybody that might have an opinion on this JDO topic...
>>>
>>> Is the JDO personality of OpenJPA still being utilized? Marc's
>>> recent
>> post
>>> about possibly pulling in javax.jdo.* packages during the
>>> enhancement
>>> processing [1] reminded me that we still have old remnants of JDO
>>> (and
>>> Kodo)
>>> in the OpenJPA code base. OpenJPA has never claimed support for
>>> JDO (nor
>>> Kodo). Way back when, BEA provided a JDO implementation as part
>>> of their
>>> offering that sat on top of OpenJPA. As far as I know, BEA (and
>>> Oracle)
>>> only support the 1.1.x service stream of OpenJPA. So, if we did
>>> this in
>>> the
>>> 2.x stream, there should be no effects to that set of users.
>>>
>>> Would there be a concern with the current users of OpenJPA to
>>> clean up the
>>> code base and remove these JDO/Kodo references? From a JPA/OpenJPA
>>> perspective, you should see no differences in functionality.
>>>
>>> Like I said, Marc's posting prompted me to revisit this topic. I'm
>> just
>>> exploring the option with no immediate plans of actually doing the
>>> work...
>>>
>>> Thanks,
>>> Kevin
>>>
>>> [1]
>>>
>>>
>> http://openjpa.208410.n2.nabble.com/weird-JDO-Exception-when-using-OpenJPA-2-Enhancer-tc6870122.html
>>>
>>
Craig L Russell
Architect, Oracle
http://db.apache.org/jdo
408 276-5638 mailto:Craig.Russell@oracle.com
P.S. A good JDO? O, Gasp!
Re: [DISCUSS] JDO usage end-of-life?
Posted by Mark Struberg <st...@yahoo.de>.
There is a JDO-3.0 spec out there since almost a year now. I'm not sure if we can/like to catch up.
Isn't there a pure JDO impl at db.apache.org which is up2date anyway?
LieGrue,
strub
----- Original Message -----
> From: Michael Dick <mi...@gmail.com>
> To: users@openjpa.apache.org; dev@openjpa.apache.org
> Cc:
> Sent: Tuesday, October 11, 2011 4:21 AM
> Subject: Re: [DISCUSS] JDO usage end-of-life?
>
>T here is at least some interest from a subset of our users. Matthew Adams
> and issue: OPENJPA-1744
> <https://issues.apache.org/jira/browse/OPENJPA-1744>to add support for
> JDO last July. I closed the issue, but Matthew responded
> and the issue was reopened.
>
> There hasn't been a lot of activity on the JIRA since then. There are some
> users watching it, but no one has voted for it. If there's an outpouring of
> support from the users list, and a committer (or aspiriing committer) is
> interested in championing the effort, I'd be all for adding a JDO persona.
> Absent a champion who is ready to dive into the code, I think that we should
> clean up the references to jdo.
>
> Even if OpenJPA removes the references to JDO, I'm sure a separate module
> could be written that sits on top of our binaries. I suspect that's what BEA
> / Oracle did.
>
> -mike
>
>
>
> On Mon, Oct 10, 2011 at 8:18 AM, Kevin Sutter <kw...@gmail.com> wrote:
>
>> Hi,
>> Sorry to cross post to both forums, but I wanted to ensure that I hit
>> everybody that might have an opinion on this JDO topic...
>>
>> Is the JDO personality of OpenJPA still being utilized? Marc's recent
> post
>> about possibly pulling in javax.jdo.* packages during the enhancement
>> processing [1] reminded me that we still have old remnants of JDO (and
>> Kodo)
>> in the OpenJPA code base. OpenJPA has never claimed support for JDO (nor
>> Kodo). Way back when, BEA provided a JDO implementation as part of their
>> offering that sat on top of OpenJPA. As far as I know, BEA (and Oracle)
>> only support the 1.1.x service stream of OpenJPA. So, if we did this in
>> the
>> 2.x stream, there should be no effects to that set of users.
>>
>> Would there be a concern with the current users of OpenJPA to clean up the
>> code base and remove these JDO/Kodo references? From a JPA/OpenJPA
>> perspective, you should see no differences in functionality.
>>
>> Like I said, Marc's posting prompted me to revisit this topic. I'm
> just
>> exploring the option with no immediate plans of actually doing the work...
>>
>> Thanks,
>> Kevin
>>
>> [1]
>>
>>
> http://openjpa.208410.n2.nabble.com/weird-JDO-Exception-when-using-OpenJPA-2-Enhancer-tc6870122.html
>>
>
Re: [DISCUSS] JDO usage end-of-life?
Posted by Mark Struberg <st...@yahoo.de>.
There is a JDO-3.0 spec out there since almost a year now. I'm not sure if we can/like to catch up.
Isn't there a pure JDO impl at db.apache.org which is up2date anyway?
LieGrue,
strub
----- Original Message -----
> From: Michael Dick <mi...@gmail.com>
> To: users@openjpa.apache.org; dev@openjpa.apache.org
> Cc:
> Sent: Tuesday, October 11, 2011 4:21 AM
> Subject: Re: [DISCUSS] JDO usage end-of-life?
>
>T here is at least some interest from a subset of our users. Matthew Adams
> and issue: OPENJPA-1744
> <https://issues.apache.org/jira/browse/OPENJPA-1744>to add support for
> JDO last July. I closed the issue, but Matthew responded
> and the issue was reopened.
>
> There hasn't been a lot of activity on the JIRA since then. There are some
> users watching it, but no one has voted for it. If there's an outpouring of
> support from the users list, and a committer (or aspiriing committer) is
> interested in championing the effort, I'd be all for adding a JDO persona.
> Absent a champion who is ready to dive into the code, I think that we should
> clean up the references to jdo.
>
> Even if OpenJPA removes the references to JDO, I'm sure a separate module
> could be written that sits on top of our binaries. I suspect that's what BEA
> / Oracle did.
>
> -mike
>
>
>
> On Mon, Oct 10, 2011 at 8:18 AM, Kevin Sutter <kw...@gmail.com> wrote:
>
>> Hi,
>> Sorry to cross post to both forums, but I wanted to ensure that I hit
>> everybody that might have an opinion on this JDO topic...
>>
>> Is the JDO personality of OpenJPA still being utilized? Marc's recent
> post
>> about possibly pulling in javax.jdo.* packages during the enhancement
>> processing [1] reminded me that we still have old remnants of JDO (and
>> Kodo)
>> in the OpenJPA code base. OpenJPA has never claimed support for JDO (nor
>> Kodo). Way back when, BEA provided a JDO implementation as part of their
>> offering that sat on top of OpenJPA. As far as I know, BEA (and Oracle)
>> only support the 1.1.x service stream of OpenJPA. So, if we did this in
>> the
>> 2.x stream, there should be no effects to that set of users.
>>
>> Would there be a concern with the current users of OpenJPA to clean up the
>> code base and remove these JDO/Kodo references? From a JPA/OpenJPA
>> perspective, you should see no differences in functionality.
>>
>> Like I said, Marc's posting prompted me to revisit this topic. I'm
> just
>> exploring the option with no immediate plans of actually doing the work...
>>
>> Thanks,
>> Kevin
>>
>> [1]
>>
>>
> http://openjpa.208410.n2.nabble.com/weird-JDO-Exception-when-using-OpenJPA-2-Enhancer-tc6870122.html
>>
>
Re: [DISCUSS] JDO usage end-of-life?
Posted by Michael Dick <mi...@gmail.com>.
There is at least some interest from a subset of our users. Matthew Adams
and issue: OPENJPA-1744
<https://issues.apache.org/jira/browse/OPENJPA-1744>to add support for
JDO last July. I closed the issue, but Matthew responded
and the issue was reopened.
There hasn't been a lot of activity on the JIRA since then. There are some
users watching it, but no one has voted for it. If there's an outpouring of
support from the users list, and a committer (or aspiriing committer) is
interested in championing the effort, I'd be all for adding a JDO persona.
Absent a champion who is ready to dive into the code, I think that we should
clean up the references to jdo.
Even if OpenJPA removes the references to JDO, I'm sure a separate module
could be written that sits on top of our binaries. I suspect that's what BEA
/ Oracle did.
-mike
On Mon, Oct 10, 2011 at 8:18 AM, Kevin Sutter <kw...@gmail.com> wrote:
> Hi,
> Sorry to cross post to both forums, but I wanted to ensure that I hit
> everybody that might have an opinion on this JDO topic...
>
> Is the JDO personality of OpenJPA still being utilized? Marc's recent post
> about possibly pulling in javax.jdo.* packages during the enhancement
> processing [1] reminded me that we still have old remnants of JDO (and
> Kodo)
> in the OpenJPA code base. OpenJPA has never claimed support for JDO (nor
> Kodo). Way back when, BEA provided a JDO implementation as part of their
> offering that sat on top of OpenJPA. As far as I know, BEA (and Oracle)
> only support the 1.1.x service stream of OpenJPA. So, if we did this in
> the
> 2.x stream, there should be no effects to that set of users.
>
> Would there be a concern with the current users of OpenJPA to clean up the
> code base and remove these JDO/Kodo references? From a JPA/OpenJPA
> perspective, you should see no differences in functionality.
>
> Like I said, Marc's posting prompted me to revisit this topic. I'm just
> exploring the option with no immediate plans of actually doing the work...
>
> Thanks,
> Kevin
>
> [1]
>
> http://openjpa.208410.n2.nabble.com/weird-JDO-Exception-when-using-OpenJPA-2-Enhancer-tc6870122.html
>
Re: [DISCUSS] JDO usage end-of-life?
Posted by Michael Dick <mi...@gmail.com>.
There is at least some interest from a subset of our users. Matthew Adams
and issue: OPENJPA-1744
<https://issues.apache.org/jira/browse/OPENJPA-1744>to add support for
JDO last July. I closed the issue, but Matthew responded
and the issue was reopened.
There hasn't been a lot of activity on the JIRA since then. There are some
users watching it, but no one has voted for it. If there's an outpouring of
support from the users list, and a committer (or aspiriing committer) is
interested in championing the effort, I'd be all for adding a JDO persona.
Absent a champion who is ready to dive into the code, I think that we should
clean up the references to jdo.
Even if OpenJPA removes the references to JDO, I'm sure a separate module
could be written that sits on top of our binaries. I suspect that's what BEA
/ Oracle did.
-mike
On Mon, Oct 10, 2011 at 8:18 AM, Kevin Sutter <kw...@gmail.com> wrote:
> Hi,
> Sorry to cross post to both forums, but I wanted to ensure that I hit
> everybody that might have an opinion on this JDO topic...
>
> Is the JDO personality of OpenJPA still being utilized? Marc's recent post
> about possibly pulling in javax.jdo.* packages during the enhancement
> processing [1] reminded me that we still have old remnants of JDO (and
> Kodo)
> in the OpenJPA code base. OpenJPA has never claimed support for JDO (nor
> Kodo). Way back when, BEA provided a JDO implementation as part of their
> offering that sat on top of OpenJPA. As far as I know, BEA (and Oracle)
> only support the 1.1.x service stream of OpenJPA. So, if we did this in
> the
> 2.x stream, there should be no effects to that set of users.
>
> Would there be a concern with the current users of OpenJPA to clean up the
> code base and remove these JDO/Kodo references? From a JPA/OpenJPA
> perspective, you should see no differences in functionality.
>
> Like I said, Marc's posting prompted me to revisit this topic. I'm just
> exploring the option with no immediate plans of actually doing the work...
>
> Thanks,
> Kevin
>
> [1]
>
> http://openjpa.208410.n2.nabble.com/weird-JDO-Exception-when-using-OpenJPA-2-Enhancer-tc6870122.html
>