I have been using SDN 2.1.0.M1 for little more than a month and I
getting closer and closer to giving up.
I find myself really struggling with the mystery going on in SDN
aspectj version. Trying to find a bug is way to complicated.
I have worked with java since 1.0 and am not used to spend so much
time trying to debug my datastore. Do not get me wrong I love neo4j
and what it contributes. The problem is the SDN aspectj layer.
We are trying to build a non trivial application based on SDN and I
soon find myself in trouble, here are some examples:
* If you declare a static field on the NodeEntities (like a logger)
and tries to access it the code crashes with strange exceptions like
"the this is not present".
* When executing a findByQuery and there are multiple results you get
an Iterator not empty exception. I get it, you get an iterator but the
error is really misleading!
* The reason there where multiple objects is that the check for
duplicates used an index lookup and the index did not get updated in
time to stop the second insert.
* The (for me) strange mix of Set and Iterables as results from the
mapping (if you like to add something) and the queries makes it more
difficult than it have to be. For me it means i have to have two
methods to convert the result into json for example.
I have many more examples but my question is this: Is anyone actually
using SDN for any production system. For me the upgrade to RC1 from M1
actually broke my code so I am just wondering if I should bother to
fix it or throw out SDN?
> I have been using SDN 2.1.0.M1 for little more than a month and I
> getting closer and closer to giving up.
> I find myself really struggling with the mystery going on in SDN
> aspectj version. Trying to find a bug is way to complicated.
> I have worked with java since 1.0 and am not used to spend so much
> time trying to debug my datastore. Do not get me wrong I love neo4j
> and what it contributes. The problem is the SDN aspectj layer.
> We are trying to build a non trivial application based on SDN and I
> soon find myself in trouble, here are some examples:
> * If you declare a static field on the NodeEntities (like a logger)
> and tries to access it the code crashes with strange exceptions like
> "the this is not present".
> I haven't seen this issue ...are you certain it's SDN related?
> * When executing a findByQuery and there are multiple results you get
> an Iterator not empty exception. I get it, you get an iterator but the
> error is really misleading!
> We generally tend to add specific methods to the repository interface and
decorate with query annotation and return Sets or Collection.
> * The reason there where multiple objects is that the check for
> duplicates used an index lookup and the index did not get updated in
> time to stop the second insert.
> Would need more info as to what you are doing.
> * The (for me) strange mix of Set and Iterables as results from the
> mapping (if you like to add something) and the queries makes it more
> difficult than it have to be. For me it means i have to have two
> methods to convert the result into json for example.
> > Have u looked at using org.codehaus.jackson Json annotations in your
> I have many more examples but my question is this: Is anyone actually
> using SDN for any production system. For me the upgrade to RC1 from M1
> actually broke my code so I am just wondering if I should bother to
> fix it or throw out SDN?
I'm in a similar situation than Stacy's one. Our production phase is
planned for June and I'm using SDN.
As he does I add specific methods to the repository interface and decorate
them with query (Cypher or Gremlin, depends of my mind) annotations and
return Sets. And I give a +1 to the Jackson solution to return JSON (I'm
also using it).
Some residual bugs can occur sometimes. It's because SDN is pretty young
for now. But Michael (Hunger, from Neo Tech) is very reactive and
accommodating. And if I remember well, Neo Tech planned to hire a full-time
SDN developer by early summer.
Cheers,
Marc
On Tue, May 22, 2012 at 5:12 PM, Stacy Young <st...@actualtechnology.net>wrote:
> Yes, we're using in a system scheduled for production in the June
> timeframe. While I cannot speak to all of your points, some notes inline.
> On Tue, May 22, 2012 at 9:57 AM, Fredrik Hultin <
> fredrik.hul...@collabra.se> wrote:
>> I have been using SDN 2.1.0.M1 for little more than a month and I
>> getting closer and closer to giving up.
>> I find myself really struggling with the mystery going on in SDN
>> aspectj version. Trying to find a bug is way to complicated.
>> I have worked with java since 1.0 and am not used to spend so much
>> time trying to debug my datastore. Do not get me wrong I love neo4j
>> and what it contributes. The problem is the SDN aspectj layer.
>> We are trying to build a non trivial application based on SDN and I
>> soon find myself in trouble, here are some examples:
>> * If you declare a static field on the NodeEntities (like a logger)
>> and tries to access it the code crashes with strange exceptions like
>> "the this is not present".
> > I haven't seen this issue ...are you certain it's SDN related?
>> * When executing a findByQuery and there are multiple results you get
>> an Iterator not empty exception. I get it, you get an iterator but the
>> error is really misleading!
> > We generally tend to add specific methods to the repository interface
> and decorate with query annotation and return Sets or Collection.
>> * The reason there where multiple objects is that the check for
>> duplicates used an index lookup and the index did not get updated in
>> time to stop the second insert.
> > Would need more info as to what you are doing.
>> * The (for me) strange mix of Set and Iterables as results from the
>> mapping (if you like to add something) and the queries makes it more
>> difficult than it have to be. For me it means i have to have two
>> methods to convert the result into json for example.
>> > Have u looked at using org.codehaus.jackson Json annotations in your
> entities? We've had a lot of success
>> I have many more examples but my question is this: Is anyone actually
>> using SDN for any production system. For me the upgrade to RC1 from M1
>> actually broke my code so I am just wondering if I should bother to
>> fix it or throw out SDN?
Great, thx for the reply.
I will look into the jackson annotations and se if it also helps me. I
just wanted to make sure I was not the only one on the ship since
there was for me some really strange behaviours.
Thanks guys ;)
/Fredrik
On May 23, 8:45 am, Marc de Verdelhan <marc.deverdel...@yahoo.com>
wrote:
> I'm in a similar situation than Stacy's one. Our production phase is
> planned for June and I'm using SDN.
> As he does I add specific methods to the repository interface and decorate
> them with query (Cypher or Gremlin, depends of my mind) annotations and
> return Sets. And I give a +1 to the Jackson solution to return JSON (I'm
> also using it).
> Some residual bugs can occur sometimes. It's because SDN is pretty young
> for now. But Michael (Hunger, from Neo Tech) is very reactive and
> accommodating. And if I remember well, Neo Tech planned to hire a full-time
> SDN developer by early summer.
> Cheers,
> Marc
> On Tue, May 22, 2012 at 5:12 PM, Stacy Young <st...@actualtechnology.net>wrote:
> > Yes, we're using in a system scheduled for production in the June
> > timeframe. While I cannot speak to all of your points, some notes inline.
> > On Tue, May 22, 2012 at 9:57 AM, Fredrik Hultin <
> > fredrik.hul...@collabra.se> wrote:
> >> I have been using SDN 2.1.0.M1 for little more than a month and I
> >> getting closer and closer to giving up.
> >> I find myself really struggling with the mystery going on in SDN
> >> aspectj version. Trying to find a bug is way to complicated.
> >> I have worked with java since 1.0 and am not used to spend so much
> >> time trying to debug my datastore. Do not get me wrong I love neo4j
> >> and what it contributes. The problem is the SDN aspectj layer.
> >> We are trying to build a non trivial application based on SDN and I
> >> soon find myself in trouble, here are some examples:
> >> * If you declare a static field on the NodeEntities (like a logger)
> >> and tries to access it the code crashes with strange exceptions like
> >> "the this is not present".
> > > I haven't seen this issue ...are you certain it's SDN related?
> >> * When executing a findByQuery and there are multiple results you get
> >> an Iterator not empty exception. I get it, you get an iterator but the
> >> error is really misleading!
> > > We generally tend to add specific methods to the repository interface
> > and decorate with query annotation and return Sets or Collection.
> >> * The reason there where multiple objects is that the check for
> >> duplicates used an index lookup and the index did not get updated in
> >> time to stop the second insert.
> > > Would need more info as to what you are doing.
> >> * The (for me) strange mix of Set and Iterables as results from the
> >> mapping (if you like to add something) and the queries makes it more
> >> difficult than it have to be. For me it means i have to have two
> >> methods to convert the result into json for example.
> >> > Have u looked at using org.codehaus.jackson Json annotations in your
> > entities? We've had a lot of success
> >> I have many more examples but my question is this: Is anyone actually
> >> using SDN for any production system. For me the upgrade to RC1 from M1
> >> actually broke my code so I am just wondering if I should bother to
> >> fix it or throw out SDN?
> Great, thx for the reply.
> I will look into the jackson annotations and se if it also helps me. I
> just wanted to make sure I was not the only one on the ship since
> there was for me some really strange behaviours.
> Thanks guys ;)
> /Fredrik
> On May 23, 8:45 am, Marc de Verdelhan <marc.deverdel...@yahoo.com>
> wrote:
>> I'm in a similar situation than Stacy's one. Our production phase is
>> planned for June and I'm using SDN.
>> As he does I add specific methods to the repository interface and decorate
>> them with query (Cypher or Gremlin, depends of my mind) annotations and
>> return Sets. And I give a +1 to the Jackson solution to return JSON (I'm
>> also using it).
>> Some residual bugs can occur sometimes. It's because SDN is pretty young
>> for now. But Michael (Hunger, from Neo Tech) is very reactive and
>> accommodating. And if I remember well, Neo Tech planned to hire a full-time
>> SDN developer by early summer.
>> Cheers,
>> Marc
>> On Tue, May 22, 2012 at 5:12 PM, Stacy Young <st...@actualtechnology.net>wrote:
>>> Yes, we're using in a system scheduled for production in the June
>>> timeframe. While I cannot speak to all of your points, some notes inline.
>>> On Tue, May 22, 2012 at 9:57 AM, Fredrik Hultin <
>>> fredrik.hul...@collabra.se> wrote:
>>>> I have been using SDN 2.1.0.M1 for little more than a month and I
>>>> getting closer and closer to giving up.
>>>> I find myself really struggling with the mystery going on in SDN
>>>> aspectj version. Trying to find a bug is way to complicated.
>>>> I have worked with java since 1.0 and am not used to spend so much
>>>> time trying to debug my datastore. Do not get me wrong I love neo4j
>>>> and what it contributes. The problem is the SDN aspectj layer.
>>>> We are trying to build a non trivial application based on SDN and I
>>>> soon find myself in trouble, here are some examples:
>>>> * If you declare a static field on the NodeEntities (like a logger)
>>>> and tries to access it the code crashes with strange exceptions like
>>>> "the this is not present".
Might be an issue with the pointcut not filtering out statics
>>>> I haven't seen this issue ...are you certain it's SDN related?
>>>> * When executing a findByQuery and there are multiple results you get
>>>> an Iterator not empty exception. I get it, you get an iterator but the
>>>> error is really misleading!
So a better DataAccessException makes sense
>>>> We generally tend to add specific methods to the repository interface
>>> and decorate with query annotation and return Sets or Collection.
>>>> * The reason there where multiple objects is that the check for
>>>> duplicates used an index lookup and the index did not get updated in
>>>> time to stop the second insert.
What do you mean here
>>>> Would need more info as to what you are doing.
>>>> * The (for me) strange mix of Set and Iterables as results from the
>>>> mapping (if you like to add something) and the queries makes it more
>>>> difficult than it have to be. For me it means i have to have two
>>>> methods to convert the result into json for example.
>>>>> Have u looked at using org.codehaus.jackson Json annotations in your
>>> entities? We've had a lot of success
>>>> I have many more examples but my question is this: Is anyone actually
>>>> using SDN for any production system. For me the upgrade to RC1 from M1
>>>> actually broke my code so I am just wondering if I should bother to
>>>> fix it or throw out SDN?
Hi Michael!
Thx for your reply. I am not sure I have time to create Jira for these
issues since it takes time from production and we are on a tight
schedule.
These are minor problems which are easy to workaround, I simply used
them since they where annoying and annoying things tend to be the
first things to get fixed ;). I just wanted to see if anyone else was
using the SDN since it is new and "untested". As you know it is not
fun to put in many hundreds (thousands) of hours and find yourself on
a sinking ship all alone.
I really think you guys are doing a great job with Neo4j. What is not
as clear is if Neo4j are as committed to SDN?
As it comes to reporting all my examples, that will also have to wait
until first release of our product unfortunately :( I will try and
report if i find showstoppers.
/Regards
Fredrik
On May 23, 9:58 am, Michael Hunger <michael.hun...@neopersistence.com>
wrote:
> As a set is also an Iterable why do you need two methods? And for custom repo methods you can use all collection interfaces.
> Sent from mobile device
> Am 23.05.2012 um 09:13 schrieb Fredrik Hultin <fredrik.hul...@collabra.se>:
> > Great, thx for the reply.
> > I will look into the jackson annotations and se if it also helps me. I
> > just wanted to make sure I was not the only one on the ship since
> > there was for me some really strange behaviours.
> > Thanks guys ;)
> > /Fredrik
> > On May 23, 8:45 am, Marc de Verdelhan <marc.deverdel...@yahoo.com>
> > wrote:
> >> I'm in a similar situation than Stacy's one. Our production phase is
> >> planned for June and I'm using SDN.
> >> As he does I add specific methods to the repository interface and decorate
> >> them with query (Cypher or Gremlin, depends of my mind) annotations and
> >> return Sets. And I give a +1 to the Jackson solution to return JSON (I'm
> >> also using it).
> >> Some residual bugs can occur sometimes. It's because SDN is pretty young
> >> for now. But Michael (Hunger, from Neo Tech) is very reactive and
> >> accommodating. And if I remember well, Neo Tech planned to hire a full-time
> >> SDN developer by early summer.
> >> Cheers,
> >> Marc
> >> On Tue, May 22, 2012 at 5:12 PM, Stacy Young <st...@actualtechnology.net>wrote:
> >>> Yes, we're using in a system scheduled for production in the June
> >>> timeframe. While I cannot speak to all of your points, some notes inline.
> >>> On Tue, May 22, 2012 at 9:57 AM, Fredrik Hultin <
> >>> fredrik.hul...@collabra.se> wrote:
> >>>> I have been using SDN 2.1.0.M1 for little more than a month and I
> >>>> getting closer and closer to giving up.
> >>>> I find myself really struggling with the mystery going on in SDN
> >>>> aspectj version. Trying to find a bug is way to complicated.
> >>>> I have worked with java since 1.0 and am not used to spend so much
> >>>> time trying to debug my datastore. Do not get me wrong I love neo4j
> >>>> and what it contributes. The problem is the SDN aspectj layer.
> >>>> We are trying to build a non trivial application based on SDN and I
> >>>> soon find myself in trouble, here are some examples:
> >>>> * If you declare a static field on the NodeEntities (like a logger)
> >>>> and tries to access it the code crashes with strange exceptions like
> >>>> "the this is not present".
> Might be an issue with the pointcut not filtering out statics
> >>>> I haven't seen this issue ...are you certain it's SDN related?
> >>>> * When executing a findByQuery and there are multiple results you get
> >>>> an Iterator not empty exception. I get it, you get an iterator but the
> >>>> error is really misleading!
> So a better DataAccessException makes sense
> >>>> We generally tend to add specific methods to the repository interface
> >>> and decorate with query annotation and return Sets or Collection.
> >>>> * The reason there where multiple objects is that the check for
> >>>> duplicates used an index lookup and the index did not get updated in
> >>>> time to stop the second insert.
> What do you mean here
> >>>> Would need more info as to what you are doing.
> >>>> * The (for me) strange mix of Set and Iterables as results from the
> >>>> mapping (if you like to add something) and the queries makes it more
> >>>> difficult than it have to be. For me it means i have to have two
> >>>> methods to convert the result into json for example.
> >>>>> Have u looked at using org.codehaus.jackson Json annotations in your
> >>> entities? We've had a lot of success
> >>>> I have many more examples but my question is this: Is anyone actually
> Can you please report all your examples?
> >>>> using SDN for any production system. For me the upgrade to RC1 from M1
> >>>> actually broke my code so I am just wondering if I should bother to
> >>>> fix it or throw out SDN?
Fredrik,
yes we are committed to SDN and actually increasing support for it
going forward (as mentioned earlier). So, you can rest assured that is
is not going to be dropped anytime soon.
<fredrik.hul...@collabra.se> wrote:
> Hi Michael!
> Thx for your reply. I am not sure I have time to create Jira for these
> issues since it takes time from production and we are on a tight
> schedule.
> These are minor problems which are easy to workaround, I simply used
> them since they where annoying and annoying things tend to be the
> first things to get fixed ;). I just wanted to see if anyone else was
> using the SDN since it is new and "untested". As you know it is not
> fun to put in many hundreds (thousands) of hours and find yourself on
> a sinking ship all alone.
> I really think you guys are doing a great job with Neo4j. What is not
> as clear is if Neo4j are as committed to SDN?
> As it comes to reporting all my examples, that will also have to wait
> until first release of our product unfortunately :( I will try and
> report if i find showstoppers.
> /Regards
> Fredrik
> On May 23, 9:58 am, Michael Hunger <michael.hun...@neopersistence.com>
> wrote:
>> Fredrick,
>> Which of these issues did you report before? I just recall one report from you (can't check right now).
>> I'd like to work with you on resolving thoses issues. I'm on vacation till thursday.
>> As a set is also an Iterable why do you need two methods? And for custom repo methods you can use all collection interfaces.
>> Sent from mobile device
>> Am 23.05.2012 um 09:13 schrieb Fredrik Hultin <fredrik.hul...@collabra.se>:
>> > Great, thx for the reply.
>> > I will look into the jackson annotations and se if it also helps me. I
>> > just wanted to make sure I was not the only one on the ship since
>> > there was for me some really strange behaviours.
>> > Thanks guys ;)
>> > /Fredrik
>> > On May 23, 8:45 am, Marc de Verdelhan <marc.deverdel...@yahoo.com>
>> > wrote:
>> >> I'm in a similar situation than Stacy's one. Our production phase is
>> >> planned for June and I'm using SDN.
>> >> As he does I add specific methods to the repository interface and decorate
>> >> them with query (Cypher or Gremlin, depends of my mind) annotations and
>> >> return Sets. And I give a +1 to the Jackson solution to return JSON (I'm
>> >> also using it).
>> >> Some residual bugs can occur sometimes. It's because SDN is pretty young
>> >> for now. But Michael (Hunger, from Neo Tech) is very reactive and
>> >> accommodating. And if I remember well, Neo Tech planned to hire a full-time
>> >> SDN developer by early summer.
>> >> Cheers,
>> >> Marc
>> >> On Tue, May 22, 2012 at 5:12 PM, Stacy Young <st...@actualtechnology.net>wrote:
>> >>> Yes, we're using in a system scheduled for production in the June
>> >>> timeframe. While I cannot speak to all of your points, some notes inline.
>> >>> On Tue, May 22, 2012 at 9:57 AM, Fredrik Hultin <
>> >>> fredrik.hul...@collabra.se> wrote:
>> >>>> I have been using SDN 2.1.0.M1 for little more than a month and I
>> >>>> getting closer and closer to giving up.
>> >>>> I find myself really struggling with the mystery going on in SDN
>> >>>> aspectj version. Trying to find a bug is way to complicated.
>> >>>> I have worked with java since 1.0 and am not used to spend so much
>> >>>> time trying to debug my datastore. Do not get me wrong I love neo4j
>> >>>> and what it contributes. The problem is the SDN aspectj layer.
>> >>>> We are trying to build a non trivial application based on SDN and I
>> >>>> soon find myself in trouble, here are some examples:
>> >>>> * If you declare a static field on the NodeEntities (like a logger)
>> >>>> and tries to access it the code crashes with strange exceptions like
>> >>>> "the this is not present".
>> Might be an issue with the pointcut not filtering out statics
>> >>>> I haven't seen this issue ...are you certain it's SDN related?
>> >>>> * When executing a findByQuery and there are multiple results you get
>> >>>> an Iterator not empty exception. I get it, you get an iterator but the
>> >>>> error is really misleading!
>> So a better DataAccessException makes sense
>> >>>> We generally tend to add specific methods to the repository interface
>> >>> and decorate with query annotation and return Sets or Collection.
>> >>>> * The reason there where multiple objects is that the check for
>> >>>> duplicates used an index lookup and the index did not get updated in
>> >>>> time to stop the second insert.
>> What do you mean here
>> >>>> Would need more info as to what you are doing.
>> >>>> * The (for me) strange mix of Set and Iterables as results from the
>> >>>> mapping (if you like to add something) and the queries makes it more
>> >>>> difficult than it have to be. For me it means i have to have two
>> >>>> methods to convert the result into json for example.
>> >>>>> Have u looked at using org.codehaus.jackson Json annotations in your
>> >>> entities? We've had a lot of success
>> >>>> I have many more examples but my question is this: Is anyone actually
>> Can you please report all your examples?
>> >>>> using SDN for any production system. For me the upgrade to RC1 from M1
>> >>>> actually broke my code so I am just wondering if I should bother to
>> >>>> fix it or throw out SDN?
> Hi Michael!
> Thx for your reply. I am not sure I have time to create Jira for these
> issues since it takes time from production and we are on a tight
> schedule.
> These are minor problems which are easy to workaround, I simply used
> them since they where annoying and annoying things tend to be the
> first things to get fixed ;). I just wanted to see if anyone else was
> using the SDN since it is new and "untested". As you know it is not
> fun to put in many hundreds (thousands) of hours and find yourself on
> a sinking ship all alone.
> I really think you guys are doing a great job with Neo4j. What is not
> as clear is if Neo4j are as committed to SDN?
> As it comes to reporting all my examples, that will also have to wait
> until first release of our product unfortunately :( I will try and
> report if i find showstoppers.
> /Regards
> Fredrik
> On May 23, 9:58 am, Michael Hunger <michael.hun...@neopersistence.com>
> wrote:
>> Fredrick,
>> Which of these issues did you report before? I just recall one report from you (can't check right now).
>> I'd like to work with you on resolving thoses issues. I'm on vacation till thursday.
>> As a set is also an Iterable why do you need two methods? And for custom repo methods you can use all collection interfaces.
>> Sent from mobile device
>> Am 23.05.2012 um 09:13 schrieb Fredrik Hultin <fredrik.hul...@collabra.se>:
>>> Great, thx for the reply.
>>> I will look into the jackson annotations and se if it also helps me. I
>>> just wanted to make sure I was not the only one on the ship since
>>> there was for me some really strange behaviours.
>>> Thanks guys ;)
>>> /Fredrik
>>> On May 23, 8:45 am, Marc de Verdelhan <marc.deverdel...@yahoo.com>
>>> wrote:
>>>> I'm in a similar situation than Stacy's one. Our production phase is
>>>> planned for June and I'm using SDN.
>>>> As he does I add specific methods to the repository interface and decorate
>>>> them with query (Cypher or Gremlin, depends of my mind) annotations and
>>>> return Sets. And I give a +1 to the Jackson solution to return JSON (I'm
>>>> also using it).
>>>> Some residual bugs can occur sometimes. It's because SDN is pretty young
>>>> for now. But Michael (Hunger, from Neo Tech) is very reactive and
>>>> accommodating. And if I remember well, Neo Tech planned to hire a full-time
>>>> SDN developer by early summer.
>>>> Cheers,
>>>> Marc
>>>> On Tue, May 22, 2012 at 5:12 PM, Stacy Young <st...@actualtechnology.net>wrote:
>>>>> Yes, we're using in a system scheduled for production in the June
>>>>> timeframe. While I cannot speak to all of your points, some notes inline.
>>>>> On Tue, May 22, 2012 at 9:57 AM, Fredrik Hultin <
>>>>> fredrik.hul...@collabra.se> wrote:
>>>>>> I have been using SDN 2.1.0.M1 for little more than a month and I
>>>>>> getting closer and closer to giving up.
>>>>>> I find myself really struggling with the mystery going on in SDN
>>>>>> aspectj version. Trying to find a bug is way to complicated.
>>>>>> I have worked with java since 1.0 and am not used to spend so much
>>>>>> time trying to debug my datastore. Do not get me wrong I love neo4j
>>>>>> and what it contributes. The problem is the SDN aspectj layer.
>>>>>> We are trying to build a non trivial application based on SDN and I
>>>>>> soon find myself in trouble, here are some examples:
>>>>>> * If you declare a static field on the NodeEntities (like a logger)
>>>>>> and tries to access it the code crashes with strange exceptions like
>>>>>> "the this is not present".
>> Might be an issue with the pointcut not filtering out statics
>>>>>> I haven't seen this issue ...are you certain it's SDN related?
>>>>>> * When executing a findByQuery and there are multiple results you get
>>>>>> an Iterator not empty exception. I get it, you get an iterator but the
>>>>>> error is really misleading!
>> So a better DataAccessException makes sense
>>>>>> We generally tend to add specific methods to the repository interface
>>>>> and decorate with query annotation and return Sets or Collection.
>>>>>> * The reason there where multiple objects is that the check for
>>>>>> duplicates used an index lookup and the index did not get updated in
>>>>>> time to stop the second insert.
>> What do you mean here
>>>>>> Would need more info as to what you are doing.
>>>>>> * The (for me) strange mix of Set and Iterables as results from the
>>>>>> mapping (if you like to add something) and the queries makes it more
>>>>>> difficult than it have to be. For me it means i have to have two
>>>>>> methods to convert the result into json for example.
>>>>>>> Have u looked at using org.codehaus.jackson Json annotations in your
>>>>> entities? We've had a lot of success
>>>>>> I have many more examples but my question is this: Is anyone actually
>> Can you please report all your examples?
>>>>>> using SDN for any production system. For me the upgrade to RC1 from M1
>>>>>> actually broke my code so I am just wondering if I should bother to
>>>>>> fix it or throw out SDN?
