[jira] Created: (OFBIZ-3520) revision 897605 breaks certain delegator.find() EntityListIterator calls

[Nicolas Malin (Jira)]

revision 897605 breaks certain delegator.find() EntityListIterator calls
------------------------------------------------------------------------

                 Key: OFBIZ-3520
                 URL: https://issues.apache.org/jira/browse/OFBIZ-3520
             Project: OFBiz
          Issue Type: Bug
          Components: framework
    Affects Versions: SVN trunk
            Reporter: Adam Heath
            Assignee: David E. Jones


We recently upgraded our internal ofbiz package to a newer trunk version, one from the end of January.  Ean then deployed that to the server it was developing on.  This broke requirements processing.  I have reduced it, however, to a simple patch, that works if I revert 897606, but breaks when it is applied.

Test case will be attached.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

[Nicolas Malin (Jira)]

     [ https://issues.apache.org/jira/browse/OFBIZ-3520?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Adam Heath updated OFBIZ-3520:
------------------------------

    Attachment: 897606-testcase.patch

> revision 897605 breaks certain delegator.find() EntityListIterator calls
> ------------------------------------------------------------------------
>
>                 Key: OFBIZ-3520
>                 URL: https://issues.apache.org/jira/browse/OFBIZ-3520
>             Project: OFBiz
>          Issue Type: Bug
>          Components: framework
>    Affects Versions: SVN trunk
>            Reporter: Adam Heath
>            Assignee: David E. Jones
>         Attachments: 897606-testcase.patch
>
>
> We recently upgraded our internal ofbiz package to a newer trunk version, one from the end of January.  Ean then deployed that to the server it was developing on.  This broke requirements processing.  I have reduced it, however, to a simple patch, that works if I revert 897606, but breaks when it is applied.
> Test case will be attached.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

[Adam Heath-2]

Adam Heath (JIRA) wrote:

>      [ https://issues.apache.org/jira/browse/OFBIZ-3520?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
>
> Adam Heath updated OFBIZ-3520:
> ------------------------------
>
>     Attachment: 897606-testcase.patch
>
>> revision 897605 breaks certain delegator.find() EntityListIterator calls
>> ------------------------------------------------------------------------
>>
>>                 Key: OFBIZ-3520
>>                 URL: https://issues.apache.org/jira/browse/OFBIZ-3520
>>             Project: OFBiz
>>          Issue Type: Bug
>>          Components: framework
>>    Affects Versions: SVN trunk
>>            Reporter: Adam Heath
>>            Assignee: David E. Jones
>>         Attachments: 897606-testcase.patch
>>
>>
>> We recently upgraded our internal ofbiz package to a newer trunk version, one from the end of January.  Ean then deployed that to the server it was developing on.  This broke requirements processing.  I have reduced it, however, to a simple patch, that works if I revert 897606, but breaks when it is applied.
>> Test case will be attached.

David, could you please take a look at this, as it is your commit that
 causes this to break.

[David E. Jones-2]

On Feb 26, 2010, at 10:33 AM, Adam Heath wrote:

> Adam Heath (JIRA) wrote:
>>     [ https://issues.apache.org/jira/browse/OFBIZ-3520?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
>>
>> Adam Heath updated OFBIZ-3520:
>> ------------------------------
>>
>>    Attachment: 897606-testcase.patch
>>
>>> revision 897605 breaks certain delegator.find() EntityListIterator calls
>>> ------------------------------------------------------------------------
>>>
>>>                Key: OFBIZ-3520
>>>                URL: https://issues.apache.org/jira/browse/OFBIZ-3520
>>>            Project: OFBiz
>>>         Issue Type: Bug
>>>         Components: framework
>>>   Affects Versions: SVN trunk
>>>           Reporter: Adam Heath
>>>           Assignee: David E. Jones
>>>        Attachments: 897606-testcase.patch
>>>
>>>
>>> We recently upgraded our internal ofbiz package to a newer trunk version, one from the end of January.  Ean then deployed that to the server it was developing on.  This broke requirements processing.  I have reduced it, however, to a simple patch, that works if I revert 897606, but breaks when it is applied.
>>> Test case will be attached.
>
> David, could you please take a look at this, as it is your commit that
> causes this to break.

Did you read the discussion about this?

Just search for the revision number.

I believe it has already been fixed, and you'll either have to update to something more recent or back-port the fix.

On a side note, so what if my commit caused it to break? Does that mean that only I can fix it? What a depressingly disempowering notion.

-David

[Adam Heath-2]

David E Jones wrote:

> On Feb 26, 2010, at 10:33 AM, Adam Heath wrote:
>
>> Adam Heath (JIRA) wrote:
>>>     [ https://issues.apache.org/jira/browse/OFBIZ-3520?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
>>>
>>> Adam Heath updated OFBIZ-3520:
>>> ------------------------------
>>>
>>>    Attachment: 897606-testcase.patch
>>>
>>>> revision 897605 breaks certain delegator.find() EntityListIterator calls
>>>> ------------------------------------------------------------------------
>>>>
>>>>                Key: OFBIZ-3520
>>>>                URL: https://issues.apache.org/jira/browse/OFBIZ-3520
>>>>            Project: OFBiz
>>>>         Issue Type: Bug
>>>>         Components: framework
>>>>   Affects Versions: SVN trunk
>>>>           Reporter: Adam Heath
>>>>           Assignee: David E. Jones
>>>>        Attachments: 897606-testcase.patch
>>>>
>>>>
>>>> We recently upgraded our internal ofbiz package to a newer trunk version, one from the end of January.  Ean then deployed that to the server it was developing on.  This broke requirements processing.  I have reduced it, however, to a simple patch, that works if I revert 897606, but breaks when it is applied.
>>>> Test case will be attached.
>> David, could you please take a look at this, as it is your commit that
>> causes this to break.
>
> Did you read the discussion about this?
>
> Just search for the revision number.
>
> I believe it has already been fixed, and you'll either have to update to something more recent or back-port the fix.
>
> On a side note, so what if my commit caused it to break? Does that mean that only I can fix it? What a depressingly disempowering notion.
>
> -David

My test case is against trunk.  It doesn't work in trunk.  When I undo
 897606, it works again.  What does that have to do with any other
kind of fix for any other part of the system?

I tried looking at the commit(it's small), but didn't fully understand
the situation in which is is useful.  Since you did the commit, I
figured you'd know more about what was going on.  Of course others can
look at it, nothing is stopping them.  But you obviously know more
about this, as you are the one who did the change.  Why do you have to
be so combative?

>
>

[Adam Heath-2]

Adam Heath wrote:

> David E Jones wrote:
>> On Feb 26, 2010, at 10:33 AM, Adam Heath wrote:
>>
>>> Adam Heath (JIRA) wrote:
>>>>     [ https://issues.apache.org/jira/browse/OFBIZ-3520?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
>>>>
>>>> Adam Heath updated OFBIZ-3520:
>>>> ------------------------------
>>>>
>>>>    Attachment: 897606-testcase.patch
>>>>
>>>>> revision 897605 breaks certain delegator.find() EntityListIterator calls
>>>>> ------------------------------------------------------------------------
>>>>>
>>>>>                Key: OFBIZ-3520
>>>>>                URL: https://issues.apache.org/jira/browse/OFBIZ-3520
>>>>>            Project: OFBiz
>>>>>         Issue Type: Bug
>>>>>         Components: framework
>>>>>   Affects Versions: SVN trunk
>>>>>           Reporter: Adam Heath
>>>>>           Assignee: David E. Jones
>>>>>        Attachments: 897606-testcase.patch
>>>>>
>>>>>
>>>>> We recently upgraded our internal ofbiz package to a newer trunk version, one from the end of January.  Ean then deployed that to the server it was developing on.  This broke requirements processing.  I have reduced it, however, to a simple patch, that works if I revert 897606, but breaks when it is applied.
>>>>> Test case will be attached.
>>> David, could you please take a look at this, as it is your commit that
>>> causes this to break.
>> Did you read the discussion about this?

Which discussion?  I don't see any discussion mentioned in the
changelog, I went back to Jan 10, don't see anything near by.

>> Just search for the revision number.

I searched my own mail archive, which goes back to 12-31, and there
are no hits on 897606, except for your original commit mail, and this
issue/thread.

>> I believe it has already been fixed, and you'll either have to update to something more recent or back-port the fix.

The issue in the requirements system is not the problem.  I have given
a test case that stands by itself, that is broken with your commit
applied, and works without it.  The test case is doing something
completely reasonable, and was supported by previous versions of ofbiz.

I have checked that I didn't miss anything, and I haven't.

[Ean Schuessler]

David E Jones wrote:
> Did you read the discussion about this?
>  
Do you have a link for that? We didn't see anything.
> Just search for the revision number.
>
> I believe it has already been fixed, and you'll either have to update to something more recent or back-port the fix.
>
> On a side note, so what if my commit caused it to break? Does that mean that only I can fix it? What a depressingly disempowering notion.
>  
It still seems to be broken in trunk. We would fix it if we understood
your change. It looks like your change breaks using a count on any
select against a view with a group by, that's kind of a big deal. For us
it breaks on PostgreSQL and Derby. Could you please check out the test case?

Thanks,
Ean

--
Ean Schuessler, CTO
[hidden email]
214-720-0700 x 315
Brainfood, Inc.
http://www.brainfood.com

[Nicolas Malin (Jira)]

    [ https://issues.apache.org/jira/browse/OFBIZ-3520?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12838969#action_12838969 ]

David E. Jones commented on OFBIZ-3520:
---------------------------------------

When you write 897606 do you mean 897605, ie the same revision as in the header?

Which revision of the trunk did you test with, the most recent? You wrote: "a newer trunk version, one from the end of January." Does that mean you updated to a trunk revision from the end of January?

Also, the test case does not have any assertions, which brings up a couple of questions:

1. what happened when you ran this? (any errors, exceptions, etc?)
2. how is that different from what you expected to happen?

One last thing (I think last anyway), since this deals with SQL generation that is database sensitive, which database were you running this against?

> revision 897605 breaks certain delegator.find() EntityListIterator calls
> ------------------------------------------------------------------------
>
>                 Key: OFBIZ-3520
>                 URL: https://issues.apache.org/jira/browse/OFBIZ-3520
>             Project: OFBiz
>          Issue Type: Bug
>          Components: framework
>    Affects Versions: SVN trunk
>            Reporter: Adam Heath
>            Assignee: David E. Jones
>         Attachments: 897606-testcase.patch
>
>
> We recently upgraded our internal ofbiz package to a newer trunk version, one from the end of January.  Ean then deployed that to the server it was developing on.  This broke requirements processing.  I have reduced it, however, to a simple patch, that works if I revert 897606, but breaks when it is applied.
> Test case will be attached.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

[David E. Jones-2]

On Feb 26, 2010, at 11:01 AM, Adam Heath wrote:

> Adam Heath wrote:
>> David E Jones wrote:
>>> On Feb 26, 2010, at 10:33 AM, Adam Heath wrote:
>>>
>>>> Adam Heath (JIRA) wrote:
>>>>>    [ https://issues.apache.org/jira/browse/OFBIZ-3520?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
>>>>>
>>>>> Adam Heath updated OFBIZ-3520:
>>>>> ------------------------------
>>>>>
>>>>>   Attachment: 897606-testcase.patch
>>>>>
>>>>>> revision 897605 breaks certain delegator.find() EntityListIterator calls
>>>>>> ------------------------------------------------------------------------
>>>>>>
>>>>>>               Key: OFBIZ-3520
>>>>>>               URL: https://issues.apache.org/jira/browse/OFBIZ-3520
>>>>>>           Project: OFBiz
>>>>>>        Issue Type: Bug
>>>>>>        Components: framework
>>>>>>  Affects Versions: SVN trunk
>>>>>>          Reporter: Adam Heath
>>>>>>          Assignee: David E. Jones
>>>>>>       Attachments: 897606-testcase.patch
>>>>>>
>>>>>>
>>>>>> We recently upgraded our internal ofbiz package to a newer trunk version, one from the end of January.  Ean then deployed that to the server it was developing on.  This broke requirements processing.  I have reduced it, however, to a simple patch, that works if I revert 897606, but breaks when it is applied.
>>>>>> Test case will be attached.
>>>> David, could you please take a look at this, as it is your commit that
>>>> causes this to break.
>>> Did you read the discussion about this?
>
> Which discussion?  I don't see any discussion mentioned in the
> changelog, I went back to Jan 10, don't see anything near by.
>
>>> Just search for the revision number.
>
> I searched my own mail archive, which goes back to 12-31, and there
> are no hits on 897606, except for your original commit mail, and this
> issue/thread.
>
>>> I believe it has already been fixed, and you'll either have to update to something more recent or back-port the fix.
>
> The issue in the requirements system is not the problem.  I have given
> a test case that stands by itself, that is broken with your commit
> applied, and works without it.  The test case is doing something
> completely reasonable, and was supported by previous versions of ofbiz.
>
> I have checked that I didn't miss anything, and I haven't.

Of course you haven't! It sounds like you have a complete understanding of the problem.

Oh, BTW the rev number is 897605, not 897606.

Please see my other comments and questions on the Jira issue, if you care to of course.

On a side note, about being combative, why spend so much time isolating a revision instead of understanding the technical problem? I guess I ask because I pretty much never bother to track down who caused the problem and when... what good does that do except allow me to play the blame game. It certainly doesn't help me fix the problem. I suppose that's just my style, and I guess that conflicts with your style. I just hope I'm always around to help you with every thing I might have worked on in OFBiz.

-David

[Nicolas Malin (Jira)]

    [ https://issues.apache.org/jira/browse/OFBIZ-3520?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12838972#action_12838972 ]

Adam Heath commented on OFBIZ-3520:
-----------------------------------

Yeah, sorry, 897605 is the one that I am talking about.

I tried initially with a revsion from a few days ago.  But the one that we were running when we discovered the problem was from the end of januanry, 902021.

The test case doesn't need an assertion.  An exception is throw, which leaves the method, and is caught by junit, causing the test case to fail.  A proper run will not throw a test case, and that is not something that you can test with junit, and still maintain 100% coverage of the test case.

This breaks with derby, haven't tried the test case against any other database.

--
[JUNIT (error)] - testFoo : org.ofbiz.entity.GenericDataSourceException: SQL Exception while executing the following:SELECT COUNT(1) FROM (SELECT COUNT(DISTINCT COUNT(DISTINCT T.TESTING_TYPE_ID)) FROM OFBIZ.TESTING T INNER JOIN OFBIZ.TESTING_NODE_MEMBER TNM ON T.TESTING_ID = TNM.TESTING_ID GROUP BY TNM.TESTING_NODE_ID) TEMP_NAME (Aggregate COUNT contains one or more aggregates.)
--

> revision 897605 breaks certain delegator.find() EntityListIterator calls
> ------------------------------------------------------------------------
>
>                 Key: OFBIZ-3520
>                 URL: https://issues.apache.org/jira/browse/OFBIZ-3520
>             Project: OFBiz
>          Issue Type: Bug
>          Components: framework
>    Affects Versions: SVN trunk
>            Reporter: Adam Heath
>            Assignee: David E. Jones
>         Attachments: 897606-testcase.patch
>
>
> We recently upgraded our internal ofbiz package to a newer trunk version, one from the end of January.  Ean then deployed that to the server it was developing on.  This broke requirements processing.  I have reduced it, however, to a simple patch, that works if I revert 897606, but breaks when it is applied.
> Test case will be attached.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

[Adam Heath-2]

David E Jones wrote:

> On Feb 26, 2010, at 11:01 AM, Adam Heath wrote:
>
>> Adam Heath wrote:
>>> David E Jones wrote:
>>>> On Feb 26, 2010, at 10:33 AM, Adam Heath wrote:
>>>>
>>>>> Adam Heath (JIRA) wrote:
>>>>>>    [ https://issues.apache.org/jira/browse/OFBIZ-3520?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
>>>>>>
>>>>>> Adam Heath updated OFBIZ-3520:
>>>>>> ------------------------------
>>>>>>
>>>>>>   Attachment: 897606-testcase.patch
>>>>>>
>>>>>>> revision 897605 breaks certain delegator.find() EntityListIterator calls
>>>>>>> ------------------------------------------------------------------------
>>>>>>>
>>>>>>>               Key: OFBIZ-3520
>>>>>>>               URL: https://issues.apache.org/jira/browse/OFBIZ-3520
>>>>>>>           Project: OFBiz
>>>>>>>        Issue Type: Bug
>>>>>>>        Components: framework
>>>>>>>  Affects Versions: SVN trunk
>>>>>>>          Reporter: Adam Heath
>>>>>>>          Assignee: David E. Jones
>>>>>>>       Attachments: 897606-testcase.patch
>>>>>>>
>>>>>>>
>>>>>>> We recently upgraded our internal ofbiz package to a newer trunk version, one from the end of January.  Ean then deployed that to the server it was developing on.  This broke requirements processing.  I have reduced it, however, to a simple patch, that works if I revert 897606, but breaks when it is applied.
>>>>>>> Test case will be attached.
>>>>> David, could you please take a look at this, as it is your commit that
>>>>> causes this to break.
>>>> Did you read the discussion about this?
>> Which discussion?  I don't see any discussion mentioned in the
>> changelog, I went back to Jan 10, don't see anything near by.
>>
>>>> Just search for the revision number.
>> I searched my own mail archive, which goes back to 12-31, and there
>> are no hits on 897606, except for your original commit mail, and this
>> issue/thread.
>>
>>>> I believe it has already been fixed, and you'll either have to update to something more recent or back-port the fix.
>> The issue in the requirements system is not the problem.  I have given
>> a test case that stands by itself, that is broken with your commit
>> applied, and works without it.  The test case is doing something
>> completely reasonable, and was supported by previous versions of ofbiz.
>>
>> I have checked that I didn't miss anything, and I haven't.
>
> Of course you haven't! It sounds like you have a complete understanding of the problem.

If I had a complete understanding of the problem, I would have fixed
it.  But I saw no discussion about this change, so I can't infer what
others haven't said.

You said previously that there was a discussion about this earlier.  I
then said that I couldn't find it.  So, what was the discussion?  Give
me something to search for.

> Oh, BTW the rev number is 897605, not 897606.

Yeah, my bad/

>
> Please see my other comments and questions on the Jira issue, if you care to of course.
>
> On a side note, about being combative, why spend so much time isolating a revision instead of understanding the technical problem? I guess I ask because I pretty much never bother to track down who caused the problem and when... what good does that do except allow me to play the blame game. It certainly doesn't help me fix the problem. I suppose that's just my style, and I guess that conflicts with your style. I just hope I'm always around to help you with every thing I might have worked on in OFBiz.

Isolating a single revision that causes a problem doesn't pin the
blame.  It shows what needs to be fixed.  Would it have been better if
I just filed a test case that didn't work, and made others
investigate?  I've saved others time by finding the single revision
that causes the case to break.

This is not a blame game, no.  I never blamed you for causing this
breakage.  You just did the original commit, so I assumed you'd know
more about the situation that your change would be useful.  Your
commit was trying to fix something else, so it's needed in some case.

[Nicolas Malin (Jira)]

    [ https://issues.apache.org/jira/browse/OFBIZ-3520?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12838978#action_12838978 ]

David E. Jones commented on OFBIZ-3520:
---------------------------------------

In rev 916781 this "COUNT(DISTINCT " stuff has been disabled with some comments about the issues. You'll want to update past 916782 as well since that undoes a stupid thing I left on accident from starting to test this, before realizing that giving up is the better approach here. There are just too many issues with it.

> revision 897605 breaks certain delegator.find() EntityListIterator calls
> ------------------------------------------------------------------------
>
>                 Key: OFBIZ-3520
>                 URL: https://issues.apache.org/jira/browse/OFBIZ-3520
>             Project: OFBiz
>          Issue Type: Bug
>          Components: framework
>    Affects Versions: SVN trunk
>            Reporter: Adam Heath
>            Assignee: David E. Jones
>         Attachments: 897606-testcase.patch
>
>
> We recently upgraded our internal ofbiz package to a newer trunk version, one from the end of January.  Ean then deployed that to the server it was developing on.  This broke requirements processing.  I have reduced it, however, to a simple patch, that works if I revert 897606, but breaks when it is applied.
> Test case will be attached.

--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

[rohit]

Hi,

I too had a problem with this commit and had reported it on this thread,

http://n4.nabble.com/RE-svn-commit-r897605-in-ofbiz-trunk-framework-entity-src-org-ofbiz-entity-datasource-GenericDAO-java-td1013432.html#a1016339

I am not sure if we are talking exactly about the same issue here, anyway I hope it helps.

Rohit

[Adam Heath-2]

rohit wrote:
> Hi,
>
> I too had a problem with this commit and had reported it on this thread,
>
> http://n4.nabble.com/RE-svn-commit-r897605-in-ofbiz-trunk-framework-entity-src-org-ofbiz-entity-datasource-GenericDAO-java-td1013432.html#a1016339
>
> I am not sure if we are talking exactly about the same issue here, anyway I
> hope it helps.

Yeah, that appears to be the same issue.  I was able to reduce it to a
very simple test case, that shows purely bad SQL being generated, that
wouldn't work in any case(the test case ran on derby, I discovered the
problem with postgres, and David tried it on mysql).

I'm still waiting on the details that caused this change to be
implemented; maybe there's another way to fix the original issue.