Here are the last two issues. Ted, 729 seems to be related to the
Jenkins build issue and needs to be fixed but 768 seems innocuous enough
to slip into 0.7. What is your assessment?

Displaying 2issues at 11/Jan/12 18:04.
Issue Type Key Summary Assignee Reporter Priority Status
Resolution Created Updated Due Date
New Feature MAHOUT-792
<https://issues.apache.org/jira/browse/MAHOUT-792> Add new stochastic
decomposition code Ted Dunning Ted Dunning Major Reopened
Unresolved 8/20/11 21:33 1/11/12 3:23
Improvement MAHOUT-768
<https://issues.apache.org/jira/browse/MAHOUT-768> Duplicated
DoubleFunction in mahout and mahout-collections (mahout.math package).
Ted Dunning Dawid Weiss Minor Open Unresolved 7/24/11 7:38 1/8/12
18:41
Generated at Wed Jan 11 18:04:58 UTC 2012 by Jeff Eastman using JIRA
4.4.1#660-r161644.

I would say the simplest solution is to remove DoubleFunction from math
since it depends on collections anyway (maven dependency). I've just ran
mvn compile on the patch attached and everything seems to work fine. mvn
test is running in the background so I don't know if this change breaks
anything.

A better solution would be to fold-in collections into the trunk again to
avoid the duplication we're not seeing, but this may be too intrusive.

The third way is to ignore 768 and just live with the duplication.

By the way -- I just checked out the trunk, trying to compile it under Java
1.7 fails because new methods appeared in the DataSource interface, just
thought I'd bring it to your attention:

[ERROR] Failed to execute goal
org.apache.maven.plugins:maven-compiler-plugin:2.3.2:compile
(default-compile) on project
 mahout-integration: Compilation failure
[ERROR]
\Work\mahout\integration\src\main\java\org\apache\mahout\cf\taste\impl\model\jdbc\ConnectionPoolDataSource.java:
[39,13] error: ConnectionPoolDataSource is not abstract and does not
override abstract method getParentLogger() in Commo
nDataSource

Dawid
On Wed, Jan 11, 2012 at 7:09 PM, Jeff Eastman <[EMAIL PROTECTED]>wrote:

>  Here are the last two issues. Ted, 729 seems to be related to the Jenkins
> build issue and needs to be fixed but 768 seems innocuous enough to slip
> into 0.7. What is your assessment?
>
>        Displaying 2 issues at 11/Jan/12 18:04.  Issue Type Key Summary
> Assignee Reporter Priority Status Resolution Created Updated Due Date  New
> Feature MAHOUT-792 <https://issues.apache.org/jira/browse/MAHOUT-792> Add
> new stochastic decomposition code Ted Dunning Ted Dunning Major Reopened
> Unresolved 8/20/11 21:33 1/11/12 3:23    Improvement MAHOUT-768<https://issues.apache.org/jira/browse/MAHOUT-768> Duplicated
> DoubleFunction in mahout and mahout-collections (mahout.math package). Ted
> Dunning Dawid Weiss Minor Open Unresolved 7/24/11 7:38 1/8/12 18:41    Generated
> at Wed Jan 11 18:04:58 UTC 2012 by Jeff Eastman using JIRA
> 4.4.1#660-r161644.
>

Maybe I should clarify that my original motivation for this issue was to
leave DoubleFunction in math and remove it from collections and this is
problematic because it creates a circular dependency. Apparently going the
other way around (removing DoubleFunction from math and leaving it in
collections) is much simpler. I don't like the idea of a single package
scattered across two projects though so in the long something better would
have to be done about it.

Dawid

On Wed, Jan 11, 2012 at 7:49 PM, Dawid Weiss
<[EMAIL PROTECTED]>wrote:

>
> I would say the simplest solution is to remove DoubleFunction from math
> since it depends on collections anyway (maven dependency). I've just ran
> mvn compile on the patch attached and everything seems to work fine. mvn
> test is running in the background so I don't know if this change breaks
> anything.
>
> A better solution would be to fold-in collections into the trunk again to
> avoid the duplication we're not seeing, but this may be too intrusive.
>
> The third way is to ignore 768 and just live with the duplication.
>
> By the way -- I just checked out the trunk, trying to compile it under
> Java 1.7 fails because new methods appeared in the DataSource interface,
> just thought I'd bring it to your attention:
>
> [ERROR] Failed to execute goal
> org.apache.maven.plugins:maven-compiler-plugin:2.3.2:compile
> (default-compile) on project
>  mahout-integration: Compilation failure
> [ERROR]
> \Work\mahout\integration\src\main\java\org\apache\mahout\cf\taste\impl\model\jdbc\ConnectionPoolDataSource.java:
> [39,13] error: ConnectionPoolDataSource is not abstract and does not
> override abstract method getParentLogger() in Commo
> nDataSource
>
> Dawid
>
>
> On Wed, Jan 11, 2012 at 7:09 PM, Jeff Eastman <[EMAIL PROTECTED]>wrote:
>
>>  Here are the last two issues. Ted, 729 seems to be related to the
>> Jenkins build issue and needs to be fixed but 768 seems innocuous enough to
>> slip into 0.7. What is your assessment?
>>
>>        Displaying 2 issues at 11/Jan/12 18:04.  Issue Type Key Summary
>> Assignee Reporter Priority Status Resolution Created Updated Due Date  New
>> Feature MAHOUT-792 <https://issues.apache.org/jira/browse/MAHOUT-792> Add
>> new stochastic decomposition code Ted Dunning Ted Dunning Major Reopened
>> Unresolved 8/20/11 21:33 1/11/12 3:23    Improvement MAHOUT-768<https://issues.apache.org/jira/browse/MAHOUT-768> Duplicated
>> DoubleFunction in mahout and mahout-collections (mahout.math package). Ted
>> Dunning Dawid Weiss Minor Open Unresolved 7/24/11 7:38 1/8/12 18:41    Generated
>> at Wed Jan 11 18:04:58 UTC 2012 by Jeff Eastman using JIRA
>> 4.4.1#660-r161644.
>>
>
>