Done: http://jira.codehaus.org/browse/MNG-2541


brettporter wrote:
> 
> Please file a JIRA. It is a simple 6 letter addition (LinkedHashMap),  
> but it would be good that we could get tests to ensure this,  
> particularly since they could again be modified elsewhere.
> 
> Thanks!
> 
> - Brett
> 
> On 30/08/2006, at 3:42 PM, Gareth Tilley wrote:
> 
>>
>> Hi All,
>>
>> I've noticed that in
>> org.apache.maven.project.inheritance.DefaultModelInheritanceAssembler, 
>>  the
>> method below:
>> private void assembleDependencyInheritance( Model child, Model  
>> parent )
>>     {
>>         Map depsMap = new HashMap();
>>
>>         List deps = parent.getDependencies();
>>
>>         if ( deps != null )
>>         {
>>             for ( Iterator it = deps.iterator(); it.hasNext(); )
>>             {
>>                 Dependency dependency = (Dependency) it.next();
>>                 depsMap.put( dependency.getManagementKey(),  
>> dependency );
>>             }
>>         }
>>
>>         deps = child.getDependencies();
>>
>>         if ( deps != null )
>>         {
>>             for ( Iterator it = deps.iterator(); it.hasNext(); )
>>             {
>>                 Dependency dependency = (Dependency) it.next();
>>                 depsMap.put( dependency.getManagementKey(),  
>> dependency );
>>             }
>>         }
>>
>>         child.setDependencies( new ArrayList( depsMap.values() ) );
>>     }
>>
>> is called by the buildFromRepository method in
>> org.apache.maven.project.DefaultProjectBuilder (and I'm sure a few  
>> other
>> places too). The effect it has, is that the dependencies in the pom  
>> of the
>> project, come back in a different order to how they were declared  
>> in the
>> pom.
>>
>> Now, I'm guessing the reason for the above code being implemented  
>> the way it
>> is, is to allow easy handling of potential duplicates, the only  
>> problem is
>> that the order is lost in the process. Or is this intentional, if  
>> it is can
>> you explain why? If its not, I would think that there would be a large
>> benefit in changing the above code so that it maintains the order.  
>> I'm happy
>> to submit a JIRA request with my patch for it if that is the case.
>>
>> Regards
>> Gareth Tilley
>> -- 
>> View this message in context: http://www.nabble.com/Dependency- 
>> order-in-DefaultModelInheritanceAssembler-tf2187909.html#a6052791
>> Sent from the Maven Developers forum at Nabble.com.
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: [EMAIL PROTECTED]
>> For additional commands, e-mail: [EMAIL PROTECTED]
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
> 
> 
> 

-- 
View this message in context: 
http://www.nabble.com/Dependency-order-in-DefaultModelInheritanceAssembler-tf2187909.html#a6059560
Sent from the Maven Developers forum at Nabble.com.


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to