You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@maven.apache.org by Mike Lenner <mi...@gmail.com> on 2010/10/06 23:19:23 UTC
Maven 3.0 reading non-maven build files
All -
I'm trying to support my company's migration to maven 3.0. In doing
so, I'd like for the existing XML build files (our own schema) to
continue to work, just with Maven reading them in. Our legacy build
files are fairly simple and map very easily to a pom.
My plan is to follow the model of pmaven-cli. That is, create a maven
binary specific to my company (that I'll distribute) which uses my own
ModelReader to process the old, legacy build files.
1. Does this sound like a reasonable approach?
2. Am I on the right track with this:
public class MyCompayMavenCli extends MavenCli {
@Override
protected ModelProcessor createModelProcessor( PlexusContainer
container ) throws ComponentLookupException {
// create a new ModelProcessor which is just like
DefaultModelProcessor except it uses
// a custom ModelReader.
}
}
Thanks,
Mike
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org
Re: Maven 3.0 reading non-maven build files
Posted by Jason van Zyl <ja...@sonatype.com>.
I said copy what's in polyglot. I just assumed you would look at delegating model processor.
You create a Mapping, Reader and Writer. The Mapping is discovered and injected into Polyglot. Polyglot walks through the Mappings and if a particular mapping can accept processing the model input the Mapping grabs its associated Reader and builds the Model object.
Copy the polyglot-groovy or polyglot-jruby module. Build it, add it to the assembly so it's on the classpath and discovered and it should work.
On Oct 14, 2010, at 4:11 PM, Mike Lenner wrote:
> It seems that overriding MavenCli.createModelProcessor as shown below
> is incorrect. That will change the ModelProcessor used by MavenCli to
> *locate* the build file, but not the ModelProcessor used by
> DefaultModelBuilder to *read* the build file and generate a Model.
>
> So, following Polyglot, it actually seems like my own ModelProcessor
> needs to be inserted into the plexus container manually to be used by
> the DefaultModelBuilder. Is this correct?
>
> ## From Polygloy:
>
> // HACK: Wedge our processor in as the default
> final ComponentDescriptor<?> source =
> container.getComponentDescriptor(
> ModelProcessor.class.getName(), "polyglot" );
> final ComponentDescriptor<?> target =
> container.getComponentDescriptor(
> ModelProcessor.class.getName(), "default" );
> target.setImplementation( source.getImplementation() );
>
> On Wed, Oct 6, 2010 at 6:05 PM, Jason van Zyl <ja...@sonatype.com> wrote:
>> Copy what's in Polyglot is what will work for now. At some point in the future we can dynamically load those extensions but they won't right now. So adding a JAR with your model reader/writer will work.
>>
>> On Oct 6, 2010, at 5:19 PM, Mike Lenner wrote:
>>
>>> All -
>>>
>>> I'm trying to support my company's migration to maven 3.0. In doing
>>> so, I'd like for the existing XML build files (our own schema) to
>>> continue to work, just with Maven reading them in. Our legacy build
>>> files are fairly simple and map very easily to a pom.
>>>
>>> My plan is to follow the model of pmaven-cli. That is, create a maven
>>> binary specific to my company (that I'll distribute) which uses my own
>>> ModelReader to process the old, legacy build files.
>>>
>>> 1. Does this sound like a reasonable approach?
>>>
>>> 2. Am I on the right track with this:
>>>
>>> public class MyCompayMavenCli extends MavenCli {
>>>
>>> @Override
>>> protected ModelProcessor createModelProcessor( PlexusContainer
>>> container ) throws ComponentLookupException {
>>> // create a new ModelProcessor which is just like
>>> DefaultModelProcessor except it uses
>>> // a custom ModelReader.
>>> }
>>> }
>>>
>>> Thanks,
>>> Mike
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
>>> For additional commands, e-mail: dev-help@maven.apache.org
>>>
>>
>> Thanks,
>>
>> Jason
>>
>> ----------------------------------------------------------
>> Jason van Zyl
>> Founder, Apache Maven
>> http://twitter.com/jvanzyl
>> ---------------------------------------------------------
>>
>> believe nothing, no matter where you read it,
>> or who has said it,
>> not even if i have said it,
>> unless it agrees with your own reason
>> and your own common sense.
>>
>> -- Buddha
>>
>>
>>
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
> For additional commands, e-mail: dev-help@maven.apache.org
>
Thanks,
Jason
----------------------------------------------------------
Jason van Zyl
Founder, Apache Maven
http://twitter.com/jvanzyl
---------------------------------------------------------
Re: Maven 3.0 reading non-maven build files
Posted by Jason van Zyl <ja...@sonatype.com>.
If you put your code somewhere for the Mapping, Reader and Writer I will take a look. I'm working on the shell and polyglot currently so it wouldn't take me long to get you going.
On Oct 14, 2010, at 4:11 PM, Mike Lenner wrote:
> It seems that overriding MavenCli.createModelProcessor as shown below
> is incorrect. That will change the ModelProcessor used by MavenCli to
> *locate* the build file, but not the ModelProcessor used by
> DefaultModelBuilder to *read* the build file and generate a Model.
>
> So, following Polyglot, it actually seems like my own ModelProcessor
> needs to be inserted into the plexus container manually to be used by
> the DefaultModelBuilder. Is this correct?
>
> ## From Polygloy:
>
> // HACK: Wedge our processor in as the default
> final ComponentDescriptor<?> source =
> container.getComponentDescriptor(
> ModelProcessor.class.getName(), "polyglot" );
> final ComponentDescriptor<?> target =
> container.getComponentDescriptor(
> ModelProcessor.class.getName(), "default" );
> target.setImplementation( source.getImplementation() );
>
> On Wed, Oct 6, 2010 at 6:05 PM, Jason van Zyl <ja...@sonatype.com> wrote:
>> Copy what's in Polyglot is what will work for now. At some point in the future we can dynamically load those extensions but they won't right now. So adding a JAR with your model reader/writer will work.
>>
>> On Oct 6, 2010, at 5:19 PM, Mike Lenner wrote:
>>
>>> All -
>>>
>>> I'm trying to support my company's migration to maven 3.0. In doing
>>> so, I'd like for the existing XML build files (our own schema) to
>>> continue to work, just with Maven reading them in. Our legacy build
>>> files are fairly simple and map very easily to a pom.
>>>
>>> My plan is to follow the model of pmaven-cli. That is, create a maven
>>> binary specific to my company (that I'll distribute) which uses my own
>>> ModelReader to process the old, legacy build files.
>>>
>>> 1. Does this sound like a reasonable approach?
>>>
>>> 2. Am I on the right track with this:
>>>
>>> public class MyCompayMavenCli extends MavenCli {
>>>
>>> @Override
>>> protected ModelProcessor createModelProcessor( PlexusContainer
>>> container ) throws ComponentLookupException {
>>> // create a new ModelProcessor which is just like
>>> DefaultModelProcessor except it uses
>>> // a custom ModelReader.
>>> }
>>> }
>>>
>>> Thanks,
>>> Mike
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
>>> For additional commands, e-mail: dev-help@maven.apache.org
>>>
>>
>> Thanks,
>>
>> Jason
>>
>> ----------------------------------------------------------
>> Jason van Zyl
>> Founder, Apache Maven
>> http://twitter.com/jvanzyl
>> ---------------------------------------------------------
>>
>> believe nothing, no matter where you read it,
>> or who has said it,
>> not even if i have said it,
>> unless it agrees with your own reason
>> and your own common sense.
>>
>> -- Buddha
>>
>>
>>
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
> For additional commands, e-mail: dev-help@maven.apache.org
>
Thanks,
Jason
----------------------------------------------------------
Jason van Zyl
Founder, Apache Maven
http://twitter.com/jvanzyl
---------------------------------------------------------
Our achievements speak for themselves. What we have to keep track
of are our failures, discouragements and doubts. We tend to forget
the past difficulties, the many false starts, and the painful
groping. We see our past achievements as the end result of a
clean forward thrust, and our present difficulties as
signs of decline and decay.
-- Eric Hoffer, Reflections on the Human Condition
Re: Maven 3.0 reading non-maven build files
Posted by Mike Lenner <mi...@gmail.com>.
It seems that overriding MavenCli.createModelProcessor as shown below
is incorrect. That will change the ModelProcessor used by MavenCli to
*locate* the build file, but not the ModelProcessor used by
DefaultModelBuilder to *read* the build file and generate a Model.
So, following Polyglot, it actually seems like my own ModelProcessor
needs to be inserted into the plexus container manually to be used by
the DefaultModelBuilder. Is this correct?
## From Polygloy:
// HACK: Wedge our processor in as the default
final ComponentDescriptor<?> source =
container.getComponentDescriptor(
ModelProcessor.class.getName(), "polyglot" );
final ComponentDescriptor<?> target =
container.getComponentDescriptor(
ModelProcessor.class.getName(), "default" );
target.setImplementation( source.getImplementation() );
On Wed, Oct 6, 2010 at 6:05 PM, Jason van Zyl <ja...@sonatype.com> wrote:
> Copy what's in Polyglot is what will work for now. At some point in the future we can dynamically load those extensions but they won't right now. So adding a JAR with your model reader/writer will work.
>
> On Oct 6, 2010, at 5:19 PM, Mike Lenner wrote:
>
>> All -
>>
>> I'm trying to support my company's migration to maven 3.0. In doing
>> so, I'd like for the existing XML build files (our own schema) to
>> continue to work, just with Maven reading them in. Our legacy build
>> files are fairly simple and map very easily to a pom.
>>
>> My plan is to follow the model of pmaven-cli. That is, create a maven
>> binary specific to my company (that I'll distribute) which uses my own
>> ModelReader to process the old, legacy build files.
>>
>> 1. Does this sound like a reasonable approach?
>>
>> 2. Am I on the right track with this:
>>
>> public class MyCompayMavenCli extends MavenCli {
>>
>> @Override
>> protected ModelProcessor createModelProcessor( PlexusContainer
>> container ) throws ComponentLookupException {
>> // create a new ModelProcessor which is just like
>> DefaultModelProcessor except it uses
>> // a custom ModelReader.
>> }
>> }
>>
>> Thanks,
>> Mike
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
>> For additional commands, e-mail: dev-help@maven.apache.org
>>
>
> Thanks,
>
> Jason
>
> ----------------------------------------------------------
> Jason van Zyl
> Founder, Apache Maven
> http://twitter.com/jvanzyl
> ---------------------------------------------------------
>
> believe nothing, no matter where you read it,
> or who has said it,
> not even if i have said it,
> unless it agrees with your own reason
> and your own common sense.
>
> -- Buddha
>
>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org
Re: Maven 3.0 reading non-maven build files
Posted by Jason van Zyl <ja...@sonatype.com>.
Copy what's in Polyglot is what will work for now. At some point in the future we can dynamically load those extensions but they won't right now. So adding a JAR with your model reader/writer will work.
On Oct 6, 2010, at 5:19 PM, Mike Lenner wrote:
> All -
>
> I'm trying to support my company's migration to maven 3.0. In doing
> so, I'd like for the existing XML build files (our own schema) to
> continue to work, just with Maven reading them in. Our legacy build
> files are fairly simple and map very easily to a pom.
>
> My plan is to follow the model of pmaven-cli. That is, create a maven
> binary specific to my company (that I'll distribute) which uses my own
> ModelReader to process the old, legacy build files.
>
> 1. Does this sound like a reasonable approach?
>
> 2. Am I on the right track with this:
>
> public class MyCompayMavenCli extends MavenCli {
>
> @Override
> protected ModelProcessor createModelProcessor( PlexusContainer
> container ) throws ComponentLookupException {
> // create a new ModelProcessor which is just like
> DefaultModelProcessor except it uses
> // a custom ModelReader.
> }
> }
>
> Thanks,
> Mike
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
> For additional commands, e-mail: dev-help@maven.apache.org
>
Thanks,
Jason
----------------------------------------------------------
Jason van Zyl
Founder, Apache Maven
http://twitter.com/jvanzyl
---------------------------------------------------------
believe nothing, no matter where you read it,
or who has said it,
not even if i have said it,
unless it agrees with your own reason
and your own common sense.
-- Buddha