Specifically making sure the core works with a set of plugins I know use p-u 
extensively and m2e.

To start looking at what the plan is. For example if you're not going to export 
any of the utility code used in the core then I can't see much of a problem. If 
it's used an exported like p-u is then we're just going to end up with the same 
problem except we'll have p-u to deal with in older versions and m-s-u in new 
versions and I don't see much benefit in that path. But if you're not that far 
then I'm happy to wait until a branch shows up. But from your comment the other 
day that said you wanted to release 3.0.5 it sounded like you were closer, 
albeit you did say you were hungover ;-)

There are the JSR330/SLF4J changes, your m-s-u changes and I have a branch for 
Eclipse Aether. I'm not sure if it's wise to roll all of that into one release. 
I'm not particular on the order, just a plan on how to do it.

On Oct 23, 2012, at 12:05 PM, Kristian Rosenvold <[email protected]> 
wrote:

> Did you think of *everything* or just core changes?? (haven't really
> started on the core bits properly, there's tons of other stuff to
> test/port first)
> 
> K
> 
> Den 22. okt. 2012 kl. 15:50 skrev Jason van Zyl <[email protected]>:
> 
>> Please put it on a branch in public view before you merge it. I would like 
>> to make sure everything works with a couple systems, but primarily m2e.
>> 
>> On Oct 22, 2012, at 1:14 AM, Kristian Rosenvold 
>> <[email protected]> wrote:
>> 
>>> It's all just on my local machine, so there's not much to look at.
>>> 
>>> It's mostly a textual replace of org.codehaus.plexus.utils with the new
>>> package name. There's a few interesting points
>>> I can mention already:
>>> 
>>> 1. HelpMojo generated by plugin-tools uses plexus. Switching to m-s-u will
>>> effectively make every plugin m-s-u dependant.
>>> 2. Invoker has an implicit import of plexus-utils into the bean shell
>>> scripts (and probably others) being used to verify.
>>>  In this case it feels correct to just add m-s-u side by side.
>>> 3. I have not looked into the plexus relationship for modello generated
>>> code but there seem to be some bindings there too.
>>> 
>>> Kristian
>>> 
>>> 
>>> 
>>> 2012/10/21 Jason van Zyl <[email protected]>
>>> 
>>>> Where's the branch that you're using for testing? I have a few questions
>>>> but I'll look at the branch first.
>>>> 
>>>> On Oct 18, 2012, at 2:11 PM, Kristian Rosenvold <
>>>> [email protected]> wrote:
>>>> 
>>>>> All the plugin IT's pass with m-s-u trunk now, and I am finshed with all
>>>>> the stuff I planned to do.
>>>>> 
>>>>> I am well in progress on a replacement for Xpp3Dom. Initially we're just
>>>>> looking at a compatible replacement
>>>>> in a different package that can be used as a replacement for 90-95% of
>>>> the
>>>>> use cases.
>>>>> 
>>>>> I think we should release "0.9" now or at least very soon, and just give
>>>>> the Xpp3Dom stuff a couple of extra weeks and we can call that version
>>>> 1.0
>>>>> ;)
>>>>> 
>>>>> Kristian
>>>>> 
>>>>> 
>>>>> 2012/10/15 Mark Struberg <[email protected]>
>>>>> 
>>>>>> Guys, you rock!
>>>>>> 
>>>>>> I also like to add my thanks to Stephen as he started the
>>>>>> plexus-utils-commons-bridge over in our sandbox. Without this work we
>>>> would
>>>>>> not have been able to do this so fast. Also a thanks to all guys who
>>>> helped
>>>>>> importing the stuff they wrote into this module.
>>>>>> 
>>>>>> LieGrue,
>>>>>> strub
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> ----- Original Message -----
>>>>>>> From: Kristian Rosenvold <[email protected]>
>>>>>>> To: Maven Developers List <[email protected]>
>>>>>>> Cc:
>>>>>>> Sent: Monday, October 15, 2012 8:37 AM
>>>>>>> Subject: Re: status update maven-shared-utils
>>>>>>> 
>>>>>>> I have been running m-s-u with the entire maven codebase loaded,
>>>>>> basically
>>>>>>> analyzing all usages
>>>>>>> and deleting any code from m-s-u that is unused in maven. In a couple
>>>> of
>>>>>>> cases I have also
>>>>>>> modified maven code to use commons code directly so we can avoid some
>>>>>> code
>>>>>>> in m-s-u.
>>>>>>> 
>>>>>>> We discussed this on IRC and those of us present figured it would be a
>>>>>> good
>>>>>>> idea to
>>>>>>> keep m-s-u at a minimum and *not* use this as a playground for adding
>>>> all
>>>>>>> sorts of
>>>>>>> other nifty features we might feel like adding, both now and in the
>>>>>> future.
>>>>>>> (I know this *sounds*
>>>>>>> so good, but it also sounds like wishful thinking ;) I also think it's
>>>>>>> desirable that m-s-u *NOT*
>>>>>>> support any use cases outside maven ;)
>>>>>>> 
>>>>>>> I have in practice migrated "most" of the maven codebase to use m-s-u,
>>>>>>> and
>>>>>>> what remains is basically Xpp3Dom and its close "friends". (So while
>>>>>>> everything in
>>>>>>> org.codehaus.plexus.util.xml is deprecated, that's really only because
>>>> we
>>>>>>> don't have
>>>>>>> Xpp3Dom yet)
>>>>>>> 
>>>>>>> I only have a few things left before I'm "1.0" ready:
>>>>>>> A) Finish analyzing all the usages so I can trim m-s-u further down. We
>>>>>> can
>>>>>>> always reinstate code if I delete too much ;)
>>>>>>> B) Make plugin it's run with m-s-u (decent progress has been made here,
>>>>>>> more or less finished)
>>>>>>> C) I am also considering just doing a clean reimplementation of Xpp3Dom
>>>>>> and
>>>>>>> its companions, realistically it's not that
>>>>>>> many lines of code. Unsure if that is "1.0" material.
>>>>>>> 
>>>>>>> Kristian
>>>>>>> 
>>>>>>> 
>>>>>>> 2012/10/14 Robert Scholte <[email protected]>
>>>>>>> 
>>>>>>>> Hi,
>>>>>>>> 
>>>>>>>> Mark, Kristian and I have made some good progress on the Maven Shared
>>>>>>>> Utils.
>>>>>>>> 
>>>>>>>> The project has now 2 compile-scoped dependencies: commons-io-2.2
>>>>>> (final
>>>>>>>> 1.5 compatible version) and jsr305-2.0.1 (for the support of @Nonnull
>>>>>> and
>>>>>>>> @Nullable)
>>>>>>>> Since we still think that Maven Shared Utils should not have any
>>>>>>>> dependencies, the commons-io is shaded.
>>>>>>>> 
>>>>>>>> ReaderFactory and WriterFactory now return a Reader or Writer, the
>>>>>> method
>>>>>>>> decides which implementation is used. Right now that is commons-io
>>>>>>>> 
>>>>>>>> CollectionUtils has been removed, since there is a very small usage of
>>>>>> it.
>>>>>>>> With generics this class has become useless.
>>>>>>>> 
>>>>>>>> ExceptionUtils is nominated to be removed, since almost every
>>>>>> Exception in
>>>>>>>> JDK5 can chain exceptions. For the few left we're looking if it is
>>>>>>> worth
>>>>>>>> to keep it here or let does project depend on the original
>>>> plexus-utils
>>>>>>>> (different
>>>>>>>> package, so no class-collision)
>>>>>>>> 
>>>>>>>> The whole org.apache.maven.shared.utils.**xml package is nominated to
>>>>>> be
>>>>>>>> removed as well, since all its classes are deprecated.
>>>>>>>> 
>>>>>>>> We're making heavy usage of generics, varArgs and other JDK5 specific
>>>>>>>> features.
>>>>>>>> 
>>>>>>>> My opinion is that we need to remove all deprecated code, solve all
>>>>>> TODO
>>>>>>>> comments before its first release.(the number of todo's is very
>>>> small,6
>>>>>>> in
>>>>>>>> main and 1 in test, and have either to do with method signatures or
>>>>>> with
>>>>>>>> <=jdk1.4 issues.)
>>>>>>>> The reason is simple: with a new package this is the best moment to
>>>> get
>>>>>>>> rid of some legacy code from the old plexus-utils which had to stay
>>>>>>>> backwards compatible.
>>>>>>>> 
>>>>>>>> If you think there are other classes/methods which need to be
>>>>>> discussed,
>>>>>>>> please let us know.
>>>>>>>> 
>>>>>>>> 
>>>>>>>> thanks,
>>>>>>>> 
>>>>>>>> Robert
>>>> ------------------------------**------------------------------**---------
>>>>>>>> To unsubscribe, e-mail:
>>>>>>> [email protected].**org<[email protected]>
>>>>>>>> For additional commands, e-mail: [email protected]
>>>>>> 
>>>>>> ---------------------------------------------------------------------
>>>>>> To unsubscribe, e-mail: [email protected]
>>>>>> For additional commands, e-mail: [email protected]
>>>> 
>>>> Thanks,
>>>> 
>>>> Jason
>>>> 
>>>> ----------------------------------------------------------
>>>> Jason van Zyl
>>>> Founder & CTO, Sonatype
>>>> 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
>> 
>> Thanks,
>> 
>> Jason
>> 
>> ----------------------------------------------------------
>> Jason van Zyl
>> Founder & CTO, Sonatype
>> Founder,  Apache Maven
>> http://twitter.com/jvanzyl
>> ---------------------------------------------------------
>> 
>> Selfish deeds are the shortest path to self destruction.
>> 
>> -- The Seven Samuari, Akira Kurosawa
>> 
>> 
>> 
>> 
>> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]
> 

Thanks,

Jason

----------------------------------------------------------
Jason van Zyl
Founder & CTO, Sonatype
Founder,  Apache Maven
http://twitter.com/jvanzyl
---------------------------------------------------------

To do two things at once is to do neither.
 
 -- Publilius Syrus, Roman slave, first century B.C.





Reply via email to