You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@commons.apache.org by Simone Tripodi <si...@apache.org> on 2012/02/14 09:50:58 UTC
[chain] towards 2.0
Hi al guys,
current 2.0 trunk has been updated to fully support Generics[1],
adding improvement as discussed on 2011 in ML - there is the cost
anyway of losing the backward compatibility, clirr detects 57 errors.
So, according to what we already experienced (digester3 ,pool2), I
would proceed on updating the package to o.a.c.chain2, groupId to
o.a.c. and artifactId commons-chain2.
Any objection?
TIA,
-Simo
[1] https://issues.apache.org/jira/browse/CHAIN-58
http://people.apache.org/~simonetripodi/
http://simonetripodi.livejournal.com/
http://twitter.com/simonetripodi
http://www.99soft.org/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org
Re: [chain] towards 2.0
Posted by Elijah Zupancic <el...@zupancic.name>.
It looks like I'm getting a test failing with the patch, I will need some
more time to investigate.
It is one of these
tests: testDefault(org.apache.commons.chain2.config.ConfigParserTestCase):
Correct command count expected:<17> but was:<8>
This error message is associated with order dependent tests on chain.
I will post an update when I have a working fix.
-Elijah
On Tue, Feb 14, 2012 at 7:02 AM, Elijah Zupancic <el...@zupancic.name>wrote:
> Hi Sebb and Simo,
>
> I just created an issue to represent the work for renaming the package and
> the artifact id.
> https://issues.apache.org/jira/browse/CHAIN-65
>
> In it, I have attached a potential patch for the rename. Please take a
> look and let me know if it works for you.
>
> Thanks,
> -Elijah
>
>
> On Tue, Feb 14, 2012 at 5:18 AM, Simone Tripodi <si...@apache.org>wrote:
>
>> Salut,
>>
>> >
>> > There's no need to update both the groupId and tthe artifactId.
>> > So long as the each unique package name relates to a unique
>> > groupId:artifactId pair, Maven should be able to resolve dependencies
>> > correctly.
>> >
>>
>> Yes it does, the issue is not technical but rather keeping coherence
>> with previous cases, indeed I mentioned past experiences with pool2
>> [1] (o.a.c:commons-pool2) and digester3 [2] (o.a.c:digester3) where we
>> agreed on updating both - what is the reason to make an exception with
>> chain?
>>
>> >
>> > Changing package name causes lots of work for downstream users.
>> >
>>
>> Yes, I agree, anyway we spoke about the plan of releasing a major
>> version of the [chain] component...
>>
>> > So are you sure that:
>> > - compatibilty has to be broken in order to support the changes that
>> > have been made?
>>
>> The main big impact is changing the Command/Chain#execute() method
>> signature supporting the Map<K,V> as context, as we discussed last
>> year - since we
>>
>> > - there aren't any other pending API changes that would require
>> > another package rename for the next release?
>> >
>>
>> Do you mean components that dependes to chain?
>>
>> Just for the record, clirr report of chain2 [3] has been updated on my
>> personal space, so people can discuss about changes.
>>
>> thanks for the feedbacks,
>> -Simo
>>
>> [1] https://svn.apache.org/repos/asf/commons/proper/pool/trunk/pom.xml
>> [2]
>> https://svn.apache.org/repos/asf/commons/proper/digester/tags/DIGESTER3_3_0/pom.xml
>> [3] http://people.apache.org/~simonetripodi/chain2/clirr-report.html
>>
>> http://people.apache.org/~simonetripodi/
>> http://simonetripodi.livejournal.com/
>> http://twitter.com/simonetripodi
>> http://www.99soft.org/
>>
>>
>>
>>
>> >
>> > There's no need to update both the groupId and tthe artifactId.
>> > So long as the each unique package name relates to a unique
>> > groupId:artifactId pair, Maven should be able to resolve dependencies
>> > correctly.
>> >
>> > However it is easier to keep the artifactId in step with the package
>> name.
>> >
>> >> Any objection?
>> >
>> > Changing package name causes lots of work for downstream users.
>> >
>> > So are you sure that:
>> > - compatibilty has to be broken in order to support the changes that
>> > have been made?
>> > - there aren't any other pending API changes that would require
>> > another package rename for the next release?
>> >
>> >> TIA,
>> >> -Simo
>> >>
>> >> [1] https://issues.apache.org/jira/browse/CHAIN-58
>> >>
>> >> http://people.apache.org/~simonetripodi/
>> >> http://simonetripodi.livejournal.com/
>> >> http://twitter.com/simonetripodi
>> >> http://www.99soft.org/
>> >>
>> >> ---------------------------------------------------------------------
>> >> To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
>> >> For additional commands, e-mail: dev-help@commons.apache.org
>> >>
>> >
>> > ---------------------------------------------------------------------
>> > To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
>> > For additional commands, e-mail: dev-help@commons.apache.org
>> >
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
>> For additional commands, e-mail: dev-help@commons.apache.org
>>
>>
>
Re: [chain] towards 2.0
Posted by Elijah Zupancic <el...@zupancic.name>.
Hi Sebb and Simo,
I just created an issue to represent the work for renaming the package and
the artifact id.
https://issues.apache.org/jira/browse/CHAIN-65
In it, I have attached a potential patch for the rename. Please take a look
and let me know if it works for you.
Thanks,
-Elijah
On Tue, Feb 14, 2012 at 5:18 AM, Simone Tripodi <si...@apache.org>wrote:
> Salut,
>
> >
> > There's no need to update both the groupId and tthe artifactId.
> > So long as the each unique package name relates to a unique
> > groupId:artifactId pair, Maven should be able to resolve dependencies
> > correctly.
> >
>
> Yes it does, the issue is not technical but rather keeping coherence
> with previous cases, indeed I mentioned past experiences with pool2
> [1] (o.a.c:commons-pool2) and digester3 [2] (o.a.c:digester3) where we
> agreed on updating both - what is the reason to make an exception with
> chain?
>
> >
> > Changing package name causes lots of work for downstream users.
> >
>
> Yes, I agree, anyway we spoke about the plan of releasing a major
> version of the [chain] component...
>
> > So are you sure that:
> > - compatibilty has to be broken in order to support the changes that
> > have been made?
>
> The main big impact is changing the Command/Chain#execute() method
> signature supporting the Map<K,V> as context, as we discussed last
> year - since we
>
> > - there aren't any other pending API changes that would require
> > another package rename for the next release?
> >
>
> Do you mean components that dependes to chain?
>
> Just for the record, clirr report of chain2 [3] has been updated on my
> personal space, so people can discuss about changes.
>
> thanks for the feedbacks,
> -Simo
>
> [1] https://svn.apache.org/repos/asf/commons/proper/pool/trunk/pom.xml
> [2]
> https://svn.apache.org/repos/asf/commons/proper/digester/tags/DIGESTER3_3_0/pom.xml
> [3] http://people.apache.org/~simonetripodi/chain2/clirr-report.html
>
> http://people.apache.org/~simonetripodi/
> http://simonetripodi.livejournal.com/
> http://twitter.com/simonetripodi
> http://www.99soft.org/
>
>
>
>
> >
> > There's no need to update both the groupId and tthe artifactId.
> > So long as the each unique package name relates to a unique
> > groupId:artifactId pair, Maven should be able to resolve dependencies
> > correctly.
> >
> > However it is easier to keep the artifactId in step with the package
> name.
> >
> >> Any objection?
> >
> > Changing package name causes lots of work for downstream users.
> >
> > So are you sure that:
> > - compatibilty has to be broken in order to support the changes that
> > have been made?
> > - there aren't any other pending API changes that would require
> > another package rename for the next release?
> >
> >> TIA,
> >> -Simo
> >>
> >> [1] https://issues.apache.org/jira/browse/CHAIN-58
> >>
> >> http://people.apache.org/~simonetripodi/
> >> http://simonetripodi.livejournal.com/
> >> http://twitter.com/simonetripodi
> >> http://www.99soft.org/
> >>
> >> ---------------------------------------------------------------------
> >> To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
> >> For additional commands, e-mail: dev-help@commons.apache.org
> >>
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
> > For additional commands, e-mail: dev-help@commons.apache.org
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
> For additional commands, e-mail: dev-help@commons.apache.org
>
>
Re: [chain] towards 2.0
Posted by Simone Tripodi <si...@apache.org>.
Salut,
>
> There's no need to update both the groupId and tthe artifactId.
> So long as the each unique package name relates to a unique
> groupId:artifactId pair, Maven should be able to resolve dependencies
> correctly.
>
Yes it does, the issue is not technical but rather keeping coherence
with previous cases, indeed I mentioned past experiences with pool2
[1] (o.a.c:commons-pool2) and digester3 [2] (o.a.c:digester3) where we
agreed on updating both - what is the reason to make an exception with
chain?
>
> Changing package name causes lots of work for downstream users.
>
Yes, I agree, anyway we spoke about the plan of releasing a major
version of the [chain] component...
> So are you sure that:
> - compatibilty has to be broken in order to support the changes that
> have been made?
The main big impact is changing the Command/Chain#execute() method
signature supporting the Map<K,V> as context, as we discussed last
year - since we
> - there aren't any other pending API changes that would require
> another package rename for the next release?
>
Do you mean components that dependes to chain?
Just for the record, clirr report of chain2 [3] has been updated on my
personal space, so people can discuss about changes.
thanks for the feedbacks,
-Simo
[1] https://svn.apache.org/repos/asf/commons/proper/pool/trunk/pom.xml
[2] https://svn.apache.org/repos/asf/commons/proper/digester/tags/DIGESTER3_3_0/pom.xml
[3] http://people.apache.org/~simonetripodi/chain2/clirr-report.html
http://people.apache.org/~simonetripodi/
http://simonetripodi.livejournal.com/
http://twitter.com/simonetripodi
http://www.99soft.org/
>
> There's no need to update both the groupId and tthe artifactId.
> So long as the each unique package name relates to a unique
> groupId:artifactId pair, Maven should be able to resolve dependencies
> correctly.
>
> However it is easier to keep the artifactId in step with the package name.
>
>> Any objection?
>
> Changing package name causes lots of work for downstream users.
>
> So are you sure that:
> - compatibilty has to be broken in order to support the changes that
> have been made?
> - there aren't any other pending API changes that would require
> another package rename for the next release?
>
>> TIA,
>> -Simo
>>
>> [1] https://issues.apache.org/jira/browse/CHAIN-58
>>
>> http://people.apache.org/~simonetripodi/
>> http://simonetripodi.livejournal.com/
>> http://twitter.com/simonetripodi
>> http://www.99soft.org/
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
>> For additional commands, e-mail: dev-help@commons.apache.org
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
> For additional commands, e-mail: dev-help@commons.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org
Re: [chain] towards 2.0
Posted by sebb <se...@gmail.com>.
On 14 February 2012 08:50, Simone Tripodi <si...@apache.org> wrote:
> Hi al guys,
>
> current 2.0 trunk has been updated to fully support Generics[1],
> adding improvement as discussed on 2011 in ML - there is the cost
> anyway of losing the backward compatibility, clirr detects 57 errors.
>
> So, according to what we already experienced (digester3 ,pool2), I
> would proceed on updating the package to o.a.c.chain2, groupId to
> o.a.c. and artifactId commons-chain2.
There's no need to update both the groupId and tthe artifactId.
So long as the each unique package name relates to a unique
groupId:artifactId pair, Maven should be able to resolve dependencies
correctly.
However it is easier to keep the artifactId in step with the package name.
> Any objection?
Changing package name causes lots of work for downstream users.
So are you sure that:
- compatibilty has to be broken in order to support the changes that
have been made?
- there aren't any other pending API changes that would require
another package rename for the next release?
> TIA,
> -Simo
>
> [1] https://issues.apache.org/jira/browse/CHAIN-58
>
> http://people.apache.org/~simonetripodi/
> http://simonetripodi.livejournal.com/
> http://twitter.com/simonetripodi
> http://www.99soft.org/
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
> For additional commands, e-mail: dev-help@commons.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org